identity performance theory

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 08 April 2024

Strategic attitude expressions as identity performance and identity creation in interaction

• Caoimhe O’Reilly   ORCID: orcid.org/0000-0001-5014-4772 1 ,
• Shane Mannion 2 ,
• Paul J. Maher 1 ,
• Elaine M. Smith   ORCID: orcid.org/0000-0001-5017-9457 1 ,
• Pádraig MacCarron 1 , 2 &
• Michael Quayle   ORCID: orcid.org/0000-0002-7497-0566 1 , 3  

Communications Psychology volume  2 , Article number:  27 ( 2024 ) Cite this article

658 Accesses

10 Altmetric

Metrics details

• Communication
• Human behaviour

We assess the strategic alignment of attitudes and the active construction of attitude-based identity across two studies. Study one assessed the twitter response (hashtags in English) to the war in Ukraine for five months after Russia’s first invasion of Ukraine 2022 ( N  = 8149). Results demonstrated that individuals publicly expressed hashtags similar to others close to them in the followership network, showing their support for Ukraine and condemnation of the Russian invasion in qualitatively different ways. Study two was a preregistered Prolific experiment with geographical European participants ran in September, 2022 ( N  = 1368). Results demonstrated that attitude interaction with ingroup members motivated interactants towards attitude alignment, and attitude alignment strengthened the identification that motivated the alignment in the first place. Results suggest that attitude expression is performative and constrained by one’s group relationship with one’s audience and the definition of social identity can be constrained by opinion-based identity performance.

Similar content being viewed by others

Determinants of behaviour and their efficacy as targets of behavioural change interventions

The carbon emissions of writing and illustrating are lower for AI than for humans

Toolbox of individual-level interventions against online misinformation

Introduction.

In periods of social change, especially those involving situations of threat, new groups based on new dimensions often emerge and it is often necessary to form new attitudes in response to novel contexts. In such periods of uncertainty, our ingroups often serve to inform us of the most appropriate norms, behaviours, and attitudes 1 . In February of 2022, the war in Ukraine escalated dramatically when Russia invaded Ukraine. This invasion sparked international outrage and condemnation, much of which was in the form of online attitude sharing. In the current research, we assess the co-ordination and alignment of online attitudes in relation to this novel context; the war in Ukraine. In our first study, we assess the online twitter response to the war in Ukraine to determine whether people tended to share similar attitudes as others in their follower network. We tracked tweets in relation to the war in Ukraine from the beginning of the war (February, 2022) and for the following five months.

In a second study, we assessed attitude identity dynamics and the strategic, performative nature of attitude-based communication in a controlled, experimental context, once again assessing attitudes related to the war in Ukraine. Importantly, we use the term performance to reflect how ‘the expression of social identity [or an identity-relevant attitude] in behaviour is affected by considerations relating to the nature of available audiences’ in line with previous theorists (ref. 2 , p. 29, 5, 4). This does not imply that attitude expressions are merely due to audience effects (cf. the power of the situation, conformity, social desirability, or inauthenticity)—rather that attitudes achieve identity functions in social interaction.

We propose that attitude expressors are sensitive to their audience, and performatively express attitudes to consolidate ingroup identity 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 . To this end, we suggest that interaction with ingroup members should motivate people towards attitude alignment. Beyond this, we assert that the attitude-identity?thyc?> relationship is a dynamic, reciprocal process of influence 2 , 4 , 5 . That is, we suggest that the social act of attitude expression can affect the construction of social identification 10 , 11 , 12 , 13 , 14 , 15 , and this social identification in turn can affect the attitudes that will be publicly expressed because of identity performance goals 2 , 4 , 5 , 16 .

We know that awareness of attitude congruence through computer mediated communication can foster a sense of ingroup identification 17 . Our main hypothesis predicted that interaction with such ingroup members can influence future attitude expression as a result of strategic attitude [dis]alignment and identity performance 2 , 5 , 18 . Our secondary hypotheses predicted that, in a dynamic way, perceptions of similarity as a result of strategic and cumulative attitude alignment can strengthen the identification that motivated the alignment in the first place 10 .

The social identity approach

In order to understand the theoretical rationale for the proposals in the current paper, we first must outline the theoretical model on which the rationale for these proposals is based: the social identity approach. The social identity approach 19 , 20 asserts that the implicit or explicit presence of others in a particular comparative context can imbue categories with social meaning. That is, ‘the mere perception of common category membership’ is sufficient and necessary for social identification and resultant group formation (ref. 21 , p. 3). Beyond the effect of mere similarity 22 , the intergroup context and awareness of a category one belongs to and another that one does not belong to, gives one’s own category social meaning and thus it moves from being a mere category to being a social identity. Social identity is defined by Tajfel as ‘the individual’s knowledge that he [or she] belongs to certain social groups together with some emotional and value significance to him [or her] of this group membership’ (ref. 21 , p. 21). Social identification was defined by Turner as ‘any social categorisation used by a person to define him-or herself and others’ or ‘the process whereby an individual internalises some form of social categorisation so that it becomes a component of the self concept, whether long-lasting or ephemeral’ (ref. 20 , p. 18). Self-categorisation theory explains that the process via which social categories become internalised as social identities is context dependent and varies (partly) as a function of the meta-contrast principle. The metacontrast principle stipulates that people cognitively structure social categories by adhering to elements in a given context that maximise relative differences between one’s ingroup and one’s outgroup and maximise the relative entitativity of one’s ingroup 20 .

In the late 90s and early 2000s there was a shift in social identity theorising towards recognising the performative, active nature of categorisation and social identity processes 6 , 8 , 9 , 23 . The notion of the purposeful deployment of categorisation was introduced, and theorists proposed that categories could be actively created, defined, and deployed 2 , 6 , 8 , 9 , 23 , 24 . That is, theorists began to recognise that categorisation is not something that just passively happens to people, but people have agency, ability, and motivation to be active agents in the construction of categories around which identities and groups revolve. Not only do ‘social identities reflect social reality’ but also ‘social identities move people to create social reality’ (ref. 9 p. 365). The social identity model of deindividuation effects (SIDE model) recognises the strategic nature of social identity 2 , 24 providing a theoretical framework for understanding how social identities can influence and create social reality via motivated identity performance.

Performative expression of attitudes depending on audience

The SIDE model was originally developed to understand the effects of anonymity and identifiability especially within the realm of computer-mediated communication 24 , 25 . It has two main components, cognitive and strategic. The cognitive SIDE asserts that anonymity eliminates visual cues to identity, thus heightening the depersonalisation process and the likelihood for group identities to be salient. In 2007, the strategic SIDE was extended by Klein and colleagues to explain how public expression of identity can be performative and serve to consolidate, as well as construct, social identity 2 . Their theoretical assertion is that individuals’ behaviour can be purposefully co-ordinated, motivated by a desire to be perceived as an ingroup member and to achieve ingroup acceptance 2 , 26 , 27 . The SIDE model provides a theoretical framework for understanding performative attitude expression. In line with the SIDE model we suggest that behavioural coordination can apply to attitudes, that attitude expression can be performative, and that people can be motivated towards attitude alignment when they express attitudes to their in-group members to publicly consolidate their identity.

This is consistent with the social identity approach at large. Self-categorisation theory asserts that, in an intergroup context, in line with the metacontrast principle, people should be motivated to be more similar to their ingroup to maximise the relative differences between themselves and their outgroup and to maximise the entitativity of their ingroup 20 . Similarly, the referent informational influence theory 20 , 28 explains that ingroup members provide information about how to think and behave appropriately as a group member. As a result, attitudes perceived as normative of ingroup members are experienced as socially valid, accurate, and appropriate because ingroup members are relied on to provide information about appropriate social norms and are prioritized as a source of valid/correct information 29 , 30 , 31 , 32 . Thus, ingroup members should be motivated towards broad attitude alignment because (a) fellow ingroup members are perceived as having valid, normative, appropriate attitudes and (b) people in intergroup situations are motivated to be more similar to their ingroups and distinct from their outgroups 20 , 28 .

Klein et al. 18 provide empirical support that attitude expression (specifically prejudice attitude expression) is constrained by one’s audience and that attitude expressions can be strategically used to publicly consolidate one’s identity 18 . In the current study, we suggest that the audience is a ‘co-present interactional reality, itself susceptible to social influence’ (ref. 5 , p.21). Recognising this, we suggest that people in interaction can engage in a performative, active process of attitude co-ordination constrained by the identity relationship between interactants 4 . We suggest that attitudes that become normative are actively defined and constructed in interaction via a process of attitude prediction, desire to publicly express in-group consistent attitudes, and strategic attitude expression constrained by the ingroup/outgroup relationship between interactants 5 .

In line with the social identity approach, we therefore propose that attitude expression can be an identity performance which is strategic and co-created depending on one’s audience. Interaction with ingroup members should motivate interactors towards attitude alignment to overtly consolidate ingroup identity, because ingroup members provide information about which attitudes are correct/appropriate, and because ingroup members see themselves as more similar to other ingroup members in an intergroup context. Note that ‘performativity’ in this sense is agnostic about whether people are expressing their ‘true’ attitudes or not; the focus of analysis is on the function of attitude expression in achieving group identity and identification.

For such motivated attitude alignment to occur, people must first be interacting with ingroup members 2 , 20 . A plethora of previous research has found that salience of attitude congruence can foster ingroup identification 10 , 11 , 12 , 13 , 17 , 33 , 34 , 35 , 36 . In interaction, opinion-based identification is often the most contextually relevant means for categorising the self and others into groups 10 , 11 , 12 , 37 . This is especially the case when considering anonymous, computer mediated communication, where communication is often attitude based (likes, shares, tweets etc.) and visual identity cues are concealed 3 , 17 . Online, attitudes are often the only cues available from which to categorise others in group terms and under such conditions of relative anonymity we would expect depersonalisation and perceptions in group terms to be heightened 17 , 24 . Thus opinion-based identities may be an important and ubiquitous group type emerging from online communication 17 , 24 .

In the current research we assert that, for identity-laden issues, computer mediated attitude congruence fosters ingroup identity with an audience whereas attitude incongruence fosters a sense of out-groupness 10 . Importantly, in the current research we are conceptualising categories and any resultant group identification, not as static objective groups (such as race, country of origin; see ref. 37 for a discussion), but as situational identities; ones that have been actively created within the interaction context, whose boundaries exist specifically and uniquely to that interactive context, and which can be updated with the introduction of new situationally relevant information (consistent with the self-categorisation theory’s metacontrast principle, see also. 7 , 8 , 10 , 37 , 38 , 39 ) Our main proposal is that interaction with people perceived as ingroup members will lead to alignment on attitudes, especially in contexts where attitudes are the primary means for achieving affiliation. Interaction with outgroup members may also motivate interactants towards attitude disalignment because people seek differentiation from outgroups, and outgroup members provide information about how not to behave 20 .

Performative attitude expression in the online context

Qualitative research has found that social identity can be actively [re]constructed and [re]defined on social media and that social media can be used as a tool to display and express one’s social identity 40 , 41 , 42 . Furthermore, previous research has found that, when expressing attitudes online, people are aware of, and attentive to, their audience 43 , 44 , 45 . That is, the influence of the audience on communication translates to the online context 46 . Although people can comprehend the limitless nature of the social media audience, people nonetheless tend to behave as if the audience is more confined and imagine a particular audience when posting on social media 43 , 47 .

In line with the social identity approach 19 , 20 , in the online context (as in the offline context) people should be motivated to be similar to their online ingroups 20 . Posting an attitude online is a social act that goes beyond any similarity motivations however. That is, it is possible to be referently influenced by one’s online ingroups without needing to post anything online 20 —the social act of sharing one’s attitude online goes further than mere referent informational influence 20 . People may be motivated to publicly express attitudes online in order to consolidate and mobilise their identity 2 , 48 . People may look to others close to them in a social network and share similar attitudes to them because they are perceived to share an online social identity and because people are motivated to consolidate and mobilise this identity.

In line with our main proposal (that interaction with people perceived as ingroup members will lead to alignment on attitudes) we thus also explore whether people in online communities tend to express similar attitudes as those close to them in the social network. The purpose of this exploration is to determine whether patterns of attitudinal expression in the large scale online social media context suggest that people could be performatively expressing attitudes similar to their online groups to consolidate their online identities. We do not directly test the motivations behind these patterns, we simply explore and observe large-scale patterns of attitude expression within online communities on Twitter.

Reciprocal relationship between performative attitude expression and opinion-based identity

As well as opinion-based identification influencing future motivated attitude alignment, we also expect this attitude alignment to strengthen the identification that motivated the alignment in the first place. Klein et al. 2 suggest that the relationship between context, identity, and identity performance is reciprocal. They explain that context influences one’s salient social identity which in turn influences one’s identity performance but that one’s identity performance can also influence the definition of social identity and the context within which identity is occurring. Quayle applies a similar logic to attitudes specifically. He proposes that the relationship between attitudes and identity is dynamic, and recursive (4, see also 16 who apply this logic to computer-mediated communication).

In the current study, in line with Klein et al. 18 and Quayle 3 , we suggest a dynamic reciprocal relationship between opinion-based identification and performative attitude expression. We propose that attitudes that are publicly expressed are contingent on the group nature of one’s audience (whether the audience is perceived as ingroup or outgroup). Since people are attuned to how they will be positioned by their utterances, performative attitude expression can thus change the attitudes that enter the social world. As groups can be based on the attitudes expressed in context 11 , 12 , performative attitude expression also has the potential to change social identification. That is, the definition of social identity can be constrained by opinion-based identity performance. Thus, the performative, strategic expression of attitudes can affect social reality itself 5 .

Research has also found a cumulative attitude congruence effect, where attitude congruence on multiple attitudes strengthens identification 10 . We therefore expect that motivated attitude alignment can strengthen identification. Specifically, we suggest that the social act of attitude expression affects the construction of social identity, this social identification in turn effects the attitudes that will be publicly expressed (motivated attitude alignment), and the attitudes publicly expressed can, in turn, affect the strength of identification. Importantly, we expect cumulative attitude congruence to strengthen identification for those interacting with opinion-based ingroup members but not those expressing attitudes to outgroups or privately expressing attitudes. We hypothesise that greater congruence will be related to greater identification for those expressing attitudes to ingroup members compared to outgroup members and those privately expressing attitudes. We also expect that attitude incongruence should weaken or even extinguish identification. Therefore, we also track change in identification as a result of changing attitude congruence. We hypothesise that change in congruence will predict change in identification over time (greater congruence will be related to greater identification) for those expressing attitudes to ingroup members.

We run two studies, one an exploratory Twitter study, and the other a preregistered experiment ( https://aspredicted.org/4dg7y.pdf ). In our first study, we use real-world data to assess whether clusters of followers online tend to share similar attitudes to one another. Specifically, we gathered tweets from Twitter based on shared hashtags related to the Ukraine Russia crisis which escalated in 2022 with the Russian invasion of Ukraine. We captured a followership network, ran community detection algorithms to identify groups, and assessed the most popular hashtags shared by each follower community to visualise how attitude alignment occurs in the wild.

While study one uses large-scale real-world data to descriptively explore attitude alignment, in study two, we use controlled, interactive, online experimental methods to quantitatively explore motivated attitude alignment. Most experimental research designs are unable to capture recursive processes. Our experimental design allows us to capture the dynamic, co-occurring process where performative attitude expression shapes identity and where identity shapes performative attitude expression. We suggest that awareness of attitude congruence through computer mediated communication can foster identification 11 , 12 , 17 this identification influences subsequent attitude expression as a result of strategic attitude alignment 2 , 18 ; and perceptions of similarity resulting from cumulative attitude alignment can strengthen identification 10 . We track change in identification over time as a result of changing attitude congruence to determine whether cumulative attitude congruence strengthens identification only when expressing attitudes to an ingroup audience. This design attempts to capture the dynamic and reciprocal nature of situated identity performance.

Study one method

As outlined in the introduction, the social identity performance model asserts that people are motivated to consolidate their identity publicly. We suggest that one way people can achieve this is by publicly expressing attitudes that align with the attitudes of our ingroup members. Public attitude expression is ubiquitous in the online context. The main purpose of study one is to observe large-scale patterns of attitude sharing in ‘real world’ online contexts (as opposed to the more constrained experimental settings) and to explore how patterns of attitudinal expression might relate to the attitudes expressed by others close to oneself in online followership networks. In exploratory analysis, we assess whether individuals tend to publicly share attitudes (in the form of tweets) that are similar to others in their followership networks on Twitter. To do this, we tracked online expressions of attitudes in the form of hashtags on Twitter that related to the war in Ukraine for the five months following Russia’s invasion of Ukraine (February, 2022).

Data were collected using the Twitter academic API specifically for the purpose of the current paper. We scraped Twitter for tweets in English containing hashtags relating to the Russia/Ukraine war, starting from the beginning of the invasion (24/02/22) and ending the day we commenced gathering data (28/06/22). Study one was not preregistered. Data and analytic code can be found here ( https://osf.io/w8shc/?view_only=34d38b51dc8a4691bf4fec2dbe3cde66 ).

Participants and procedure

We searched Twitter to identify popular hashtags related to the ongoing invasion of Ukraine, as well as hashtags that were ‘trending’ at the time (see Supplementary Note  2 for full list of hashtags), attempting to identify as many dimensions of the online discourse as possible. Participants were included in the dataset if they had tweeted any of the hashtags on the hashtag list between February and June 2022 (from Russia’s invasion of Ukraine and for five months after this). This yielded 5,178,508 tweets from 1,053,860 users. We then examined the frequencies of hashtag expressions for each tweet and chose six hashtags that had high tweet or retweet frequencies. Our final sample consisted of a subset of users who had tweeted one of these hashtags (Supplementary Note  4 ). We then created a dataset of all tweets in the original dataset from users who were in our final sample. Next, we gathered the follower list for each of these users, discarding the data for any user whose account information was unavailable through the Twitter API (e.g., restricted, banned, or deleted accounts). We also discarded any user in the follower lists for whom we had no tweets. This left us with 8149 users. We then built a followership network of these users, where users were linked to every user they were followed by, if that user had also tweeted or retweeted one of the hashtags.

The followership network of Twitter users is a directed network (i.e., there is a directionality associated with followership, unlike, for example, Facebook friendships) We thus employed the Infomap community detection algorithm to the directed version of the network 49 . The Infomap algorithm identifies communities based on how information may flow through the network, taking into account the directionality of the followership edges 50 . We then identified the five most common hashtags shared in each community detected by the algorithm. We excluded several hashtags such as #Ukraine, #Russia (Supplementary Note  3 ) because these were hashtags used by all communities and therefore not that informative of differences between the communities. That is, these hashtags did little more than signify that people were discussing the war in Ukraine crisis at large but were not informative of any qualitative nuances in the discourse. Using AFINN sentiment lexicon 51 , we obtained a sentiment score from the tweets of each user.

Statistics and reproducibility

The alpha level for all tests was .05 and all tests were two-tailed. Parametric tests were run on normal data and non-parametric tests were run when normality was violated. Data was analysed using SPSS version 26, R, and Python. Sufficient detail is included to enable reproducibility of this study (including open code and open data), although the particular context in which the Twitter data was collected will of course, not be reproducible (see also discussion section).

Using the Infomap community detection method we detected 611 communities, of which seven had more than 100 users (see Fig.  1 ). We then assessed the frequencies of the five most common hashtags for each of these seven communities. Supplementary note  5 represents the frequencies of people who shared each of these hashtags for all communities. It should be noted that in the communities containing the hashtag ‘#sanctionskill’ (the blue community in both Figs.  1 and 2 ), many of the tweets were referring to sanctions against other countries, not necessarily Russia. However, as we can see in the network diagram (Fig.  1 ), this community is very densely connected to the rest of the network and so we did not remove these users.

Community 0 is represented with bluish green nodes. Community 1 is represented with orange nodes. Community 2 is represented with blue nodes. Community 3 is represented with yellow nodes. Community 4 is represented with redish purple nodes. Community 5 is represented with sky blue nodes. Community 6 is represented with black nodes.

The dots in each cell represent the relative magnitude of the corresponding component i.e., larger dots represent greater frequencies.

We then compared the distribution of each hashtag across communities statistically using a significant chi-square goodness of fit statistic. We also ran individual ANOVA’s to assess the frequency of individuals’ hashtag use across all communities.

On top of the aforementioned aggregate level analysis, we also conducted individual level analysis. Using AFINN sentiment lexicon 51 , we obtained a sentiment score from the tweets of each user and compared the average sentiment across the three largest communities using a one-way ANOVA.

Study two method

Study one allowed us to visualise real-world large-scale attitude expression patterns. This has the advantage of capturing real-world behaviour, however, this data is descriptive and, although it seems that attitude alignment occurs when publicly expressing attitudes to our followers, if we are to assert that attitude alignment strategically occurs when expressing attitudes to ingroup members it is necessary to test this in a more stringent, controlled environment. This is the purpose of study two.

As outlined in the introduction, in (implicit or explicit) intergroup contexts, people are motivated to be similar to their ingroups to maximise intergroup differences and intragroup entitativity (metacontrast theory, 20 ) and because ingroups are perceived to hold appropriate norms and accurate information (referent informational influence theory, 20 ). Furthermore, people are motivated to overtly display their identities via identity performance to consolidate and mobilise these identities (social identity performance model,. 2 ) In study two, we therefore propose that ingroup members should be motivated towards attitude alignment in line with the metacontrast theory and the referent informational influence theory. We also propose that performative attitude expression (in the form of attitude alignment with ingroup members) is one way that people can perform their identities in line with the social identity performance model.

Our main preregistered hypothesis (hypothesis one) predicted that attitude alignment would be greater in the ingroup experimental condition compared to the control condition and the out-group experimental condition. Our secondary hypotheses predicted that, hypothesis two: condition (ingroup vs outgroup vs control) would moderate the effect of congruence on identification. We expected congruence to predict identification better in the experimental (ingroup) condition, versus the outgroup and control, conditions. Hypothesis three predicted that c hange in congruence would predict change in identification over time (greater congruence would be related to greater identification) for the ingroup experimental condition. We expected congruence to predict identification better in the ingroup experimental, versus the outgroup and control conditions.

For the purpose of study two, it is important to distinguish between circumstantial attitude alignment and motivated attitude alignment. Of course, people who hold certain attitudes will be likely to hold other, related attitudes 52 . For example, if a person believes that their country of residence should give money towards Ukraine forces, they may also believe that their country of residence should give arms/weapons to Ukrainian forces because these attitudes are qualitatively similar. This is circumstantial attitude alignment—alignment that occurs because certain attitudes are interconnected and more likely to be co-held. Separate from this is motivated attitude alignment, which is co-ordinated, performative, strategic attitude alignment and which we suggest occurs when interacting in opinion-based intergroup terms. In study two, we compare circumstantial attitude [dis]alignment (the control group) to strategic attitude [dis]alignment (the experimental group) to ensure that any results found do not simply reflect patterns of qualitatively interlinked attitudes. Participants answer attitudes about the current crisis in Ukraine (2022–23).

In the experimental condition, participants share their attitude answers. In the control condition participants answer attitudes privately. In this way, we can compare motivated alignment occurring as a result of attitude expression and awareness of the intergroup context of one’s audience (experimental group) to circumstantial alignment occurring as a result of holding qualitatively interrelated attitudes (control group). We compare circumstantial alignment to motivated alignment to assess whether interactants strategically align their attitudes depending on their group relationship with their audience in public attitude expression. We also compare opinion-based congruence to arbitrary non-opinion-based congruence to ascertain whether attitude congruence, rather than mere similarity, strengthens identification. We track change in identification over time as a result of changing congruence to assess how self-categories can be updated in context with the introduction of new, relevant information, and how attitude alignment can strengthen identification, whereas attitude disalignment can weaken it.

The study design, hypotheses, sample size, exclusion criteria and analyses were preregistered on the 12th of September 2022, prior to data collection ( https://aspredicted.org/4dg7y.pdf ). All analyses described in the preregistration are reported in the article or supplementary materials. All data, study materials, and analytic code are openly available here ( https://osf.io/w8shc/?view_only=34d38b51dc8a4691bf4fec2dbe3cde66 ).

The study received ethical approval from the University of Limerick Committee (19_06_2019). All research is conducted ethically, results are reported honestly, and the submitted work is original and not plagiarised. On September 19th, 2022, data were collected via Prolific (version 2022) using Qualtrics (Version 2022). Prior to data collection, individuals who had participated in our previous studies were precluded (to avoid practice effects). The survey was made available to individuals who self-declared that their first language was English, who had a prolific approval rate greater than 94, who had a minimum of 50 prior survey submissions, and who were currently located in Europe (Supplementary Note  1 ). All participants who were included first gave informed consent.

Prolific samples are susceptible to several biases including first come first serve response bias, WEIRD bias, selection bias, and reward per hour bias (see ref. 53 for more detail). Nonetheless, crowdsourced data has been found to be of high quality 54 and Prolific data has been found to have higher quality of data than other crowd sourcing platforms 55 . It allows us to gather data from wide geographical areas, and data seems to function similarly well to laboratory studies 56 .

Participants

After preregistered exclusions (those who did not give consent ( N  = 1), who progressed less than 95% in the study ( N  = 0), who had less than two people in their dyads ( N  = 27), and who give the incorrect response to ‘What is your letter?’ ( N  = 9)), there were 1368 participants ranging in age from 18 to 75 ( M  = 36.07, SD = 12.27). Of these, 722 self-identified as female, 639 as male and 7 as non-binary. All participants were paid £1.80.

We conducted a sensitivity power analysis 57 with Anova (fixed effects one way) as the statistical test. Based on N  = 1368, with alpha significance criterion .05 two-tailed and a standard power criterion of 80%, the analysis had power to detect an effect size of f  = 0.08 (equivalent to d  = 0.16).

After preregistered exclusions there were 432 people in the control group and 936 in the experimental group, of which 502 were in the experimental ingroup condition and 434 in the experimental outgroup condition.

The exact survey can be found in the supplementary materials. All measures were assessed using distinct samples but identification was measured repeatedly at 11 timepoints.

Participants began by reading an information sheet which outlined a cover story explaining that the survey was about memory and attention. Next, they answered questions pertaining to informed consent, gender (male, female, non-binary), and age. Information on race/ethnicity was not collected.

All participants were asked 11 attitudes pertaining to their opinions about the Ukraine crisis (2022, 2023). Responses were binary ( yes / no; agree/disagree ). Table  1 outlines all attitudes asked and percentage who agreed and disagreed.

Participants in the control group were also asked 11 arbitrary information questions. An example is ‘Is the third item in your prolific identification a letter or a number?’ Responses were binary ( Letter / Number ).

Identification questions were phrased in group terms, for example ‘I identify with the group my partner is representative of’. Identification was measured at eleven different time points (after each exposure/feedback round) with three items ‘I identify with the group my partner is representative of’ 58 , ‘I feel solidarity with the group my partner is representative of’, ‘I feel strong ties with the group my partner is representative of’ 59 . All responses ranged from 1 strongly disagree to 7 strongly agree .

The current study has both an experimental and quasi experimental design. The experimental manipulation separated participants into the experimental condition who observed each other’s attitude answers, and the control condition who observed each other’s answers to the arbitrary non-attitude information questions. The quasi-experimental element involved further subdivision of the control and experimental groups based on their attitude congruence on the first attitude, into; experimental ingroup (those who are aware they have attitude congruence on the first attitude); experimental outgroup (those who are aware they have attitude incongruence on the first attitude); control ingroup (those who had attitude congruence on the first attitude but were unaware of this congruence); and control outgroup (those who had attitude incongruence on the first attitude but were unaware of this incongruence). Participants were not explicitly told they had been categorised into a group.

Congruence was a constructed variable based on whether a dyad answered the attitude (experimental group) or the arbitrary information question (control group) in the same way. If they answered the same, they got a score of one, if they answered differently, they got a score of zero. Total congruence was a sum of all 11 congruence scores thus, a total score of zero meant the two participants did not answer any question in the same way, a score of 11 meant that they answered all 11 questions in the same way.

Attitude congruence for the experimental groups were the same as the congruence variable described above. For the control groups, if both members of the dyad had the same attitude answer they got a score of one, and if they had different answers they got a score of zero. Importantly, the control group participants were unaware of their partner’s attitude answer. Attitude alignment was a sum of all 11 attitude congruence scores.

As well as assessing attitude alignment within dyads, we also assessed bipartite attitude alignment within conditions. To do this, we created bipartite graphs which connected users via attitude agreement 60 , 61 . In these bipartite graphs there were two types of nodes, one representing attitudes, and the other representing each participant. Participant nodes were linked to attitude nodes via positive edges if a participant agreed with an attitude, or via negative edges if a participant disagreed with an attitude. Next, following MacCarron et al. 60 , network projections were created which connected each participant to other participants based on how much attitude congruence they had with each other (60:61). Participants would be very close in the network if they had lots of the same attitudes, and further away if they had lots of different attitudes. We created four network projections (bipartite graphs), one for each condition. We then compared the attitude alignment across conditions statistically. After creating the networks where the edges represent shared agreement of participants, we computed the clustering coefficient and the average path length. The clustering coefficient is the fraction of closed triangles in the network, if a node is connected to two neighbours, it gives the probability those neighbours are also connected (see ref. 62 for further explanation). The average path length represents the mean shortest number of steps between a pair of participants. In social networks this is related to the idea of six degrees of separation which hypothesises that the mean number of steps between any two people is six.

The current study involved computer-mediated interaction. Participants participated in pairs, into which they were randomly matched via Qualtrics programming (following 63 ) Participants read an information sheet, gave informed consent, and answered demographic questions. We recognise that individuals interacting in dyads can perceive themselves to be engaged in intergroup interaction, just as individuals interacting in groups can perceive themselves as individuals rather than as group members 64 , 65 . To maximise the likelihood of intergroup rather than interpersonal interaction participants were told ‘You have been matched with another participant. This participant has been chosen because they are representative of a particular sample of people’. Next, they engaged in a round of instant messaging for 180s with their interactive partner, the purpose of which was to help participants understand they were interacting with a real person (which they were).

Participants then gave an attitude about the Ukraine crisis (2022/2023). Participants in the control group also answered an arbitrary information question. Here the experimental manipulation occurred. Participants in the experimental group observed each other’s answers to the Ukraine attitude, whereas participants in the control group observed each other’s answers to the arbitrary information question. Thus, the experimental group shared attitude answers (which allowed them to observe whether they had attitude congruence or not), while the control group privately answered the same attitudes but shared arbitrary information (which allowed them to observe whether they had congruence on the arbitrary information or not). This allowed us to determine whether any attitude alignment and identification effects found were as a result of similarity, generally, or whether they were a result of attitude similarity in particular. All participants answered attitude one, and participants in the control group additionally answered arbitrary information question one. Following this, all participants were asked their identification with the sample that the other participant represented. Participants then engaged in ten more rounds of information sharing and identification reporting. Finally, participants answered items on superordinate opinion-based social identification, activism intentions, and their country of birth and current residence, before being debriefed.

The alpha level for all tests was 0.05 and all tests were two-tailed. Parametric tests were run on normal data and non-parametric tests were run when normality was violated. Data was analysed using SPSS version 26, R, and Python. Sufficient detail is included to enable reproducibility of experiments, including open materials open analytic code, and open data.

As a manipulation check, to assess whether the experimental ingroup had higher identification after the experimental manipulation than the experimental outgroup, we conducted a one-way ANOVA, with condition as the independent variable and identification at time one as the dependent variable.

To test hypothesis one we conducted a one-way ANOVA with Bonferroni post hoc tests, with condition as the independent variable and total attitude congruence as the dependent variable. We deviated slightly from the preregistration by subdividing the control group in the same way as the experimental group. Analysis without this subdivision is available in the supplementary materials but we caution that this analysis fails to control for circumstantial attitude alignment.

As well as assessing attitude alignment within dyads, we also assessed overall bipartite attitude alignment within each condition via generating bipartite graphs which connect participants via the attitudes they share (Fig.  3 ) 60 , 61 . Firstly, we connected participants who had congruence on all 11 items. We then lowered the agreement threshold until a majority of participants were in a giant connected component. The threshold was nine items. We then assessed two network quantities; path lengths and clustering coefficients. The path length assesses how ‘coherent’ the attitude space is, so shorter path lengths (shorter distances between nodes) represent higher attitude congruence, and longer path lengths (longer distances between nodes) represent lower attitude congruence. We then assessed whether the average path lengths differed across condition via one-way ANOVA. The clustering coefficient is robust to unequal sample sizes and assesses how ‘clustered’ a group is or how many triangles there are. If one’s neighbours in the attitude network all agree with each other, this will result higher clustering, if not then we would observe lower clustering. As the data were not normally distributed, we assessed differences in clustering across condition via a Kruskal-Wallis.

The network for the experimental ingroup condition is more dense with a lower mean path length. This suggests that there is greater attitude alignment in the experimental ingroup condition. The nodes are coloured based on their identification score represented by a rainbow spectrum An edge is created when the participants agree on X or more of the 11 items. a represents the experimental ingroup, b the experimental outgroup, c the control ingroup and d the control outgroup.

Hypothesis two predicted that condition (ingroup vs outgroup vs control) would moderate the effect of congruence on identification. We expected congruence to predict identification better in the experimental (ingroup), versus the outgroup and control, condition. We conducted a linear model with fixed effects of total congruence, condition, and their interaction on total identification averaged across the rounds.

Hypothesis three predicted that change in congruence would predict change in identification over time (greater congruence will be related to greater identification) for the ingroup experimental condition. We expected congruence to predict identification better in the ingroup experimental, versus the outgroup and control conditions. In order to test this, a linear mixed model using maximum likelihood estimation was conducted with identification and congruence at each time point, nested within participants. A variable denoting the eleven rounds was included as a repeated effect. A random intercept was specified in order to account for individual variation in the tendency for identification. Fixed effects of congruence, condition, time, and their interactions were included. No random slopes were specified in order to retain a more parsimonious model. Importantly, we coded congruence as present 1 or not(0) at each point rather than using total congruence scores.

Reporting summary

Further information on research design is available in the  Nature Portfolio Reporting Summary linked to this article.

Results demonstrated that the five most popular hashtags were not the same in each community (Table  2 ; Fig.  2 ). Descriptive results demonstrated that the algorithm detected three large communities each with different popular hashtags, with little to no overlap (Supplementary Note  5 ). The algorithm also detected several smaller communities with some overlap, but still with hashtags unique to their community.

The distribution of hashtags was not even across communities and hashtag use characterised these communities (Table  3 ). This demonstrated that the results depicted in Table  2 differed significantly from a random distribution. ANOVA’s demonstrated that 20 out of 21 showed statistically significant differences (Table  4 ). Thus, followership communities on Twitter are characterised and clearly distinguished by the combination of hashtags that they publicly expressed. Taken together, these results all suggest that groups of followers tend to use similar hashtags depending on what ‘group/community’ they are part of.

A one-way ANOVA (see supplementary note  6 , supplementary note  7 ) demonstrated that the average sentiment for individuals in the followership network was significantly different across communities (community zero: M  = −1.35, SD  = 1.63; community one: M  = −0.51, SD  = 1.60; community two: M  = −0.85, SD  = 1.68), F (2, 4290) = 78.86, p  < 0.001 η p 2  = 0.035, 95% CI 0.03, 0.05]). These results demonstrate that, on average, variation in sentiment is larger across communities than it is within communities

Manipulation check

As expected, identification was significantly higher in the experimental ingroup condition ( M  = 5.21, SD  = 1.02, 95% CI 5.12, 5.30]), compared to the experimental outgroup condition ( M  = 3.54, SD  = 1.11, 95% CI 3.43, 3.64]; F (1, 935) = 575.10, p  < 0.001, η p 2  = 0.38, 95% CI [0.34, 0.42]).

Main hypothesis: Hypothesis one

Attitude alignment was highest in the experimental ingroup condition ( n  = 502), followed by the control ingroup ( n  = 220), the control outgroup ( n  = 212), and lowest in the experimental outgroup ( n  = 434) condition: F (3, 1362) = 37.44, p  < 0.001 η p 2  = 0.08, 95% CI [0.05, 0.10] as shown in Table  5 . Results when randomly selecting one person from each dyad were the same (Supplementary Note  8 ). As expected, the experimental ingroup condition had significantly higher attitude alignment than all other conditions (vs the experimental outgroup condition ( p  < 0.001; 95% CI [0.79, 1.37]); vs the control ingroup condition ( p  = 0.02; 95% CI [0.04, 0.76]); and vs the control outgroup condition ( p  < 0.001; 95% CI [0.59, 1.31]) indicating that publicly expressing attitudes to an ingroup member and seeing their attitudes leads to some kind of strategic motivation for attitude alignment. The null hypothesis was rejected. In our preregistration, we noted that we suspected that experimental outgroup attitude alignment would be lower than the control group. As suspected, Bonferroni comparisons demonstrated that the experimental outgroup condition had significantly lower attitude alignment than the control ingroup condition ( p  < 0.001; 95% CI [−1.05, −0.32]) suggesting that interaction with an outgroup member leads to strategic motivation for attitude disalignment.

Bipartite attitude alignment

Results demonstrated that the experimental outgroup has statistically significantly shorter path lengths ( M  = 4.63) than the experimental ingroup ( M  = 4.67, p  < 0.001), F (2, 1362) = 12.33, p  < 0.001 η p 2  < 0.001, 95% CI [<0.001, <0.001]). This does not support our preregistered hypothesis, however, path length analysis is sensitive to unequal sample sizes and there were 502 people in the experimental ingroup and 434 in the experimental outgroup. The longest path length possible was 15 and the shortest was 1. Such a small difference between means (.03), although statistically significant, because of the unequal sample sizes, is likely to reflect the path length’s sensitivity to sample size rather than reflecting any meaningful difference. Clearly, more work is needed to develop network methods to assess psychologically relevant features, and real-life data which is inevitably susceptible to unequal sample sizes.

Results demonstrated that the experimental ingroup has statistically significantly higher clustering ( Md  = 0.40) than all other conditions (experimental outgroup: Md  = 0.35, control ingroup: Md  = 0.36, control outgroup: Md  = 0.36); H 3  = 25.37, p  < 0.001). The density of edges was higher in the experimental ingroup condition (.05) compared to the control condition (0.04). This indicates that attitude alignment is highest in the experimental ingroup.

Hypothesis Two

No statistically significant main effect of total congruence ( b  = 0.02, SE = 0.03, t  = 0.92, p  = 0.36, 95% CI [−0.03, 0.08]) was found. Condition was dummy-coded, with control group used as a comparison. There was a significant main effect of being in the experimental ingroup (vs control) ( b  = −1.08, SE  = 0.24, t  = −4.48, p  < 0.001, 95% CI [−1.56, −0.61]) and the experimental outgroup (vs control) ( b  = −1.13, SE = 0.23, t  = −4.83, p  < 0.001, 95% CI [−1.58, −0.67]). There was a significant experimental ingroup (vs control) × congruence interaction ( b  = 0.25, SE = 0.04, t  = 6.86, p  < 0.001, 95% CI [0.18, 0.32]) and an experimental outgroup (vs control) × congruence interaction ( b  = 0.18, SE  = 0.04, t  = 4.81, p  < 0.001, 95% CI [0.11, 0.26]). The relationship between congruence and identification was strongest in the ingroup, and stronger in the outgroup condition than in the control condition (see Fig.  4 ). Thus, hypothesis two was supported.

The blue solid line represents the experimental ingroup. The orange long-dashed line represents the experimental outgroup. The green short-dashed line represents the control group.

Hypothesis three

The linear mixed model revealed a significant main effect of congruence ( b  = 0.14, SE = 0.05, t  = 3.01, p  = 0.003 95% CI [0.05, 0.23]) and time ( b  = −0.02, SE = 0.01, t  = −3.97, p  < 0.001, 95% CI [−0.03, −0.01]). Condition was dummy-coded, with the control group used as a comparison. There was a significant main effect of being in the experimental ingroup (vs control) ( b  = 0.29, SE = 0.08, t  = 3.58, p  < 0.001, 95% CI [0.13, 0.45]), and the experimental outgroup (vs control) ( b  = −0.67, SE = 0.08, t  = −8.48, p  < 0.001, 95% CI [−0.83, −0.52]). There was a significant experimental ingroup (vs control) × congruence interaction ( b  = 0.60, SE = 0.07, t  = 9.04, p  < 0.001, 95% CI [0.47, 0.73]), and an experimental outgroup (vs control) × congruence interaction ( b  = 0.83, SE = 0.07, t  = 12.67, p <0.001, 95% CI [0.70, 0.96]). The effect of congruence was stronger in each experimental condition, versus the control condition. This is the effect of sharing attitudes within dyads.

The experimental ingroup (vs control) × time interaction was not statistically significant ( b  = −0.00, SE = 0.01, t  = −0.05, p  = 0.97, 95% CI [−0.02, 0.02]). The experimental outgroup (vs control) × time interaction was significant ( b  = 0.03, SE = 0.007, t  = 3.92, p  < 0.001, 95% CI [0.01, 0.04]). This indicates that the difference in identity between the control group and the outgroup changed over time.

The congruence by time interaction was not statistically significant ( b  = 0.004, SE = 0.01, t  = 0.63, p  = 0.53, 95% CI [−0.01, 0.02]). The experimental ingroup condition/time/congruence interaction was not statistically significant ( b  = 0.01, SE = 0.01, t  = 1.02, p  = 0.31, 95% CI [−0.01, 0.03]), nor was the experimental outgroup condition/time/congruence interaction ( b  = 0.02, SE = 0.01, t  = 0.71, p  = 0.48, 95% CI [−0.01, .03]). Probing these results with analysis split by condition revealed that congruence predicted identification over time for the experimental ingroup ( b  = 0.02, SE = 0.007, t  = 2.11, p  = 0.04, 95% CI [0.001, 0.03]) but not for the experimental outgroup ( b  = 0.01, SE  = 0.008, t  = 1.40, p  = 0.16, 95% CI [−0.01, 0.03]) or the control group ( b  = 0.005, SE = 0.005, t  = 0.90, p  = 0.37, 95% CI [−0.01, 0.02]). Figure  5 illustrates this effect by showing how congruence accumulated over time. It is clear from the graph that in the ingroup condition in particular, this accumulation corresponded with an increase in identification. These results suggest that our analysis did not have sufficient power to detect a three-way interaction but the split group analysis and Fig.  5 clearly demonstrate support for our hypothesis. Figure  6 demonstrates that congruence (vs incongruence) was associated with higher identification for all conditions, but the highest identification related to attitude congruence was in the experimental ingroup condition, and the lowest identification associated with attitude incongruence was in the experimental outgroup condition.

Higher values on the graph are represented by increasingly saturated purple nodes. Green represents the experimental ingroup. Yellow represents the experimental outgroup. Purple represents the control group.

‘Cong’ refers to when individuals have congruence with their dyadic partner on a given attitude. ‘Incong’ refers to when individuals have incongruence with their dyadic partner on a given attitude. Error bars represent 95% confidence intervals ( n  = 836). Green represents the experimental ingroup, yellow represents the experimental outgroup, grey represents the control group. Average identification begins at the highest point in the ingroup and despite differences in congruence over time, congruent outgroup members never reach an average identification as high as any congruent ingroup members. Incongruent ingroup members never reach an average identification level as low as incongruent outgroup members. This displays the impact of the initial opinion-based group identification.

As can be seen in Fig.  6 (see also Supplementary Note  9 ), at every time point participants in the experimental ingroup with congruence had significantly higher identification than all other conditions. With the exception of two comparisons, the experimental outgroup with incongruence had significantly lower identification than all other conditions. Thus, having attitude incongruence had the strongest identification weakening effects for those interacting with outgroup members.

Participants in the experimental ingroup with incongruence had higher identification than participants in the experimental outgroup with incongruence with the exception of time 7. Participants in the experimental outgroup with incongruence had lower identification than participants in the experimental ingroup with incongruence at each time point.

The studies in the current paper were carried out in the context of the ongoing war in Ukraine to assess attitude identity dynamics as they emerge in novel contexts and the strategic, performative nature of attitude-based communication. We assessed whether attitudes in the real-world online context are aligned with attitudes of those in our followership network by assessing the most popular hashtags shared by follower communities on twitter in the first five months of the war in Ukraine (2022). We found that individuals publicly expressed hashtags that were similar to others close to them in the follower network, thus showing their support for Ukraine and condemnation of the Russian invasion but in qualitatively different ways. These results imply that public expression of attitudes online is influenced by, or at least associated with, one’s followership community.

These results demonstrate that individuals tend to publicly express hashtags that others in their follower communities also express. When looking at tweets about the Ukraine crisis, several communities expressed their support for Ukraine (#standwithukraine) but in qualitatively different ways. Some expressed support by tweeting attitudes relating to arming Ukraine (#armukrainenow), some focused on raising awareness about Russian war crimes (#russianwarcrimes) or raising awareness about Ukrainian casualties (#genocideofukrainians), others advocated to sanction Russia (#boycottrussianoilnow), or to stop Russia or Putin (#stoprussia, #stopputin) whilst others focused on advocating for peace or Ukraine’s NATO status (#peace, #nato). Importantly, the public expression of one’s support for Ukraine was similar to those closest to oneself in the followership community. This implies that public expression of attitudes on twitter is affected by one’s follower network—as different clusters of followers publicly expressed tweets that were qualitatively similar to others close to them in the follower network. According to the social identity approach at large, and the social identity performance model, people should be motivated to share attitudes that are similar to their ingroups in order to consolidate and mobilise their identity. The attitude expression patterns that we have observed in the Twitter context in the 5 month aftermath of Russia’s first invasion of Ukraine are consistent with this interpretation as people tended to share attitudes similar to those closest to them in the twitter followership network. Thus, using large-scale, ‘real world’, naturalistic data we have observed patterns of attitude sharing consistent with the social identity performance model implying that people’s public attitude expression is performative and constrained by identity motivations. Nonetheless, we have not directly measured or manipulated identification thus, although the patterns are consistent with the social identity performance model, we can only speculate why these attitudinal expression patterns have occurred.

In a more controlled experiment, we then assessed whether public expression of attitudes can be considered as identity performance constrained by the identity relationship of expressor and audience. In particular, we assessed whether people were motivated towards attitude alignment when interacting with ingroup members. We also assessed the dynamic relationship between attitudes and identities to determine whether motivated attitude alignment could strengthen the identification which motivated the attitude alignment in the first place. We used an interactive experimental design allowing us to observe the attitude identity dynamic as it changed over time in a controlled environment, a process difficult to capture in the experimental context, but crucial if we are to understand the active processes involved in social identity and social reality development over time 7 , 8 .

We found support for our preregistered predictions. Firstly, we found that public expression of attitudes was contingent on the identity relationship between interactants, implying that attitude expression is performative and constrained by, or tailored to, one’s audience. Interaction with ingroup members motivated interactors towards attitude alignment whereas interaction with outgroup members motivated interactors towards attitude disalignment. Importantly, this motivated alignment was different to circumstantial alignment patterns found in the control group who privately expressed their attitudes. On top of finding higher attitude alignment within dyads in the experimental group, we found higher attitude alignment between individuals in the experimental ingroup more so than mere situational alignment (control ingroup condition) or interaction with an outgroup member. These findings suggest that interaction with an ingroup member can lead to alignment of an entire opinion space. Nonetheless, we were unable to robustly assess path length comparisons across groups due to unequal sample sizes thus caution is needed when interpreting these results and future research should assess this using equal samples.

That the entire ingroup aligned their attitudes after interaction with only one ingroup member, could potentially reflect individual ingroup members’ strategic tailoring of expressed attitudes to be consistent with the attitude they perceive to most prototypical. That is, bipartite attitude alignment may be representing a shift towards perceived ingroup prototypical attitudes. In our study, people began with either congruence or incongruence on their attitude about whether their country of origin should extend unlimited welcome to refugees. Perhaps the bipartite attitude alignment for those with attitude congruence on this item represented participants’ movement towards what they perceived to be the most prototypical subsequent attitudes held by people who hold a particular attitude on refugees. If the first attitude question had been different, the form that attitude alignment took may also have been different as the perceived prototypical attitudes of a group revolving around a different attitude may be different. Future research should explore whether attitude alignment with ingroups reflects a convergence towards the attitudes perceived to be most prototypical for that particular attitude-based group. Future research should also explore whether any subsequent attitude expressions are effected by differences in the original attitude around which groups revolve.

We also found that attitude alignment strengthened identification for those interacting with ingroup members. This provides supportive evidence for a dynamic relationship between attitudes and identification, where attitude congruence can foster identification, motivate subsequent attitude alignment, and where subsequent attitude alignment can strengthen the identification that motivated the alignment in the first place. It also seems that belonging to an opinion-based group provides some resilience to any identity weakening effects of attitude incongruence 10 . Thus, people are active agents in interaction and in the construction of social identities 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 . These results provide empirical evidence supporting theoretical models such as the extension of the strategic SIDE 2 Quayle’s performative network theory of attitudes 3 , Durrheim et al.’s identity performance theory 5 and the social identity approach at large 19 .

It seems that participants may have been able to use the knowledge they had of the opinion space to locate their partner in it and performatively [dis]align their own attitudes in a process of joint positioning (although we note that further research is needed here). These results suggest that the attitudes that enter the social world and become normative can be defined in interaction through an active process of attitude co-ordination or polarisation depending on one’s interactive audience. Furthermore, identity performance through attitude expression influences the definition and strength of social identity because it constrains which attitudes can be recognised as normative for one’s group (the ones that get expressed), and because cumulative attitude alignment strengthens identification 10 . This also implies that attitude alignment/consensus is vulnerable to identity-based polarisation as consensus is identity laden. Thus, performative attitude expression has the potential to change social reality itself 5 .

Attitudes are often understood as individual phenomena. Classical ‘individualist’ perspectives on attitudes (e.g., social cognitive theory) conceptualise attitudes as fixed cognitive structures; as stable, internal ways of thinking about target stimuli relatively unconstrained by social context. In contrast to this classical conceptualisation, our research demonstrates that attitudes are collective constructs jointly produced in interaction and which facilitate the formation, evolution, expression, consolidation, and mobilisation of shared social identities as they develop in response to contextual factors 2 , 39 . Our results also support the idea that individual cognition is a function of public attitude expression in line with the original stipulations of the self-categorisation theory. Turner and colleagues state that ‘Social reality testing (consensual validation, seeking the agreement of ingroup others) …are interdependent aspects of achieving a valid social cognition. Individual perception and cognition rest on socially validated knowledge, theories, methods, and categories, just as the power of social consensus to define reality for group members makes sense only if the individual views that make up the consensus have been independently tested.’ 65 . Thus, the dynamic attitude-identity relationship and the identity motivated strategic expression of attitudes should effect individual perceptions and cognition, not merely attitudes expressed publicly.

It is crucial to recognise the dynamic nature of social influence effects, and to recognise that situational identities and expressed attitudes can be actively defined and constructed in interaction 5 , are constrained by the ingroup/outgroup relationship between interactants 2 , and are important processes that can produce social change 5 . It would be interesting for future research to further explore the idea that people are active agents in the creation of identities which can be fostered in interaction via strategic use of contextually relevant attributes, in this case, attitudes. The experimental method used in study two provides an example of one way in which opinion-identity dynamics in interaction, and identity emergence in the online context, can be explored.

Practical implications

Understanding why and how attitudes become aligned is fundamental to understanding processes of social influence, including promoting positive social influence (e.g., vaccine uptake) and resisting negative social influence (misinformation, bot influence). The resolution of many global problems (such as climate change) requires group attitude alignment to be successful. Speculatively speaking, our results suggest that tackling large scale social issues that require attitude synchronisation for their efficacy (e.g., vaccine uptake) may need to go beyond logical or rational arguments, and consider the group nature of attitudes and attitude interactions. Our results suggest that sharing/expressing commonly held attitudes may heighten the likelihood of attitude alignment on a subsequent target attitude one is hoping to change/influence, as it may foster a sense of ingroupness based on congruence on the first attitude. Furthermore, attitudes shared by others already perceived as ingroup members may be particularly influential.

Of course, the same processes of influence could also be harnessed by those attempting to influence attitudes in a socially harmful way (e.g., vaccine conspiracies) but also to protect against such attempts 66 . These social psychological mechanisms are likely exploited by ‘bots’ on social media (automated; semi-automated or sock-puppet accounts): a bot that has previously shared attitudes or tweets that we agree with may have particularly strong influence on future attitudes. Our results also suggest that purposeful misinformation can undermine consensus, especially if that misinformation is coming from accounts that have previously shared attitudes congruent with one’s own attitudes, or those accounts are perceived to represent ingroup members.

Recognising that people may publicly share attitudes online to produce and consolidate group identity may also help to explain why certain attitudes in particular are spread and go viral. It is crucial to understand how attitude alignment or synchronisation in social networks occurs, as social movements created from such attitude alignment are powerful mobilisers of social change and social action 16 . Important large scale social movements have mobilised on Twitter through attitude expression in the form of hashtags (for example #Metoo; #Notallmen and #BlackLivesMatter; #AllLivesMatter, 17, 16, 48). In study one, we captured a mobilisation of attitudes in response to a situation of threat (the Russian invasion of Ukraine, 2022). The mobilisation we observed on Twitter reflected very strong international support for Ukraine at a particular moment in time. Importantly, the particular attitudes that became aligned and that were spread on social media depended on the network structure, implying that the attitude alignment process can be produced and constructed through interaction and can be constrained by one’s audience (those one is connected with on social media). Of course, this research was exploratory and thus future research should verify findings.

In study two, we demonstrated that attitude expression is strategic, and constrained by the identity relationship between interactants. We also found that the opinion space aligned when individuals publicly expressed attitudes in interaction with opinion-based ingroup members. That is, the attitudes that enter the social world, and thus social reality itself, seem to be contingent on the identity relationships between interactants 5 . More generally, the results support the assertion that groupness is central to the attitudinal divides seen in relation to the war in Ukraine. This implies the expression of attitudes via computer-mediated communication is directed by the online behaviour of those we interact with, and identity-laden attitudes should be most amenable to online propagation.

That attitudes are constrained by the group relationship between interactants implies that the attitudes that enter the social world are moulded by the identity performative nature of attitude expression. Thus, the attitudes that gain social power are likely constrained by social identity. This has knock on societal effects as the attitudes that people choose to share on social media have macro-level consequences 48 , 67 , 68 . Identity performance increases the likelihood of publicly expressing identity-laden attitudes on social media and in the process changes the agenda for which attitudes have macrolevel outcomes. e.g., sharing attitudes about sanctioning Russia on Twitter likely had knock on effects such as companies pulling out of Russia. Such macrolevel behavioural effects of attitude-based groups are difficult to capture by measuring individual’s activism and behavioural intentions. Future research should attempt to assess macrolevel behavioural outcomes using nationally representative longitudinal studies, or using large-scale social media studies tracking attitude expression and its correspondence to the mobilisation of action over time.

The opinion space on social media (and in general life) is constantly being produced and reproduced via interaction. In the current study, we captured attitude alignment and propagation process on an upward trajectory but care needs to be taken to retain such propagation tendencies. With fatigue, motivation for attitude propagation in relation to a particular topic can unwind and other, more topical issues can gain momentum. Similarly, in a different context, or in a different moment in time, the mobilisation process we observed, and the particular attitudes that became consensualised, could have been different. The opinion space can thus reconfigure, resulting in the alignment and propagation of different attitudes. Of course, our results cannot fully explain why particular attitudes gain momentum on social media at certain moments in time, but our results strongly imply that to fully understand such processes we need to account for groupness. Attitudes that clearly reflect or represent group identity may be more amenable to online propagation. Future research should recognise the importance of groupness when attempting to understand virality in the online context.

More speculatively, situations of uncertainty and threat, such as the Russian invasion of Ukraine, may perpetuate attitude propagation. When interacting with an ingroup, we may be motivated towards attitudinal alignment, not only to pursue goals of ingroup inclusion, but also because ingroups have collective power only when they achieve consensus or cohesion on core dimensions. When attitudes become coordinated and collective, they have the power to transform general social opinion on issues (e.g. women’s/LQBTQ rights movements) as well as ignite collective action and transform macro level societal structures 4 , 5 . As group cohesion maximises the collective agency and power of a group 5 , 12 , 38 , 69 , the strategic alignment of attitudes should be especially pertinent in situations of threat, such as the Russian invasion of Ukraine. Finding that those who engaged in attitudinal interaction aligned their attitudes more than those who privately expressed their attitudes in the context of the Ukraine crisis, may provide preliminary support that people are motivated to align their attitudes to achieve collective power. Indeed, this would reflect the identity mobilisation dimension of the social identity performance model 2 . Klein and colleagues explain that, not only can identity performance serve to consolidate identity but also to mobilise it. That is, identity performance can serve to foster ingroup related action. It is difficult to ascertain from the current results whether or not people performatively expressed attitudes to mobilise identity, and future research is needed to probe these effects further.

Limitations

In study one, we captured tweets only in the English language. This is a one-sided, biased representation of online interaction revolving around the war in Ukraine. In particular, it fails to account for the languages of the two countries directly involved in the war. Future research should assess these dynamics accounting for tweets and hashtags that were popular in other languages. Furthermore, we may not have identified all popular English hashtags prior to data collection, thus certain attitudes may have been excluded at the data collection stage. Nonetheless, it is not necessary to capture every attitude expressed online to see a pattern of attitude responses thus, the patterns we observe should not be detrimentally affected by any bias introduced at the data collection stage.

In the current paper, we conceptualise hashtags as attitude expressions. While hashtags can be considered to serve the same communicative function as attitude expressions, it is important to note that this is not always the case (see ref. 70 for a discussion on the communicative functions of hashtags).

Theoretically, we expect that, in line with the social identity performance model 2 , people can be motivated to publicly express attitudes online in order to consolidate and mobilise their identity. We expect that people can look to others close to them in a social network and share similar attitudes to them because they are perceived to share an online social identity and because people are motivated to consolidate and mobilise this identity. It is important to note that, while the online patterns of attitude-sharing we observed in the Twitter study are in line with these theoretical suggestions, we have not ruled out alternative explanations as this was merely an exploratory observational study of large-scale attitude sharing patterns.

In study two we capture dyadic interaction across 11 rounds of attitude sharing. Of course, real-life interaction online often involves more indirect interaction, much larger opinion systems, and with much larger numbers of people. While study two does capture naturally occurring attitudinal interactions it does not fully capture the dynamics of online attitude expression in the real world. Although we do capture these dynamics in the more ecologically valid twitter context, future research should attempt to replicate these results with larger audience sizes in a controlled experimental context.

Furthermore, in real-life contexts where people are repeatedly exposed to streams of opinions (likes, shares, tweets, upvotes etc.), opinion congruence may have a larger impact than our eleven-round experiment implies. This is because frequently repeated small exposures can have large influences in iterative systems via cyclical processes 71 . Future research should assess whether repeated attitude exposure across numerous iterations can compound to have large-scale effects.

Regarding replicability, we expect the process of synchronization of attitudes and identity in groups to be quite universal and highly replicable, although the specific form and functions of synchronization in a given context will be subject to abstract agreements on group norms and self-stereotypes. For example, academics value independence, and therefore disagreement (within bounds, and not on core issues) is normative for such groups in academic contexts. Therefore, we expect the very general process we capture in this paper to be creatively and strategically adapted by different groups in different contexts, but for the process of synchronization on core attitudes to be important for group emergence and identification of individuals to groups in most contexts.

Additionally, attitudes dynamically oscillate in and out of group identities; and opinion-based groups, and inter-group alliances, can be ephemeral 39 , 65 , 72 . Therefore, we do not expect this study to replicate directly, at least, not for long. We have captured emergent dynamic opinion-based groups by identifying a topic of great social importance where these processes were clearly at work at the time of the study. When the same design, with the same items, is run at a different moment of history, agreement may not have quite the same import. A full replication would require redesigning the study around whatever issues people are cohering around at that time. On the other hand, we do also expect people to form ephemeral opinion-based groups and identifications on less important, or even arbitrary, topics, as in minimal-group settings 10 . However, people would not be able to recognize allies based on their first answers if they are not already familiar with the emergent group structure related to the topic, and our key manipulation would be less potent.

Conclusions

These findings provide some insight into how particular attitudes enter the social world, why certain attitudes propagate (go viral), and how social identities are constructed and defined in the online context through attitude expression. Broadly speaking, these results provide evidence to suggest that expressed attitudes are constrained by the group dynamic between attitude expressors and audiences, and that attitude expressors are active agents in the construction of identity, expressing attitudes performatively to construct and consolidate identities. These results also have practical implications, strongly suggesting that online attitude expressions involving attitudes that clearly reflect or represent group identity may be more amenable to online propagation and have particularly strong influence on future attitudes.

Data availability

All data generated and used in the current studies are available at the following link. ( https://osf.io/w8shc/?view_only=34d38b51dc8a4691bf4fec2dbe3cde66 ).

Code availability

All analytic code is available at the following link. ( https://osf.io/w8shc/?view_only=34d38b51dc8a4691bf4fec2dbe3cde66 ).

Hogg, M. A. Uncertainty-identity theory. In M. Zanna (Ed.), Advances in Experimental Social Psychology , 39 , 69–126 (Elsevier, 2007).

Klein, O., Spears, R. & Reicher, S. Social identity performance: extending the strategic side of SIDE. Personal. Social Psychol. Rev. 11 , 28–45 (2007).

Article   Google Scholar  

Durrheim, K., Quayle, M. & Dixon, J. The struggle for the nature of ‘prejudice’: ‘prejudice’ expression as identity performance. Political Psychol. 37 , 17–35 (2016).

Quayle, M. A performative network theory of attitudes PsyArxiv , https://doi.org/10.31234/osf.io/mh4z8 (2020).

Figgou, L. & Condor, S. Irrational categorization, natural intolerance and reasonable discrimination: lay representations of prejudice and racism. Br. J. Soc. Psychol. 45 , 219–243 (2006).

Article   PubMed   Google Scholar  

Reicher, S. The context of social identity: domination, resistance, and change. Political Psychol. 25 , 921–945 (2004).

Haslam, S. A. & Reicher, S. Identity entrepreneurship and the consequences of identity failure: the dynamics of leadership in the BBC prison study. Soc. Psychol. Q. 70 , 125–147 (2007).

Reicher, S. D., Spears, R. & Postmes, T. A social identity model of deindividuation phenomena. Eur. Rev. Soc. Psychol. 6 , 161–198 (1995).

Reicher, S. D., Haslam, S. A., Spears, R. & Reynolds, K. J. A social mind: the context of John Turner’s work and its influence. Eur. Rev. Soc. Psychol. 23 , 344–385 (2012).

O’Reilly, C., Maher, P. J., Smith, E. M., MacCarron, P. & Quayle, M. Social identity emergence in attitude interactions and the identity strengthening effects of cumulative attitude agreement. Eur. J. Soc. Psychol . https://doi.org/10.1002/ejsp.3000 (2023).

Bliuc, A. M., McGarty, C., Reynolds, K. & Muntele, D. Opinion‐based group membership as a predictor of commitment to political action. Eur. J. Soc. Psychol. 37 , 19–32 (2007).

Bliuc, A. et al. Public division about climate change rooted in conflicting socio-political identities. Nat. Clim. Change 5 , 226–229 (2015).

Thomas, E. F. & McGarty, C. A. The role of efficacy and moral outrage norms in creating the potential for international development activism through group-based interaction. Br. J. Soc. Psychol. 48 , 115–134 (2009).

Thomas, E. F., McGarty, C. & Mavor, K. Group interaction as the crucible of social identity formation: a glimpse at the foundations of social identities for collective action. Group Process. Intergroup Relat. 19 , 137–151 (2015).

Thomas, E. F., McGarty, C., Stuart, A., Smith, L. G. & Bourgeois, L. Reaching consensus promotes the internalization of commitment to social change. Group Process. Intergroup Relat. 22 , 615–630 (2019).

Lüders, A., Dinkelberg, A. & Quayle, M. Becoming ‘us’ in digital spaces: how online users creatively and strategically exploit social media affordances to build up social identity. Acta Psychol . 228 . https://doi.org/10.1016/j.actpsy.2022.103643 (2022).

O’Reilly, C., Maher, P. J., Lüders, A. & Quayle, M. Sharing is caring: how sharing opinions online can connect people into groups and foster identification. Acta Psychol. 230 , 103751 (2022).

Klein, O., Licata, L., Azzi, A. E. & Durala, I. ‘How European am I?’: Prejudice expression and the presentation of social identity. Self Identity 2 , 251–264 (2003).

Tajfel, H. & Turner, J. C. An integrative theory of intergroup conflict, in W. G. Austin & S. Worchel (eds), The Social Psychology of Intergroup Relations. 33–47 (Monterey, 1979).

Turner, J. C. Introducing the problem: individual and group. Rediscovering the Social Group: A Self-Categorization Theory, 1–18. (Blackwell, 1987).

Turner, J. C. Towards a cognitive redefinition of the social group. Cahiers Psychol. Cognit./Current Psychology of Cognition 1 , 93–118 (1981).

Google Scholar  

Billig, M. & Tajfel, H. Social categorization and similarity in intergroup behaviour. Eur. J. Soc. Psychol. 3 , 27–52 (1973).

Spears, R. The interaction between the individual and the collective self: self-categorization in context. In C. Sedikides & M. B. Brewer (Eds.), Individual Self, Relational Self, Collective Self 171–198 (Psychology Press, 2001).

Reicher, S. & Haslam, S. A. Towards a ’science of movement’: identity, authority and influence in the production of social stability and social change. J. Soc. Political Psychol. 1 , 112–131 (2013).

Spears, R. Social identity model of deindividuation effects. The International Encyclopedia of Media Effects 1–9 (Wiley, 2017).

Noel, J. G., Wann, D. L. & Branscombe, N. R. Peripheral ingroup membership status and public negativity toward outgroups. J. Personal. Soc. Psychol. 68 , 127–137 (1995).

Douglas, K. M. & McGarty, C. Identifiability and selfpresentation: computer-mediated communication and intergroup communication. Br. J. Soc. Psychol. 40 , 399–416 (2001).

Smith, J. (2010). Referent informational influence theory. In J. M. Levine & M. A. Hogg. (Eds), Encyclopedia of Group Processes and Intergroup Relations . 687–689 (Sage, 2010).

Livingstone, A. G., Haslam, S. A., Postmes, T. & Jetten, J. ‘We are, therefore we should’: evidence that in‐group identification mediates the acquisition of in‐group norms 1. J. Appl. Soc. Psychol. 41 , 1857–1876 (2011).

McGarty, C. & Turner, J. C. The effects of categorization on social judgement. Br. J. Soc. Psychol. 31 , 253–268 (1992).

Abrams, D., Wetherell, M., Cochrane, S., Hogg, M. A. & Turner, J. C. Knowing what to think by knowing who you are: Self‐categorization and the nature of norm formation, conformity and group polarization. Br. J. Soc. Psychol. 29 , 97–119 (1990).

Turner, J. C. Social Influence (Open University Press, 1991).

Akfırat, S. et al. Social identification and collective action participation in the internet age: a meta-analysis. Cyberpsychology. J. Psychosoc. Res. Cyberspace , 15 , https://doi.org/10.5817/CP2021-4-10 (2020).

Musgrove, L. & McGarty, C. Opinion-based group membership as a predictor of collective emotional responses and support for pro-and anti-war action. Soc. Psychol. 39 , 37–47 (2008).

Maher, P. J., MacCarron, P. & Quayle, M. Mapping public health responses with attitude networks: the emergence of opinion‐based groups in the UK’s early COVID‐19 response phase. Br. J. Soc. Psychol. 59 , 641–652 (2020).

Article   PubMed   PubMed Central   Google Scholar  

Bongiorno, R., McGarty, C., Kurz, T., Haslam, S. A. & Sibley, C. G. Mobilizing cause supporters through group‐based interaction. J. Appl. Soc. Psychol. 46 , 203–215 (2016).

McGarty, C., Bliuc, A. M., Thomas, E. & Bongiorno, R. Collective action as the material expression of opinion-based group membership. J. Soc. Issues 65 , 839 (2009).

Reicher, S. & Haslam, S. A. Rethinking the psychology of tyranny: the BBC prison study. Br. J. Soc. Psychol. 45 , 1–40 (2006).

Oakes, P. J., Haslam, S. A. & Turner, J. C. Stereotyping and Social Reality. (Blackwell, 1994).

Karizat, N., Delmonaco, D., Eslami, M. & Andalibi, N. Algorithmic folk theories and identity: How TikTok users co-produce knowledge of identity and engage in algorithmic resistance. Proc. ACM on Hum.-Comput. Interaction, 5(CSCW2), 5 , 1–44 (2021).

Jenkins, E. M., Zaher, Z., Tikkanen, S. A. & Ford, J. L. Creative identity (re) Construction, creative community building, and creative resistance: a qualitative analysis of queer ingroup members’ tweets after the Orlando Shooting. Comput. Hum. Behav. 101 , 14–21 (2019).

Talbot, C. V., Talbot, A., Roe, D. J. & Briggs, P. The management of LGBTQ+ identities on social media: a student perspective. New Media Soc. 24 , 1729–1750 (2022).

Marwick, A. E. & Boyd, D. I tweet honestly, I tweet passionately: twitter users, context collapse, and the imagined audience. New Media Soc. 13 , 114–133 (2011).

Boyd, D. Friends, friendsters, and top 8: writing community into being on social network sites. First monday. https://doi.org/10.5210/fm.v11i12.1418 (2006).

Ellison, N., Heino, R. & Gibbs, J. Managing impressions online: Self-presentation processes in the online dating environment. J. Comput.-Mediat. Commun. 11 , 415–441 (2006).

Mishra, S. & Ismail, A. Performances of the online self for networked audiences: an introduction to the special issue. J. Hum. Values , 24 , vii–xiii (2018).

Litt, E. & Hargittai, E. The imagined audience on social network sites. Social Media+ Society, 2, 2056305116633482 (2016).

Mooijman, M., Hoover, J., Lin, Y., Ji, H. & Dehghani, M. Moralization in social networks and the emergence of violence during protests. Nat. Hum. Behav. 2 , 389–396 (2018).

Traag, V. A., Waltman, L. & Van Eck, N. J. From Louvain to Leiden: guaranteeing well-connected communities. Sci. Rep. 9 , 5233 (2019).

Rosvall, M., Axelsson, D. & Bergstrom, C. T. The map equation. Eur. Phys. J. Special Top. 178 , 13–23 (2009).

Nielsen, F. Å. A new ANEW: evaluation of a word list for sentiment analysis in microblogs. arXiv preprint arXiv 1103, 2903 (2011).

Brandt, M. J., Sibley, C. G. & Osborne, D. What is central to political belief system networks? Personal. Soc. Psychol. Bull. 45 , 1352–1364 (2019).

Prolific team. Is online crowdsourcing a legitimate alternative to lab-based research? https://researcher-help.prolific.co/hc/en-gb/articles/360009223133-Is-online-crowdsourcing-a-legitimate-alternative-to-lab-based-research-#heading-0 (2023).

Buhrmester, M., Kwang, T. & Gosling, S. D. Amazon’s Mechanical Turk: a new source of inexpensive, yet high-quality data? Perspect. Psychol. Sci. 6 , https://doi.org/10.1177/174569161039398 (2016).

Peer, E., Rothschild, D., Gordon, A., Evernden, Z. & Damer, E. Data quality of platforms and panels for online behavioral research. Behav. Res. Methods , 54 , 1–20 (2021).

Peer, E., Brandimarte, L., Samat, S. & Acquisti, A. Beyond the Turk: alternative platforms for crowdsourcing behavioral research. J. Exp. Soc. Psychol. 70 , 153–163 (2017).

Faul, F., Erdfelder, E., Buchner, A. & Lang, A. G. Statistical power analyses using G* Power 3.1: tests for correlation and regression analyses. Behav. Res. Methods 41 , 1149–1160 (2009).

Postmes, T., Haslam, S. A. & Jans, L. A single‐item measure of social identification: reliability, validity, and utility. Br. J. Soc. Psychol. 52 , 597–617 (2013).

Doosje, B., Ellemers, N. & Spears, R. Perceived intragroup variability as a function of group status and identification. J. Exp. Soc. Psychol. 31 , 410–436 (1995).

MacCarron, P., Maher, P. J. & Quayle, M. ‘Identifying opinion-based groups from survey data: a bipartite network approach.’. arXiv preprint arXiv 2012, 11392 (2020).

Dinkelberg, A., O’Sullivan, D. J. P., Quayle, M. & MacCaron, P. (2021). Detecting opinion-based groups and polarization in survey-based attitude networks and estimating question relevance. Advances in Complex Systems, 24. https://doi.org/10.1142/S0219525921500065 .

Newman, M. Networks (Oxford University Press, 2018).

Molnar, A. How to implement real-time interaction between participants in online surveys: a practical guide to SMARTRIQS. PsyArXiv. https://doi.org/10.31234/osf.io/6wmny (2019).

McGarty, C. Categorization in social psychology. Categorization in Social Psychology, 1–320 (1999).

Turner, J. C., Oakes, P. J., Haslam, S. A. & McGarty, C. Self and collective: cognition and social context. Personal. Soc. Psychol. Bull. 20 , 454–463 (1994).

Carpentras, D., Lüders, A. & Quayle, M. Mapping the global opinion space to explain anti-vaccine attraction. Sci. Rep. 12 , 6188 (2022).

Hopp, F. R., Fisher, J. T., Cornell, D., Huskey, R. & Weber, R. The extended Moral Foundations Dictionary (eMFD): development and applications of a crowd-sourced approach to extracting moral intuitions from text. Behav. Res. Methods 53 , 232–246 (2021).

Brady, W. J., Crockett, M. J. & Van Bavel, J. J. The MAD model of moral contagion: the role of motivation, attention, and design in the spread of moralized content online. Perspect. Psychol. Sci. 15 , 978–1010 (2020).

Billig, M. et al. Ideological Dilemmas: A Social Psychology of Everyday Thinking. (Sage Publications, Inc, 1988).

Laucuka, A. Communicative functions of hashtags. Econ. Cult. 15 , 56–62 (2018).

Slater, M. D. Reinforcing spirals model: conceptualizing the relationship between media content exposure and the development and maintenance of attitudes. Media Psychol. 18 , 370–395 (2015).

Smith, E. R. & Zárate, M. A. Exemplar-based model of social judgment. Psychol. Rev. 99 , 321 (1992).

Download references

Acknowledgements

We would like to acknowledge and thank Professor Ailish Hannigan for her expertise and advice regarding the statistical methods used in this paper. We also thank the Irish Research Council [grant number GOIPG/2022/545], the Science Foundation Ireland [Grant number 18/CRT/6049], and the European Research Council [grant number 802421, 2020] who provided funding for this work. Note that these funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.

Author information

Authors and affiliations.

Department of Psychology, University of Limerick, Limerick, Ireland

Caoimhe O’Reilly, Paul J. Maher, Elaine M. Smith, Pádraig MacCarron & Michael Quayle

Department of Maths, University of Limerick, Limerick, Ireland

Shane Mannion & Pádraig MacCarron

Department of Psychology, School of Applied Human Sciences, University of KwaZulu-Ng community into being on social network atal, Durban, South Africa

Michael Quayle

You can also search for this author in PubMed   Google Scholar

Contributions

Caoimhe O’Reilly: Conceptualization, Methodology, Writing – Original draft preparation, Formal analysis, Visualisation, Investigation, Software, Data curation, Project Administration. Shane Mannion: Formal analysis, Visualisation, Data curation, Writing – review and editing Paul Maher: Conceptualization, Methodology, Writing – review and editing, Supervision. Elaine M Smith: Conceptualization, Formal analysis, Writing – review and editing. Pádraig MacCarron: Formal analysis, Visualisation, Writing – review and editing. Michael Quayle: Conceptualization, Methodology, Writing – review and editing, Supervision, Funding acquisition.

Corresponding author

Correspondence to Caoimhe O’Reilly .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Peer review

Peer review information.

Communications Psychology thanks Ilka Gleibs, Olivier Klein, Andrew Pilny and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editor: Antonia Eisenkoeck and Marike Schiffer. A peer review file is available

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary information, peer review file, reporting summary, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

O’Reilly, C., Mannion, S., Maher, P.J. et al. Strategic attitude expressions as identity performance and identity creation in interaction. Commun Psychol 2 , 27 (2024). https://doi.org/10.1038/s44271-024-00076-7

Download citation

Received : 03 April 2023

Accepted : 13 March 2024

Published : 08 April 2024

DOI : https://doi.org/10.1038/s44271-024-00076-7

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

• Search Menu

Sign in through your institution

• Browse content in Arts and Humanities
• Browse content in Archaeology
• Anglo-Saxon and Medieval Archaeology
• Archaeological Methodology and Techniques
• Archaeology by Region
• Archaeology of Religion
• Archaeology of Trade and Exchange
• Biblical Archaeology
• Contemporary and Public Archaeology
• Environmental Archaeology
• Historical Archaeology
• History and Theory of Archaeology
• Industrial Archaeology
• Landscape Archaeology
• Mortuary Archaeology
• Prehistoric Archaeology
• Underwater Archaeology
• Zooarchaeology
• Browse content in Architecture
• Architectural Structure and Design
• History of Architecture
• Residential and Domestic Buildings
• Theory of Architecture
• Browse content in Art
• Art Subjects and Themes
• History of Art
• Industrial and Commercial Art
• Theory of Art
• Biographical Studies
• Byzantine Studies
• Browse content in Classical Studies
• Classical History
• Classical Philosophy
• Classical Mythology
• Classical Literature
• Classical Reception
• Classical Art and Architecture
• Classical Oratory and Rhetoric
• Greek and Roman Papyrology
• Greek and Roman Epigraphy
• Greek and Roman Law
• Greek and Roman Archaeology
• Late Antiquity
• Religion in the Ancient World
• Digital Humanities
• Browse content in History
• Colonialism and Imperialism
• Diplomatic History
• Environmental History
• Genealogy, Heraldry, Names, and Honours
• Genocide and Ethnic Cleansing
• Historical Geography
• History by Period
• History of Emotions
• History of Agriculture
• History of Education
• History of Gender and Sexuality
• Industrial History
• Intellectual History
• International History
• Labour History
• Legal and Constitutional History
• Local and Family History
• Maritime History
• Military History
• National Liberation and Post-Colonialism
• Oral History
• Political History
• Public History
• Regional and National History
• Revolutions and Rebellions
• Slavery and Abolition of Slavery
• Social and Cultural History
• Theory, Methods, and Historiography
• Urban History
• World History
• Browse content in Language Teaching and Learning
• Language Learning (Specific Skills)
• Language Teaching Theory and Methods
• Browse content in Linguistics
• Applied Linguistics
• Cognitive Linguistics
• Computational Linguistics
• Forensic Linguistics
• Grammar, Syntax and Morphology
• Historical and Diachronic Linguistics
• History of English
• Language Evolution
• Language Reference
• Language Acquisition
• Language Variation
• Language Families
• Lexicography
• Linguistic Anthropology
• Linguistic Theories
• Linguistic Typology
• Phonetics and Phonology
• Psycholinguistics
• Sociolinguistics
• Translation and Interpretation
• Writing Systems
• Browse content in Literature
• Bibliography
• Children's Literature Studies
• Literary Studies (Romanticism)
• Literary Studies (American)
• Literary Studies (Asian)
• Literary Studies (European)
• Literary Studies (Eco-criticism)
• Literary Studies (Modernism)
• Literary Studies - World
• Literary Studies (1500 to 1800)
• Literary Studies (19th Century)
• Literary Studies (20th Century onwards)
• Literary Studies (African American Literature)
• Literary Studies (British and Irish)
• Literary Studies (Early and Medieval)
• Literary Studies (Fiction, Novelists, and Prose Writers)
• Literary Studies (Gender Studies)
• Literary Studies (Graphic Novels)
• Literary Studies (History of the Book)
• Literary Studies (Plays and Playwrights)
• Literary Studies (Poetry and Poets)
• Literary Studies (Postcolonial Literature)
• Literary Studies (Queer Studies)
• Literary Studies (Science Fiction)
• Literary Studies (Travel Literature)
• Literary Studies (War Literature)
• Literary Studies (Women's Writing)
• Literary Theory and Cultural Studies
• Mythology and Folklore
• Shakespeare Studies and Criticism
• Browse content in Media Studies
• Browse content in Music
• Applied Music
• Dance and Music
• Ethics in Music
• Ethnomusicology
• Gender and Sexuality in Music
• Medicine and Music
• Music Cultures
• Music and Media
• Music and Religion
• Music and Culture
• Music Education and Pedagogy
• Music Theory and Analysis
• Musical Scores, Lyrics, and Libretti
• Musical Structures, Styles, and Techniques
• Musicology and Music History
• Performance Practice and Studies
• Race and Ethnicity in Music
• Sound Studies
• Browse content in Performing Arts
• Browse content in Philosophy
• Aesthetics and Philosophy of Art
• Epistemology
• Feminist Philosophy
• History of Western Philosophy
• Metaphysics
• Moral Philosophy
• Non-Western Philosophy
• Philosophy of Language
• Philosophy of Mind
• Philosophy of Perception
• Philosophy of Science
• Philosophy of Action
• Philosophy of Law
• Philosophy of Religion
• Philosophy of Mathematics and Logic
• Practical Ethics
• Social and Political Philosophy
• Browse content in Religion
• Biblical Studies
• Christianity
• East Asian Religions
• History of Religion
• Judaism and Jewish Studies
• Qumran Studies
• Religion and Education
• Religion and Health
• Religion and Politics
• Religion and Science
• Religion and Law
• Religion and Art, Literature, and Music
• Religious Studies
• Browse content in Society and Culture
• Cookery, Food, and Drink
• Cultural Studies
• Customs and Traditions
• Ethical Issues and Debates
• Hobbies, Games, Arts and Crafts
• Natural world, Country Life, and Pets
• Popular Beliefs and Controversial Knowledge
• Sports and Outdoor Recreation
• Technology and Society
• Travel and Holiday
• Visual Culture
• Browse content in Law
• Arbitration
• Browse content in Company and Commercial Law
• Commercial Law
• Company Law
• Browse content in Comparative Law
• Systems of Law
• Competition Law
• Browse content in Constitutional and Administrative Law
• Government Powers
• Judicial Review
• Local Government Law
• Military and Defence Law
• Parliamentary and Legislative Practice
• Construction Law
• Contract Law
• Browse content in Criminal Law
• Criminal Procedure
• Criminal Evidence Law
• Sentencing and Punishment
• Employment and Labour Law
• Environment and Energy Law
• Browse content in Financial Law
• Banking Law
• Insolvency Law
• History of Law
• Human Rights and Immigration
• Intellectual Property Law
• Browse content in International Law
• Private International Law and Conflict of Laws
• Public International Law
• IT and Communications Law
• Jurisprudence and Philosophy of Law
• Law and Politics
• Law and Society
• Browse content in Legal System and Practice
• Courts and Procedure
• Legal Skills and Practice
• Primary Sources of Law
• Regulation of Legal Profession
• Medical and Healthcare Law
• Browse content in Policing
• Criminal Investigation and Detection
• Police and Security Services
• Police Procedure and Law
• Police Regional Planning
• Browse content in Property Law
• Personal Property Law
• Study and Revision
• Terrorism and National Security Law
• Browse content in Trusts Law
• Wills and Probate or Succession
• Browse content in Medicine and Health
• Browse content in Allied Health Professions
• Arts Therapies
• Clinical Science
• Dietetics and Nutrition
• Occupational Therapy
• Operating Department Practice
• Physiotherapy
• Radiography
• Speech and Language Therapy
• Browse content in Anaesthetics
• General Anaesthesia
• Neuroanaesthesia
• Clinical Neuroscience
• Browse content in Clinical Medicine
• Acute Medicine
• Cardiovascular Medicine
• Clinical Genetics
• Clinical Pharmacology and Therapeutics
• Dermatology
• Endocrinology and Diabetes
• Gastroenterology
• Genito-urinary Medicine
• Geriatric Medicine
• Infectious Diseases
• Medical Toxicology
• Medical Oncology
• Pain Medicine
• Palliative Medicine
• Rehabilitation Medicine
• Respiratory Medicine and Pulmonology
• Rheumatology
• Sleep Medicine
• Sports and Exercise Medicine
• Community Medical Services
• Critical Care
• Emergency Medicine
• Forensic Medicine
• Haematology
• History of Medicine
• Browse content in Medical Skills
• Clinical Skills
• Communication Skills
• Nursing Skills
• Surgical Skills
• Browse content in Medical Dentistry
• Oral and Maxillofacial Surgery
• Paediatric Dentistry
• Restorative Dentistry and Orthodontics
• Surgical Dentistry
• Medical Ethics
• Medical Statistics and Methodology
• Browse content in Neurology
• Clinical Neurophysiology
• Neuropathology
• Nursing Studies
• Browse content in Obstetrics and Gynaecology
• Gynaecology
• Occupational Medicine
• Ophthalmology
• Otolaryngology (ENT)
• Browse content in Paediatrics
• Neonatology
• Browse content in Pathology
• Chemical Pathology
• Clinical Cytogenetics and Molecular Genetics
• Histopathology
• Medical Microbiology and Virology
• Patient Education and Information
• Browse content in Pharmacology
• Psychopharmacology
• Browse content in Popular Health
• Caring for Others
• Complementary and Alternative Medicine
• Self-help and Personal Development
• Browse content in Preclinical Medicine
• Cell Biology
• Molecular Biology and Genetics
• Reproduction, Growth and Development
• Primary Care
• Professional Development in Medicine
• Browse content in Psychiatry
• Addiction Medicine
• Child and Adolescent Psychiatry
• Forensic Psychiatry
• Learning Disabilities
• Old Age Psychiatry
• Psychotherapy
• Browse content in Public Health and Epidemiology
• Epidemiology
• Public Health
• Browse content in Radiology
• Clinical Radiology
• Interventional Radiology
• Nuclear Medicine
• Radiation Oncology
• Reproductive Medicine
• Browse content in Surgery
• Cardiothoracic Surgery
• Gastro-intestinal and Colorectal Surgery
• General Surgery
• Neurosurgery
• Paediatric Surgery
• Peri-operative Care
• Plastic and Reconstructive Surgery
• Surgical Oncology
• Transplant Surgery
• Trauma and Orthopaedic Surgery
• Vascular Surgery
• Browse content in Science and Mathematics
• Browse content in Biological Sciences
• Aquatic Biology
• Biochemistry
• Bioinformatics and Computational Biology
• Developmental Biology
• Ecology and Conservation
• Evolutionary Biology
• Genetics and Genomics
• Microbiology
• Molecular and Cell Biology
• Natural History
• Plant Sciences and Forestry
• Research Methods in Life Sciences
• Structural Biology
• Systems Biology
• Zoology and Animal Sciences
• Browse content in Chemistry
• Analytical Chemistry
• Computational Chemistry
• Crystallography
• Environmental Chemistry
• Industrial Chemistry
• Inorganic Chemistry
• Materials Chemistry
• Medicinal Chemistry
• Mineralogy and Gems
• Organic Chemistry
• Physical Chemistry
• Polymer Chemistry
• Study and Communication Skills in Chemistry
• Theoretical Chemistry
• Browse content in Computer Science
• Artificial Intelligence
• Computer Architecture and Logic Design
• Game Studies
• Human-Computer Interaction
• Mathematical Theory of Computation
• Programming Languages
• Software Engineering
• Systems Analysis and Design
• Virtual Reality
• Browse content in Computing
• Business Applications
• Computer Security
• Computer Games
• Computer Networking and Communications
• Digital Lifestyle
• Graphical and Digital Media Applications
• Operating Systems
• Browse content in Earth Sciences and Geography
• Atmospheric Sciences
• Environmental Geography
• Geology and the Lithosphere
• Maps and Map-making
• Meteorology and Climatology
• Oceanography and Hydrology
• Palaeontology
• Physical Geography and Topography
• Regional Geography
• Soil Science
• Urban Geography
• Browse content in Engineering and Technology
• Agriculture and Farming
• Biological Engineering
• Civil Engineering, Surveying, and Building
• Electronics and Communications Engineering
• Energy Technology
• Engineering (General)
• Environmental Science, Engineering, and Technology
• History of Engineering and Technology
• Mechanical Engineering and Materials
• Technology of Industrial Chemistry
• Transport Technology and Trades
• Browse content in Environmental Science
• Applied Ecology (Environmental Science)
• Conservation of the Environment (Environmental Science)
• Environmental Sustainability
• Environmentalist Thought and Ideology (Environmental Science)
• Management of Land and Natural Resources (Environmental Science)
• Natural Disasters (Environmental Science)
• Nuclear Issues (Environmental Science)
• Pollution and Threats to the Environment (Environmental Science)
• Social Impact of Environmental Issues (Environmental Science)
• History of Science and Technology
• Browse content in Materials Science
• Ceramics and Glasses
• Composite Materials
• Metals, Alloying, and Corrosion
• Nanotechnology
• Browse content in Mathematics
• Applied Mathematics
• Biomathematics and Statistics
• History of Mathematics
• Mathematical Education
• Mathematical Finance
• Mathematical Analysis
• Numerical and Computational Mathematics
• Probability and Statistics
• Pure Mathematics
• Browse content in Neuroscience
• Cognition and Behavioural Neuroscience
• Development of the Nervous System
• Disorders of the Nervous System
• History of Neuroscience
• Invertebrate Neurobiology
• Molecular and Cellular Systems
• Neuroendocrinology and Autonomic Nervous System
• Neuroscientific Techniques
• Sensory and Motor Systems
• Browse content in Physics
• Astronomy and Astrophysics
• Atomic, Molecular, and Optical Physics
• Biological and Medical Physics
• Classical Mechanics
• Computational Physics
• Condensed Matter Physics
• Electromagnetism, Optics, and Acoustics
• History of Physics
• Mathematical and Statistical Physics
• Measurement Science
• Nuclear Physics
• Particles and Fields
• Plasma Physics
• Quantum Physics
• Relativity and Gravitation
• Semiconductor and Mesoscopic Physics
• Browse content in Psychology
• Affective Sciences
• Clinical Psychology
• Cognitive Psychology
• Cognitive Neuroscience
• Criminal and Forensic Psychology
• Developmental Psychology
• Educational Psychology
• Evolutionary Psychology
• Health Psychology
• History and Systems in Psychology
• Music Psychology
• Neuropsychology
• Organizational Psychology
• Psychological Assessment and Testing
• Psychology of Human-Technology Interaction
• Psychology Professional Development and Training
• Research Methods in Psychology
• Social Psychology
• Browse content in Social Sciences
• Browse content in Anthropology
• Anthropology of Religion
• Human Evolution
• Medical Anthropology
• Physical Anthropology
• Regional Anthropology
• Social and Cultural Anthropology
• Theory and Practice of Anthropology
• Browse content in Business and Management
• Business Ethics
• Business Strategy
• Business History
• Business and Technology
• Business and Government
• Business and the Environment
• Comparative Management
• Corporate Governance
• Corporate Social Responsibility
• Entrepreneurship
• Health Management
• Human Resource Management
• Industrial and Employment Relations
• Industry Studies
• Information and Communication Technologies
• International Business
• Knowledge Management
• Management and Management Techniques
• Operations Management
• Organizational Theory and Behaviour
• Pensions and Pension Management
• Public and Nonprofit Management
• Strategic Management
• Supply Chain Management
• Browse content in Criminology and Criminal Justice
• Criminal Justice
• Criminology
• Forms of Crime
• International and Comparative Criminology
• Youth Violence and Juvenile Justice
• Development Studies
• Browse content in Economics
• Agricultural, Environmental, and Natural Resource Economics
• Asian Economics
• Behavioural Finance
• Behavioural Economics and Neuroeconomics
• Econometrics and Mathematical Economics
• Economic History
• Economic Systems
• Economic Methodology
• Economic Development and Growth
• Financial Markets
• Financial Institutions and Services
• General Economics and Teaching
• Health, Education, and Welfare
• History of Economic Thought
• International Economics
• Labour and Demographic Economics
• Law and Economics
• Macroeconomics and Monetary Economics
• Microeconomics
• Public Economics
• Urban, Rural, and Regional Economics
• Welfare Economics
• Browse content in Education
• Adult Education and Continuous Learning
• Care and Counselling of Students
• Early Childhood and Elementary Education
• Educational Equipment and Technology
• Educational Strategies and Policy
• Higher and Further Education
• Organization and Management of Education
• Philosophy and Theory of Education
• Schools Studies
• Secondary Education
• Teaching of a Specific Subject
• Teaching of Specific Groups and Special Educational Needs
• Teaching Skills and Techniques
• Browse content in Environment
• Applied Ecology (Social Science)
• Climate Change
• Conservation of the Environment (Social Science)
• Environmentalist Thought and Ideology (Social Science)
• Natural Disasters (Environment)
• Social Impact of Environmental Issues (Social Science)
• Browse content in Human Geography
• Cultural Geography
• Economic Geography
• Political Geography
• Browse content in Interdisciplinary Studies
• Communication Studies
• Museums, Libraries, and Information Sciences
• Browse content in Politics
• African Politics
• Asian Politics
• Chinese Politics
• Comparative Politics
• Conflict Politics
• Elections and Electoral Studies
• Environmental Politics
• European Union
• Foreign Policy
• Gender and Politics
• Human Rights and Politics
• Indian Politics
• International Relations
• International Organization (Politics)
• International Political Economy
• Irish Politics
• Latin American Politics
• Middle Eastern Politics
• Political Behaviour
• Political Economy
• Political Institutions
• Political Methodology
• Political Communication
• Political Philosophy
• Political Sociology
• Political Theory
• Politics and Law
• Politics of Development
• Public Policy
• Public Administration
• Quantitative Political Methodology
• Regional Political Studies
• Russian Politics
• Security Studies
• State and Local Government
• UK Politics
• US Politics
• Browse content in Regional and Area Studies
• African Studies
• Asian Studies
• East Asian Studies
• Japanese Studies
• Latin American Studies
• Middle Eastern Studies
• Native American Studies
• Scottish Studies
• Browse content in Research and Information
• Research Methods
• Browse content in Social Work
• Addictions and Substance Misuse
• Adoption and Fostering
• Care of the Elderly
• Child and Adolescent Social Work
• Couple and Family Social Work
• Direct Practice and Clinical Social Work
• Emergency Services
• Human Behaviour and the Social Environment
• International and Global Issues in Social Work
• Mental and Behavioural Health
• Social Justice and Human Rights
• Social Policy and Advocacy
• Social Work and Crime and Justice
• Social Work Macro Practice
• Social Work Practice Settings
• Social Work Research and Evidence-based Practice
• Welfare and Benefit Systems
• Browse content in Sociology
• Childhood Studies
• Community Development
• Comparative and Historical Sociology
• Economic Sociology
• Gender and Sexuality
• Gerontology and Ageing
• Health, Illness, and Medicine
• Marriage and the Family
• Migration Studies
• Occupations, Professions, and Work
• Organizations
• Population and Demography
• Race and Ethnicity
• Social Theory
• Social Movements and Social Change
• Social Research and Statistics
• Social Stratification, Inequality, and Mobility
• Sociology of Religion
• Sociology of Education
• Sport and Leisure
• Urban and Rural Studies
• Browse content in Warfare and Defence
• Defence Strategy, Planning, and Research
• Land Forces and Warfare
• Military Administration
• Military Life and Institutions
• Naval Forces and Warfare
• Other Warfare and Defence Issues
• Peace Studies and Conflict Resolution
• Weapons and Equipment

• < Previous chapter
• Next chapter >

6 Identity Theory in a Digital Age

• Published: January 2016
• Cite Icon Cite
• Permissions Icon Permissions

This chapter shows how identity theory is useful in theorizing digitally mediated sociality and how new technological advancements push the identity theory model. Beginning with social structures, it examines how social media platforms are at once “open” and “closed,” posing questions about the availability of identities for social media users. It addresses the identity processes of performance and verification, disentangling the ways that users both gain and relinquish control over their identity meanings. It shows how the outcomes of identity verification processes—self-worth, self-efficacy, and authenticity—help explain complex and sometimes contradictory findings about the relationship between digitally mediated interaction and psychological well-being. It concludes with two areas of emergent research within identity theory: multiple identities and identity change, for which social media provoke essential questions and provide important sources of data. The chapter shows that advances in social media studies dovetail in theoretically fruitful ways with advances in identity theory.

Signed in as

Institutional accounts.

• Google Scholar Indexing
• GoogleCrawler [DO NOT DELETE]

Personal account

• Sign in with email/username & password
• Get email alerts
• Save searches
• Purchase content
• Activate your purchase/trial code
• Add your ORCID iD

Institutional access

Sign in with a library card.

• Sign in with username/password
• Recommend to your librarian
• Institutional account management
• Get help with access

Access to content on Oxford Academic is often provided through institutional subscriptions and purchases. If you are a member of an institution with an active account, you may be able to access content in one of the following ways:

IP based access

Typically, access is provided across an institutional network to a range of IP addresses. This authentication occurs automatically, and it is not possible to sign out of an IP authenticated account.

Choose this option to get remote access when outside your institution. Shibboleth/Open Athens technology is used to provide single sign-on between your institution’s website and Oxford Academic.

• Click Sign in through your institution.
• Select your institution from the list provided, which will take you to your institution's website to sign in.
• When on the institution site, please use the credentials provided by your institution. Do not use an Oxford Academic personal account.
• Following successful sign in, you will be returned to Oxford Academic.

If your institution is not listed or you cannot sign in to your institution’s website, please contact your librarian or administrator.

Enter your library card number to sign in. If you cannot sign in, please contact your librarian.

Society Members

Society member access to a journal is achieved in one of the following ways:

Sign in through society site

Many societies offer single sign-on between the society website and Oxford Academic. If you see ‘Sign in through society site’ in the sign in pane within a journal:

• Click Sign in through society site.
• When on the society site, please use the credentials provided by that society. Do not use an Oxford Academic personal account.

If you do not have a society account or have forgotten your username or password, please contact your society.

Sign in using a personal account

Some societies use Oxford Academic personal accounts to provide access to their members. See below.

A personal account can be used to get email alerts, save searches, purchase content, and activate subscriptions.

Some societies use Oxford Academic personal accounts to provide access to their members.

Viewing your signed in accounts

Click the account icon in the top right to:

• View your signed in personal account and access account management features.
• View the institutional accounts that are providing access.

Signed in but can't access content

Oxford Academic is home to a wide variety of products. The institutional subscription may not cover the content that you are trying to access. If you believe you should have access to that content, please contact your librarian.

For librarians and administrators, your personal account also provides access to institutional account management. Here you will find options to view and activate subscriptions, manage institutional settings and access options, access usage statistics, and more.

Our books are available by subscription or purchase to libraries and institutions.

• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Rights and permissions
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

The Presentation of Self in Everyday Life

Theo Wargo / Getty Images

• Recommended Reading
• Key Concepts
• Major Sociologists
• News & Issues
• Research, Samples, and Statistics
• Archaeology

The Presentation of Self in Everyday Life is a book that was published in the U.S. in 1959, written by sociologist  Erving Goffman . In it, Goffman uses the imagery of theater in order to portray the nuances and significance of face-to-face social interaction. Goffman puts forth a theory of social interaction that he refers to as the dramaturgical model of social life.

According to Goffman, social interaction may be likened to a theater, and people in everyday life to actors on a stage, each playing a variety of roles. The audience consists of other individuals who observe the role-playing and react to the performances. In social interaction, like in theatrical performances, there is a 'front stage' region where the actors are on stage  before an audience, and their consciousness of that audience and the audience's expectations for the role they should play influence the actor's behavior. There is also a back region, or 'backstage,' where individuals can relax, be themselves, and the role or identity that they play when they are in front of others.

Central to the book and Goffman's theory is the idea that people, as they interact together in social settings, are constantly engaged in the process of "impression management," wherein each tries to present themselves and behave in a way that will prevent the embarrassment of themselves or others. This is primarily done by each person that is part of the interaction working to ensure that all parties have the same "definition of the situation," meaning that all understand what is meant to happen in that situation, what to expect from the others involved, and thus how they themselves should behave.

Though written over half a century ago,  The Presentation of Self in Everday Life  remains one of the most famous and widely taught sociology books, which was listed as the 10th most important sociology book of the twentieth century by the International Sociological Association in 1998.

Performance

Goffman uses the term ‘performance’ to refer to all the activity of an individual in front of a particular set of observers, or audience. Through this performance, the individual, or actor, gives meaning to themselves, to others, and to their situation. These performances deliver impressions to others, which communicates information that confirms the identity of the actor in that situation. The actor may or may not be aware of their performance or have an objective for their performance, however, the audience is constantly attributing meaning to it and to the actor.

The setting for the performance includes the scenery, props, and location in which the interaction takes place. Different settings will have different audiences and will thus require the actor to alter his performances for each setting.

Appearance functions to portray to the audience the performer’s social statuses. Appearance also tells us of the individual’s temporary social state or role, for example, whether he is engaging in work (by wearing a uniform), informal recreation, or a formal social activity. Here, dress and props serve to communicate things that have socially ascribed meaning, like gender , status, occupation, age, and personal commitments.

Manner refers to how the individual plays the role and functions to warn the audience of how the performer will act or seek to act in a role (for example, dominant, aggressive, receptive, etc.). Inconsistency and contradiction between appearance and manner may occur and will confuse and upset an audience. This can happen, for example, when one does not present himself or behave in accordance with his perceived social status or position.

The actor’s front, as labeled by Goffman, is the part of the individual’s performance which functions to define the situation for the audience. It is the image or impression he or she gives off to the audience. A social front can also be thought of like a script. Certain social scripts tend to become institutionalized in terms of the stereotyped expectations it contains. Certain situations or scenarios have social scripts that suggest how the actor should behave or interact in that situation. If the individual takes on a task or role that is new to him, he or she may find that there are already several well-established fronts among which he must choose. According to Goffman, when a task is given a new front or script, we rarely find that the script itself is completely new. Individuals commonly use pre-established scripts to follow for new situations, even if it is not completely appropriate or desired for that situation.

Front Stage, Back Stage, and Off Stage

In stage drama, as in everyday interactions, according to Goffman, there are three regions, each with different effects on an individual’s performance: front stage, backstage, and off-stage. The front stage is where the actor formally performs and adheres to conventions that have particular meaning for the audience. The actor knows he or she is being watched and acts accordingly.

When in the backstage region, the actor may behave differently than when in front of the audience on the front stage. This is where the individual truly gets to be herself and get rid of the roles that she plays when she is in front of other people.

Finally, the off-stage region is where individual actors meet the audience members independently of the team performance on the front stage. Specific performances may be given when the audience is segmented as such.

• Goffman's Front Stage and Back Stage Behavior
• Examples of Chemical Reactions in Everyday Life
• The Meaning and Purpose of the Dramaturgical Perspective
• A Biography of Erving Goffman
• How Our Aligning Behavior Shapes Everyday Life
• "Cheating Out," "Breaking Curtain," and More Curious Theatre Jargon
• Ancient Greek Tragedy
• 15 Major Sociological Studies and Publications
• Stage Directions for Actors: The Basics
• Choose the Right Setting for Your Play
• Aristotle's Tragedy Terminology
• Stigma: Notes on the Management of Spoiled Identity
• What Is a Total Institution?
• Famous Sociologists
• What Is Role Conflict in Sociology?
• Assessing a Situation, in Terms of Sociology

Identity, Performance and Technology

Practices of Empowerment, Embodiment and Technicity

• © 2012
• Susan Broadhurst ,
• Josephine Machon

You can also search for this editor in PubMed   Google Scholar

Part of the book series: Palgrave Studies in Performance and Technology (PSPT)

5774 Accesses

22 Citations

3 Altmetric

This is a preview of subscription content, log in via an institution to check access.

Access this book

• Available as EPUB and PDF
• Read on any device
• Instant download
• Own it forever
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info
• Durable hardcover edition

Tax calculation will be finalised at checkout

Other ways to access

Licence this eBook for your library

Institutional subscriptions

Table of contents (15 chapters)

Front matter, introduction.

• Susan Broadhurst, Josephine Machon

Identities — New Epistemologies and Ontologies

Improvising artists, embodied technology and emergent techniques.

• Andrew Bucksbarg, Selene Carter

Shifting Listening Identities – Towards a Fluidity of Form in Digital Music

• Franziska Schroeder

Art as Eudaimonia: Embodied Identities and the Return Beat

Graeae: an aesthetic of access — (de)cluttering the clutter.

• Jenny Sealey, Carissa Hope Lynch

(Ex)Posing Identity — Embodied Art Practices

Woven bodies, woven cultures.

• Ghislaine Boddington
• Susan Broadhurst

Experiential Identities in the Work of Marisa Carnesky

Lynn hershman and the creation of multiple robertas.

• Roberta Mock

Empowerment/Disempowerment in Digital Performance

(be)longing : a case study of recording and representation.

• Leslie Hill

Pluralistic Presence: Practising Embodiment with My Avatar

• Sita Popat, Kelly Preece

The Silent Screen/Scream: A Sensual Exploration of the Interior/Exterior Screens of the (Dis)Closing Subject

• Paul Woodward

Blurring the Boundaries: The Delimited Self

Below the surface.

• Helen Paris

Stelarc’s Mystical Body

• Charlie Gere

Borderless Bodies — The Cellulardata Body

• performance

About this book

'...this is an important addition to the material on performance and technology and its impact on embodied performance practices. In its ambition and scope this volume will be of interest to those concerned with somatic practice and the range of emergent thinking and multiple impacts of digital technologies deployed in contemporary performance.' - Scott Palmer, New Theatre Quarterly

'...a solid collection of interesting and insightful case studies of performance events and practices incorporating media technologies.' - David Z. Saltz, Theatre Journal

'This book has something to offer those interested in the confluence of performance and technology from practitioners to scholars of all levels, as well as those outside the academy with an active interest in this type of performance work.' - Sophie Lally, Contemporary Theatre Review

About the authors

Bibliographic information.

Book Title : Identity, Performance and Technology

Book Subtitle : Practices of Empowerment, Embodiment and Technicity

Editors : Susan Broadhurst, Josephine Machon

Series Title : Palgrave Studies in Performance and Technology

DOI : https://doi.org/10.1057/9781137284440

Publisher : Palgrave Macmillan London

eBook Packages : Palgrave Theatre & Performance Collection , Literature, Cultural and Media Studies (R0)

Copyright Information : Palgrave Macmillan, a division of Macmillan Publishers Limited 2012

Hardcover ISBN : 978-0-230-29888-0 Published: 23 October 2012

Softcover ISBN : 978-1-349-33510-7 Published: 01 January 2012

eBook ISBN : 978-1-137-28444-0 Published: 23 October 2012

Series ISSN : 2947-5848

Series E-ISSN : 2947-5856

Edition Number : 1

Number of Pages : XXI, 234

Topics : Performing Arts , Theatre History , Theatre and Performance Studies

• Publish with us

Policies and ethics

• Find a journal
• Track your research

Literary Theory and Criticism

Home › Judith Butler’s Concept of Performativity

Judith Butler’s Concept of Performativity

By NASRULLAH MAMBROL on October 10, 2016 • ( 7 )

Claiming that “Identity is performatively constituted”, Judith Butler in her path breaking Gender Trouble (1990) formulated a postmodernist notion of gender, in line with the deconstructive ethos and contradictory to the traditional notion’, that genders are fixed categories. Butler defined gender as a social role performed/enacted by the individuals, and validated and accepted by society. According to Butler, the meaning of gender depends on the cultural framework within which it is performed, and hence it defies fixities and universalities, because gender is a continuous performance, acquiring new meaning with each repeated performances or citations depending on the context in which it occurs. Refusing fixities, Butler sees gender as provisional, shifting, contingent and performed. This view also rejects essentialisms and stable identities and meanings, while also eschewing notions of authenticity, authority, universality and objectivity.

Judith Butler and Performativity for Beginners (mostly in her own words)

• A central concept of the theory is that your gender is constructed through your own repetitive performance of gender . This is related to the idea that discourse creates subject positions for your self to occupy—linguistic structures construct the self. The structure or discourse of gender for Butler, however, is bodily and nonverbal. Butler’s theory does not accept stable and coherent gender identity. Gender is “a stylized repetition of acts . . . which are internally discontinuous . . .[so that] the appearance of substance is precisely that, a constructed identity, a performative accomplishment which the mundane social audience, including the actors themselves, come to believe and to perform in the mode of belief” ( Gender Trouble ). To say that gender is performative is to argue that gender is “real only to the extent that it is performed” ( Gender Trouble ).
• There is no self preceding or outside a gendered self. Butler writes, “ . . . if gender is constructed, it is not necessarily constructed by an ‘I’ or a ‘we’ who stands before that construction in any spatial or temporal sense of ‘before.’ Indeed, it is unclear that there can be an ‘I’ or a “we” who had not been submitted, subjected to gender, where gendering is, among other things, the differentiating relations by which speaking subjects come into being . . . the ‘I’ neither precedes nor follows the process of this gendering, but emerges only within the matrix of gender relations themselves” ( Bodies that Matter ).

• Performativity of gender is a stylized repetition of acts, an imitation or miming of the dominant conventions of gender. Butler argues that “the act that one does, the act that one performs is, in a sense, an act that’s been going on before one arrived on the scene” ( Gender Trouble ). “Gender is an impersonation . . . becoming gendered involves impersonating an ideal that nobody actually inhabits” (interview with Liz Kotz in Artforum ).
• Biological sex is also a social construction—gender subsumes sex. “According to this view, then, the social construction of the natural presupposes the cancellation of the natural by the social. Insofar as it relies on this construal, the sex/gender distinction founders . . . if gender is the social significance that sex assumes within a given culture . . . then what, if anything, is left of ‘sex’ once it has assumed its social character as gender? . . . If gender consists of the social meanings that sex assumes, then sex does not accrue social meanings as additive properties, but rather is replaced by the social meanings it takes on; sex is relinquished in the course of that assumption, and gender emerges, not as a term in a continued relationship of opposition to sex, but as the term which absorbs and displaces “sex” ( Bodies that Matter ). Butler also writes “I think for a woman to identify as a woman is a culturally enforced effect. I don’t think that it’s a given that on the basis of a given anatomy, an identification will follow. I think that ‘coherent identification’ has to be cultivated, policed, and enforced; and that the violation of that has to be punished, usually through shame” (interview with Liz Kotz in Artforum).
• What is at stake in gender roles is the ideology of heterosexuality. “To claim that all gender is like drag, or is drag, is to suggest that ‘imitation’ is at the heart of the heterosexual project and its gender binarism, that drag is not a secondary imitation that presupposes a prior and original gender, but that hegemonic heterosexuality is itself a constant and repeated effort to imitate its own idealizations. That it must repeat this imitation, that it sets up pathologizing practices and normalizing sciences in order to produce and consecrate its own claim on originality and propriety, suggests that heterosexual performativity is beset by an anxiety that it can never fully overcome….that its effort to become its own idealizations can never be finally or fully achieved, and that it is constantly haunted by that domain of sexual possibility that must be excluded for heterosexualized gender to produce itself” ( Bodies that Matter ).
• Performativity of Gender (drag) can be subversive. “Drag is subversive to the extent that it reflects on the imitative structure by which hegemonic gender is itself produced and disputes heterosexuality’s claim on naturalness and originality” ( Bodies that Matter ).
• But subversion through performance isn’t automatic or easy. Indeed, Butler complains that people have misread her book Gender Trouble. “The bad reading goes something like this: I can get up in the morning, look in my closet, and decide which gender I want to be today. I can take out a piece of clothing and change my gender, stylize it, and then that evening I can change it again and be something radically other, so that what you get is something like the comodification of gender, and the understanding of taking on a gender as a kind of consumerism. . . . [treating] gender deliberately, as if it’s an object out there, when my whole point was that the very formation of subjects, the very formation of persons, presupposes gender in a certain way—that gender is not to be chosen and that ‘performativity’ is not radical choice and its not voluntarism . . . Performativity has to do with repetition, very often the repetition of oppressive and painful gender norms . . . This is not freedom, but a question of how to work the trap that one is inevitably in” (interview with Liz Kotz in Artforum). Butler also writes that “it seems to me that there is no easy way to know whether something is subversive. Subversiveness is not something that can be gauged or calculated . . . I do think that for a copy to be subversive of heterosexual hegemony it has to both mime and displace its conventions” (interview with Liz Kotz in Artforum ).

Share this:

Categories: Uncategorized

Tags: Bodies that Matter , Feminism , Gender Performativity , Gender Trouble , Judith Butler , Linguistics , Literary Criticism , Literary Theory

Related Articles

• Gender and Transgender Criticism – Literary Theory and Criticism Notes
• Spatial Criticism: Critical Geography, Space, Place and Textuality – Literary Theory and Criticism Notes
• Lesbian Continuum: A Brief Note – Literary Theory and Criticism Notes
• Heterosexism – Literary Theory and Criticism Notes
• Queer Theory | Literary Theory and Criticism
• Spatial Criticism: Critical Geography, Space, Place and Textuality | Literary Theory and Criticism
• Reading Notes: Judith Butler on Gender Performativity and Social Construction of the Self | Psychedelic Librarian

Leave a Reply Cancel reply

You must be logged in to post a comment.

• Open access
• Published: 06 October 2020

Young people’s tech identity performances: why materiality matters

• Spela Godec   ORCID: orcid.org/0000-0002-9729-8549 1 ,
• Uma Patel 1 ,
• Louise Archer 1 &
• Emily Dawson 2  

International Journal of STEM Education volume  7 , Article number:  51 ( 2020 ) Cite this article

5437 Accesses

5 Citations

6 Altmetric

Metrics details

Identity provides a useful conceptual lens for understanding educational inequalities in science, technology, engineering and mathematics (STEM). In this paper, we examine how paying attention to physical and digital ‘materiality’ enriches our understanding of identity work, by going beyond the spoken, written and embodied dimensions of identity performances that currently dominate the area of STEM identity scholarship. We draw on a multimodal ethnographic study with 36 young people aged 11–14 carried out over the course of one year at four UK-based informal STEM learning settings. Data collection included a series of interviews, observations and youth-created portfolios focused on STEM experiences. Illustrative case studies of two young men who took part in a community-based digital arts centre are discussed in detail through the theoretical lenses of Judith Butler’s identity performativity and Karen Barad’s intra-action .

We argue that physical and digital materiality mattered for the performances of ‘tech identity’ in that (i) the focus on the material changed our understanding of tech identity performances; (ii) digital spaces supported identity performances alongside, with and beyond physical bodies, and drew attention to new forms of identity recognition; (iii) identity performances across spaces were unpredictable and contained by the limits of material possibilities; and (iv) particular identity performances associated with technology were aligned with dominant enactments of masculinity and might thus be less accessible to some young people.

We conclude the paper by suggesting that accounting for materiality in STEM identity research not only guides researchers in going beyond what participants say and are observed doing (and thus engendering richer insights), but also offers more equitable ways of enacting research. Further, we argue that more needs to be done to support the translation of identity resources across spaces, such as between experiences within informal and online spaces, on the one hand, and formal education, on the other.

Introduction

Camera, phone, school, avatar, animation, animation speed, self-portrait, pride, frustration, pleasure, friendship, humour, traffic, weight, bulk, silence, talk, hesitations, absence, attendance, silver casing, black box, 500 Gigs SSD, terabytes, YouTube and selfies … perhaps all this stuff does not belong in the data, but I think we are missing something if it is left out. This is real and the stuff of young people’s lived experience of technology and STEM futures. (Data memo notes)

Identity has been of growing interest in STEM education research (Simpson & Bouhafa, 2020 ). Whether young people identify with science/STEM (e.g. see science as being ‘for me’, see themselves and are seen by others as a ‘science person’) has been found to relate to their aspirations, engagement and participation in science (e.g. Archer et al., 2013 ). Understanding how young people navigate their identity work is particularly important for addressing the persistent inequalities in science/STEM participation, which are shaped by gender, social class, ethnicity, dis/ability and other social axes. There is an abundance of evidence showing that it is easier for some young people to see themselves in science/STEM than it is for others, which has implications for their educational and professional trajectories, as well as for engaging with the subject(s) in everyday life (Carlone & Johnson, 2007 ; Godec, 2018 ; Gonsalves, 2014 ; Gonsalves, Silfver, Danielsson, & Berge, 2019 ; Kim, Sinatra, & Seyranian, 2018 ; Mendick, 2005 ; Rainey, Dancy, Mickelson, Stearns, & Moller, 2018 ; Vincent-Ruz & Schunn, 2018 ; Wong, 2016 ).

There has been an emphasis in STEM identity scholarship on discourse and, specifically, language. While there is a growing number of STEM identity studies (either relating to a specific subject like physics or considering the broader ‘STEM identity’), research findings have to date largely been based on the participants’ narrated accounts. Verbal and written articulations of identity have become a methodological norm, as evident from the dominance of studies utilising interviews, discussion groups and surveys (Varelas, 2012 ). Such privileging of language has been critiqued in the wider literature for possibly silencing the complexity of lived experiences, for it ‘fail[s] to fully account for the complex materiality of life’ (de Freitas & Curinga, 2015 , p. 249). Barad ( 2007 ), whose work has guided the recent ‘new materialist’ movement in social research and some of whose concepts we utilise in this paper, has advocated for close reading of wide-ranging data, including the entanglement of language and other ‘things’. With seemingly polemical remarks like ‘[l]anguage has been granted too much power’ in social research (Barad, 2007 , p. 132), Barad is problematizing the state of affairs where every ‘thing’, even materiality, is tied to language representations.

The focus on language arguably normalises and valorises a particular (classed, gendered and racialized) ability to ‘tell the self’ (Skeggs, 2004 ), while also providing limited opportunities for some younger participants and non-native speakers to express themselves. Research participants who might not possess specific resources of privilege to perform identity verbally might therefore risk being invisible, which raises concerns about the equity of research endeavours. To this point, we would add that not only does language have a somewhat privileged place in social research, but also some language/s and language forms tend to be more powerful and more privileged than others (e.g. Foucault, 1988 ; Hall & Gay, 1996 ).

We acknowledge that STEM identity research has not exclusively relied on language and recognise the important work that has employed broader methodological approaches. A number of researchers have, for example, focused on observations of embodied performances of identity (e.g. Archer et al., 2016 ; Danielsson, 2011 ; Dawson et al., 2020 ; Mendick, 2005 ). These studies have looked at the ways that participants perform their identities through physical appearance, behaviours and interactions. Attending to embodied enactments of identity has contributed rich insights about how some ways of being and doing were congruent with STEM, while others posed challenges for identity negotiations. The studies referred to here have provided valuable insights about how, for instance, performances of femininity (e.g. physical appearance and ‘typical’ feminine behaviours) tend to sit uneasily with the dominantly valued performances of STEM, with technology and engineering being particularly problematic (Danielsson, 2011 ).

While a growing body of STEM identity scholarship has considered embodied performances (i.e. focusing on the materiality of human bodies), we suggest that there is further opportunity to draw on new materialist insights and extend the focus beyond the body. In their edited book Material Practice and Materiality: Too Long Ignored in Science Education , Milne and Scantlebury ( 2019 ) have pointed out that the lack of attention to materiality is prevalent across science education research; the gap is thus not specific to science/STEM identity scholarship. While studies attending to non-human materiality have been scarce, we found a number of examples that showcase the potential of this direction (e.g. Calabrese Barton et al., 2013 ; Gonsalves & Danielsson, 2017 ; Talafian, Moy, Woodard, & Foster, 2019 ). Calabrese Barton et al. ( 2013 ), for instance, have utilised the concept of ‘identity artifacts’ (drawing on the work of Leander, 2002 ) to demonstrate how a material focus can enrich our understanding of science identity work. Calabrese Barton and her colleagues examined how young people’s various physical artefacts, such as an award-winning rocket or a poem, mediated their identity work in sometimes productive and other times constraining ways as the young people moved between different educational settings.

Theoretical framework

In thinking about the role of materiality in young people’s identity performances, we draw on the conceptual work of Judith Butler ( 1990 , 1993 ) and Karen Barad ( 2007 ). Despite some tension between the two scholars (see below), we suggest that bringing the two theoretical frameworks together is productive for including materiality. It is relevant to mention here that we did not start this piece of research by prior, foundational commitment to this specific conceptual orientation. Instead, we turned to Butler’s work on identity performance and Barad’s new materialism as a way to help us explain the repeated presence (and seeming importance) of materiality within young people’s STEM identifications that we observed in our data, as exemplified by the data memo notes that opened this paper. In this sense, we selected our theoretical framework in response to what we were noticing in the data. Yet, this does not mean that the process was solely data-driven and unidirectional. Rather, our process was guided by assemblages of ‘things’, as the research team mobilised theoretical resources, ethnographic accounts, case studies, photographs and field notes to develop articulations of the relations between human, material and conceptual phenomena. Next, we introduce our theoretical resources in more depth, highlighting in particular how these have dealt with materiality.

Butler’s identity performativity

Butler ( 1990 , 1993 ) has argued that identity is ‘performative’, a ‘doing’ rather than a ‘being’. She has pointed out that the performative ‘doing’ of identity depends on the particular discourses that are prevalent within a given space and the physical (material) bodies that enact these performances. For example, performative acts and their recognition are likely to differ in the context of mature women and that of queer boys. In her work, Butler ( 1993 ) has explicitly acknowledged the importance of materiality, writing, for example, that ‘[i]t must be possible to concede and affirm an array of “materialities” that pertain to the body’ (p. 66).

Despite considering materiality in her work, Butler has predominantly focused on embodiment (i.e. the materiality of human bodies) and paid less attention to the role of other, ‘non-human’ objects. This limitation of Butler’s performativity work has been noted by Barad:

Judith Butler’s performative account of mattering thinks the matter of materiality and signification together in their indissolubility; however, Butler’s concern is limited to the production of human bodies. (Barad, 2007 , p. 145)

While some critique from new materialism scholars has perhaps been unjustified, such as that discursive work is essentially a-material (Barad, 2007 ), we agree that non-human materiality could be more extensively considered in discursive research (and in the specific area of science/STEM identity research, as we argued above).

Barad’s concept of intra-action

While we regard Butler’s identity performativity lens compelling to think with, we agree that her work alone might ‘fail to provide an adequate account of the relationship between discursive practices and material phenomena’ (Barad, 2007 , p. 146), particularly for thinking beyond embodied practices. Thus, we found Barad’s work on new materialism helpful for extending our thinking about the role of the matter in STEM identity work. We consider Barad’s concept of intra-action and her thinking about agency to provide useful insights in our theoretical framework. Barad has introduced the term intra-action as an alternative to ‘interaction’, which she has regarded as relating to pre-established bodies and ‘things’ that are in action with each other. She has used intra-action to instead emphasise that meaning and matter are essentially entangled.

For Barad, all phenomena are co-constructed enactments between human actors and non-human things (see also Latour, 2005 ; Law & Mol, 2002 ). In relation to this, Barad ( 2003 ) has advocated that agency is not an individual property, but rather ‘agency is a matter of intra-acting; it is an enactment, not something that someone or something has’ (Barad, 2007 , p. 235). In other words, she has proposed that agency is located in the actions between material ‘things’, calling for particular attention to be paid to the said intra-actions between human and non-human entities. From this perspective, Barad has critiqued poststructuralist perspective on agency, arguing, for instance, that ‘for both Butler and Foucault, agency belongs only to the human domain’ (2007, p. 145). Barad’s notion of ‘distributed agency’ has been productively taken up in different ways in existing scholarship across various research domains, such as to explore performative pedagogies in getting university students to narrate the distributed agency of buildings (Mcphie, 2018 ) and to examine the power dynamics of the classroom (Murris, 2016 ).

Coming from a sociological perspective and taking a strong equitable stance in our own research work, we see Barad’s intra-actions as encompassing multiple entanglements between actors, objects and spaces, which are intertwined with power relations and related to the issues around im/possibilities. We argue that Barad’s work productively complements a Butlerian lens because, as Palmer ( 2011 , p. 5) has noted in her work on mathematical subjectivities, Barad ‘sharpens the theoretical tool of [Butler’s] performativity’ through taking an interest in the intra-activity of matter in this process (see also Ringrose & Rawlings, 2015 ).

Focus of this study

In the new materialism literature, the emphasis has largely been on the philosophical ‘mattering of matter’ and on theorising the relations between things human and non-human (Barad, 2007 ). We advocate that in a more practical sense, improving STEM participation and engagement, and indeed what counts as STEM, is tied to the progress in understanding the complexities of how particular STEM identities become and endure, or not. Considering the current science/STEM identity scholarship, we suggest that more attention might be usefully given to non-human materiality. With this paper, we thus seek to extend the current research by focusing on physical and digital materiality and its role in young people’s STEM identity negotiations. Specifically, we address the following research questions:

How do young people engage with physical and digital (non-human) materiality in ways that support, or not, their identity performances in relation to technology?

How does paying attention to materiality enrich our understanding of STEM identity work?

While within the wider project, we focus on STEM and locate our work within the wider science and STEM education research, the two case studies we discuss in depth in this paper predominantly relate to technology.

Data collection: multimodal portfolios and ethnographic field notes

In this paper, we draw on multimodal ethnographic data: field notes from ‘portfolio sessions’ (see below), multimodal portfolios, semi-structured and unstructured interviews, discussion groups and photographs of artefacts and spaces. Data were collected as part of Youth Equity + STEM (YESTEM) project that, broadly, aims to understand young people’s engagement with informal STEM learning and explore equitable practices within this space. We recruited 36 young people who voluntarily participated in programmes across four informal STEM learning settings; we worked with between six and 13 young people per setting. The selection criteria for recruitment into the study included age, gender, ethnicity, socio-economic background and the length of the young people’s involvement with the particular setting. Where possible, we were interested in recruiting young people with more substantive/longer-term involvement. Young people in this study were aged between 11 and 14, 22 were girls and 14 were boys, 16 self-identified as White British, seven as Black, two as Asian and the remaining 11 as mixed or other ethnicities.

The participants were initially approached to take part in the research by the educators working within the four organisations, and later met with a member of the research team who explained the nature of participation and ensured that the ethical guidelines were being met. The research project was presented to the young people as an opportunity to share views about their STEM engagement in and out of school, with the aim to help make STEM a more welcoming place for more young people like them. In addition, the research team emphasised to the young people that they were experts in their own lives and that understanding their experiences and perspectives would be highly valued and useful for informing future provision of STEM education. All the young people whom we invited to the study agreed to take part.

Young people were involved in the study as co-researchers, meaning that they participated in collecting and co-producing data (see examples of some artefacts they produced in the “Results and discussion” section). We organised between six and ten small-group sessions at each setting, where young people worked on constructing portfolios of their STEM-related experiences and discussed various aspects of their STEM engagement. These sessions were co-led by educators and researchers. The portfolio sessions involved, for instance, creating a self-portrait and documenting everyday STEM engagement (e.g. young people were asked to research and document their STEM experiences in their home, school and other settings).

During the portfolio sessions, we kept detailed field notes, took photographs and recorded conversations. We paid particular attention to what the young people were doing and how, when and where different materials were being mobilised in communicating with us and others. We also attended to how different physical and digital spaces supported or hindered STEM engagement and identity work, through investigating the social context and the specific material aspects, such as the availability of technological resources and access to additional communication channels (e.g. the online forums and social media). As researchers, we were entangled with the research process; we were not simply uninvolved observers, but took part in running the portfolio sessions where we interacted with the young people, such as through encouraging them and complimenting them on their artefacts.

During the portfolio sessions, we regularly engaged with young people in unstructured conversations and, after the sessions, conducted individual interviews following a semi-structured interview schedule. Questions most relevant to this paper included, for example: Would you say you are a science/tech person? Why, why not? What makes someone a science/tech person? Can you think of anyone who sees you as a science/tech person? Finally, we met young people for a follow-up discussion six months after the end of portfolio sessions. All interviews and discussion groups were audio recorded and transcribed. We combined individual and group data collection sessions in order to facilitate both peer discussions and provide space for young people to converse with us directly. The latter enabled some quieter and less confident participants to share their experiences in a different setting. The data that we ultimately co-constructed with participants varied across the cohort; in some cases, data for an individual participant consisted mostly of audio transcripts, with a few drawings or photographs to accompany them. In other cases, verbal and written accounts were minimal and the data were primarily in the form of digital and physical artefacts.

Our data collection approach was purposefully open and flexible, which was motivated by a growing body of literature advocating for the need to enable multiple modes of communication, particularly when working with younger participants and/or people who might struggle to articulate and present their thinking through language (Thomson, 2008 ). All young people we worked with in this study received a tablet or an action camera, which some used to take photos and videos to include in their portfolios (e.g. photos of science and technology in their lives). In some cases, additional technology and skills support was available to facilitate other techniques through which young people could perform themselves, such as digital drawing tools and stop motion animation.

The creation and interaction of various modes of communication were often synchronous. This meant that, for example, a young person might have been creating a product, such as an animation, while talking to us about a different aspect of their STEM involvement. Although material artefacts like drawings, photographs and other visual forms served as valuable prompts for conversations, we also thought of these as giving insight into richer, different meanings in and of themselves—and were especially valuable for young people who were less keen or comfortable expressing themselves through language. During the portfolio sessions, we observed the connectivity of various modes of communication and the ways young people interacted with the material/medium. The ‘mode’ of communication relates to the senses, so can be visual, linguistic, aural, spatial, gestural and so forth. At the same time, the term mode also has a lineage in computer design where the same actions produce different effects in different modes. The term ‘medium’ is linked to media (e.g. print, animation) and has a lineage in media studies heralded by McLuhan and Fiore’s ( 1967 ) famous book The Medium is the Massage . In our data, dis-aggregating mode and medium distracts from the focus on what is being mobilised and to what effect. To avoid linguistic gymnastics, we found ourselves conflating the terms mode and medium and in doing so, have come to recognise this as part of the process of analysis and writing intra-action .

Data analysis

We started data analysis by reading the data within the research team. We looked at how young people ‘narrated the self’ (verbally performed their identity), i.e. how they talked about their identity in relation to STEM. This generated accounts of how identity was discursively performed through language (Butler, 1990 , 1993 ) and involved how young people spoke about science/tech/STEM as being for them or not, the recognition they received from others, and what it meant to be a science/tech/STEM person. Our analytical process was iterative; the assemblage (Barad, 2007 ) included data, theory and concepts that we found to be useful for shedding the light on the role of materiality in STEM identity research.

Next, we moved to mapping out how the material conditions shaped identity work by looking at the ‘material moments’ (Taylor, 2013 ), i.e. moments that we interpreted as signifying intra-action between our participants and the digital and physical materiality. Following Barad, we engaged in close reading of relational effects, such as how things were connected and emerged as important. This meant paying attention to intra-action dynamics of where, when, who and what that was inclusive of language, academic text and the researcher, while being mindful of the power and privilege of particular types of data. The analysis involved a close reading of the rich ethnographic accounts generated by the researchers and data in photographs and portfolios from the young people. We attended to what digital and physical matter the young people shared with us and/or mobilised during the sessions, and how these examples were then actively referenced or used.

Inspired by other researchers, we also explored the affective relations within the data and selected particular segments that ‘glowed’ at us to examine further (MacLure, 2013 ); this process informed our case study selection. For the segments that ‘glowed’ (i.e. appeared particularly interesting or significant for answering our research questions), we mapped out the collection of things human and non-human that were making specific enactments possible. This inferencing went beyond the dialogic analysis; it was integrated into the process as a series of accounts that were in effect descriptions in the spirit of ‘look what is going on here’.

The analysis involved a critical reading of what role different physical, digital and social spaces had in supporting or hindering young people’s identity performances. Particular attention was paid to the material-discursive intra-actions where the material came into play in relation to recognition, competence and self-identification. This involved, for instance, how the young people used material objects to perform their identity in relation to technology. We would suggest that the physical presence of material objects, for example, a home-built computer (see Fig.  1 ), was vital for making the specific intra-action possible. It was the physical presence of this object that facilitated Black-and-White’s prolonged explanation of the computer parts and capabilities, which we interpreted as enactments of tech identity performance. Such rich enactment, we suggest, would be impossible without the specific material object. Thus, in our analysis, we paid close attention to what was being gathered at any specific moment in time, e.g. the researcher, the educator, the portfolio and the technology-rich educational space.

Black-and-White’s home-built computer

Some of the questions that guided our analytic process included the following: What materials seemed to be essential to a particular STEM identity performance? What made specific enactments possible or impossible? Where and how were specific identity performances viable? Where did they appear to stumble, falter or be dismissed? What were the relational shifts over time and space? Following the analysis, we reflected our interpretation back to the young people and invited comment and critique. This was done through creating a printed booklet for each young person that included our analysis about their STEM experiences, identity and aspirations, along with the various data we collected (e.g. photos from the portfolio sessions, quotes from the interviews). As part of the interviews, young people were encouraged to offer their own interpretations and challenge us on our ongoing analysis presented in the booklet.

Selecting the case studies

Attending to materiality seemed to be particularly valuable in the case of young people who felt less comfortable or less keen on expressing themselves verbally or in writing. The accounts of these young people may have been missed had we taken a more traditional qualitative research approach that relied on language as the primary mode of expression. For instance, when we asked the young people about their engagement and identification with science and technology (e.g. would you say you were a science/tech person? Why/not?), some found their answers difficult to articulate (see Ginger’s example below).

For this paper, we selected two young men (Ginger and Black-and-White, both 11 years old; pseudonyms were chosen by the participants) from our larger study to present in greater depth. Having reviewed all the cases, these two appeared to be the clearest and strongest examples of how non-human materiality mattered for their tech identity performances, as well as being particularly interesting in terms of illustrating what might have been invisible or anomalous if we focused in our research only on language. Both young men found expressing themselves verbally and/or in writing challenging in different ways (e.g. Black-and-White asked us to avoid writing and Ginger’s responses to our questions were mostly monosyllabic, as he admitted that ‘it’s quite hard for me to answer these questions’; he preferred to write ‘when I type it I’m just so much better at it’). At the same time, these two young men produced rich and meaningful data through other forms of self-expression during portfolio sessions. A further reason for selecting these two participants was that the setting within which this study was located (a community-based digital arts centre) appeared to have played a particularly important role in their identity negotiations. We also found different and complex manifestations of the two young men’s identity work in other spaces, such as at home and at school, which made them particularly interesting to focus on.

Results and discussion

Reading material-discursive entanglements as identity performances.

We start by presenting two vignettes of ‘material moments’ (Taylor, 2013 ) that stood out to us as particularly interesting in terms of materiality being intimately related with the young men’s identity performances and recognition. During one of the portfolio sessions, Black-and-White brought along a home-built computer (see Fig.  1 ), telling us that the computer showed ‘the tech I do in my life’ and explained with technical terminology what the significant features were. The field notes below summarise the occasion.

We arrive at the centre early and wait in the lobby for the participants to arrive. The first to come are brothers Spuggs and Black-and-White. Black-and-White is carrying a large black box, which I notice is computer hardware. As we inquire what this is (and why he brought this with him – it looks heavy!), he responds that it is to show us what he built at home. During the portfolio session, Black-and-White is not keen to get involved with making an animation, which is an optional activity today. Instead, he chats to us about the computer. He insists that we make sure to record all specifications correctly. He dictates us to write: ‘there is a lot of memory in there, two terabytes, 500 Gig SSD, that’s a lot of memory’, pointing to the various components that are the memory as he talks. He visibly labours across the room with the computer that he’s built, refuses help, and puts it under the camera (which others are using to make animation) so he can take a picture of it for the project portfolio. (Constructed from field notes, November 2017)

We suggest that Black-and-White’s display of his home-built computer can be interpreted as a performance of his skills and knowledge related to technology. He had briefly spoken to us previously about his involvement with technology, such as that he was the fastest in the weekly sessions to assemble robots. This point that was also recognised by others within the programme at the time, e.g. loud comments made by the educator during the weekly sessions ‘Black-and-White in the corner is our tiddlybot expert’ (‘tiddlybot’ is a Raspberry Pi robot kit). Notably, it was evident that the materiality of the computer further supported and was a key part of Black-and-White’s performance of tech identity. The intra-action between the non-human (home-built computer) and human (Black-and-White) was enacted with pride, respect and recognition—we speculate that in the absence of the computer, the enactment may have taken a different form, perhaps one that offered a more limited insight to Black-and-White’s performance of tech identity.

We suggest that the hand on the computer case in Fig.  1 might give some indication of the physical effort involved in Black-and-White’s ‘owning’ the casement of the computer and the delicate attention to detail within the limited space of the casing. He engaged with a female researcher in a masculine demonstration of power and pride (akin to the performance of ‘muscular intellect’, involving a confident display of expertise, see Archer et al., 2017 ; Mac An Ghaill, 1994 ). Black-and-White’s expertise appeared to be vital in how he negotiated competence and identity work; ‘I know I’m good at building things because I am. I build PCs’. His involvement with technology also featured in his future aspirations; he hoped to ‘build PCs’ when he got older, intimating the importance of physical materiality for this continuous, long-term engagement with technology.

Ginger’s modality of choice, on the other hand, was digital. During one of the portfolio sessions, he created an elaborate stop motion animation (see Fig.  2 ), assembling his drawings and writings together in a short animation about ‘my life in STEM’.

In the computer room, four ‘animation stations’ are set up – one for each participant, should the boys choose to create an animation as a way of communicating their STEM engagement (not everyone decides to). The practitioner gives a brief introduction, i.e. that they could focus on their journey through STEM, but encourages creativity. Ginger sets up his station (an iPad facing down, recording the images) and starts drawing and writing. He does not require any assistance. He seems to be in the flow and we don’t interrupt. This is his first time using this specific technology setup and software, and we comment a couple of times how quickly he is picking up the skills. Ginger makes an impressive 30-page animation, drawing and writing non-stop for 90 minutes. The animation shows a story of how he first found out about the community centre, got into coding, what he has been doing since, and how these activities related to his future aspirations and life outside. (Constructed from field notes, November 2017)

Ginger’s stop motion animation

Figure 2 illustrates the complexity of the intra-action : the elements of light, camera focused on a drawing pad, action on the surface that is being digitised on a screen. Later, these images and words were manipulated into a timeline to communicate Ginger’s STEM journey. At the time the image was captured, Ginger was not particularly interested in talking to the researcher but nor was he indifferent to showing us what he knew, thought about and was able to do, using his preferred mode of communication. The functionality of the software and input devices appeared to be as important to Ginger’s identity performance as the content; virtual space was brought into physical space as a way of communicating. The animation provided a medium through which Ginger was comfortable expressing his thoughts and experiences. We suggest that there were additional possibilities that became evident through the conceptual lens of intra-action , which enabled a display of skills through the creation of a digital artefact. The meaningful modality was a digitised product, which supported identity performance beyond words and beyond the traditional form of a pen-and-paper.

For both Ginger and Black-and-White, materiality mattered in a sense of being intimately entangled in the production of their tech identity; we argue that it was not separate and not just a vehicle. Digital and physical materials were ‘vital players’ (Bennett, 2010 ) in how the two young men produced their identities. Both were making choices in terms of how to communicate. These choices included not just content but also the medium, i.e. multimodal options as they choose between available material and linguistic resources to construct and animate (in some cases, literally) their tech identity. As Miller ( 2010 , p. 60) has argued, ‘objects make us, as part of the very same process by which we make them’. The mode of presentation in the cases was deeply entangled with technology. Through creating and interacting with the technology, Ginger and Black-and-White were constructing and performing their STEM identity in ways that language alone may not have supported. Their agency was emergent from intra-action with matter, which was central to being recognised for their expertise and contributing to their identity performance (Carlone & Johnson, 2007 ).

Beyond physical bodies and spaces—digital spaces and identity recognition

Spaces mattered for negotiation of tech identity performances. For Ginger, the intra-actions with the digital, online space seemed to play a particularly important role in his identity work (see Leonardi ( 2010 ) for a similar discussion). Ginger would often share with us the digital products he created outside the project sessions (e.g. online games, which led to chats with other gamers). He also had several online channels where he would post his work, which he spoke about with pride (‘I just share it with the whole entire world’; ‘I get people from far away following me!’).

Ginger’s tech involvement appeared to be closely connected to his future aspirations; he told us that he wanted to be a games designer and added ‘I already am, really’, highlighting his experience with online games design. We argue that Ginger’s identity work appeared to be performed largely through, and in turn enhanced by, the digital medium where he regularly gained recognition for his work (‘people think because of how advanced my games are that I’m much older’), indicating the importance of the particular space in shaping identity performance. In this way, we suggest that his identity performance occurred in relation with the online community. As others have pointed out, online involvement includes performative acts in and of themselves, providing powerful tools and a space to perform and develop tech identity (Cover, 2012 ).

Ginger’s enactment of identity in digital spaces can be seen as going beyond the confines of both his physical body and his physical location. The expanded intra-actions of Ginger’s identity performance can also be interpreted as enabling translocational and transformational identity work (Anthias, 2002 , 2008 ). Anthias ( 2008 ) has, for instance, proposed that different spaces (and times) matter for people’s identity work, as people are inevitably positioned differently. Ginger’s engagement with the digital world could, therefore, be interpreted as a form of such translocation (i.e. being positioned differently in different spaces), enabling him to perform expertise and be recognised in ways that supported his present and future identity work. In this way, the material aspect of space (in this instance, digital) played a crucial role in Ginger’s performance and recognition of tech identity.

Limited transferability of tech identity performances across spaces

It is well established that identity performances shift and change across space and time and depend on which enactments are ‘intelligible’ (Butler, 1993 ), supported and recognised within particular contexts (for some examples in science education, see Archer et al., 2017 ; Carlone, Scott, & Lowder, 2014 ). In this study, materiality played a role in supporting, and sometimes hindering, tech identity work as the two young men moved across different physical and digital spaces.

Ginger and Black-and-White both appeared to successfully perform tech identity in some spaces (namely, the informal STEM learning setting, to some extent at home and in Ginger’s case, additionally, in the online space). The physical environment of the community-based digital arts centre, the technology available there and the support and recognition the young men received from their peers and educators played an important role in their tech identity performances. However, these successful performances appeared to be limited to a specific time and space or what could be interpreted as an example of Barad’s ( 2007 ) ‘timespacemattering’. For Ginger, digital performances (e.g. his coding and games design, with associated recognition) seemed to only partly translate into the spaces of home and school. For instance, he would regularly speak about not being able to engage with technology at school (‘they don’t really care about it at our school’), displaying a strong critique of the absence of opportunities:

They want to encourage people to do tech and be a tech person but then in school, it’s not an option to do ICT [Information and Communication Technology] until you’re year 8, literally halfway through year 8! School is definitely not ready for the internet age! (Ginger, interview)

Ginger also admitted that he rarely talked about technology at home, commenting that his family found his involvement with technology ‘a bit odd’ (or what might in Butler’s terms be labelled ‘unintelligible’) and his parents did not understand much of what he is doing; ‘sometimes my dad helps me come up with ideas for the games, but that’s pretty much it, really’. Black-and-White, similarly, commented that he got little recognition for his technical skills at school, where access to technology was limited (‘laptops at school don’t even work’; ‘computers are rubbish’). The availability of technology was crucial for both Ginger’s and Black-and-White’s engagement and performances of expertise and identity. Its absence, likewise, hindered opportunities for tech identity performativity and recognition.

The young men were able to perform tech identity within the informal STEM learning setting, but less so at school. The disjuncture between the identity enactments that were possible or impossible in particular spaces appeared to have implications for the young men’s tech/STEM engagement and raised a concern about their future trajectories. Ginger, for instance, spoke about starting to doubt his abilities for the future: ‘I’ll do whatever the school thinks I’m capable of doing’; ‘I might be a coder, I don’t really know, depends on what people say I could do’.

The inconsistencies of opportunities for identity enactment, we suggest, were experienced by the young people as troubling. We interpreted Ginger’s statements as possibly complicating his tech identity—Ginger struggled to see his future self in technology. Similarly, we suggest that Black-and-White’s difficulties at school and the lack of opportunities to engage with technology there made it challenging for him to perform tech identity in ways he did within the informal STEM learning setting. The sense of powerlessness, we suggest, raises concerns about the resilience and longevity of tech identity performances when these are weakly supported within the mainstream educational spaces. The findings indicate that physical and digital spaces, time and matter, might enable the longer-term potential for these performances to be sustained (and possibly consolidated into STEM trajectories). The young men’s performances were ‘read’ differently in different spaces, such as by their parents and their teachers, who the young men intimated did not extensively recognise and support their involvements. In the same way as being positively recognised across multiple spaces might amplify tech identity work, dissonance between spaces is likely to cancel out the positive effects from one space as a person moves to another.

Were technology-supported tech identity performances open to everyone?

The young men’s performances of technical knowledge and mastery/expertise, such as demonstrated by Black-and-White’s technical explanations and intra-actions with the computer he brought in to show us, and by Ginger’s frequent references to specificities of his coding work, could be interpreted as exemplifying dominant alignments with masculinity. Similar dominant associations of computing, coding and tech with masculinity have been widely noted in the literature (Francis, 2000 ) and have been regarded as reproducing exclusivity and stereotypical associations of technology and tech identities (Charles & Thébaud, 2018 ). We suggest that these ways of performing tech identity might risk reinforcing popular geeky, a-social notions of who participates in technology and who performs tech identities, which in turn restricts the extent to which tech identity might be open to other youth (Varma, 2007 , 2010 ).

Equity issues around who might be able to perform technology-supported tech identity performances were raised across our wider study cohort. For instance, we offered all young people in our study ( n  = 36) the use of multimodal, technology-enhanced tools to author their identities (e.g. providing tablets and action cameras), but comparatively few did so. This was especially telling for the group of young women who took part in an all-women STEM club that included coding events, given the programme’s explicit focus on technology. These young women appeared to prefer verbal, pen-and-paper and photographic modes to construct their portfolios. They tended to use tablets mostly for selfies, performing identity through social media and digital spaces, which we suggest tended to be social rather than tech identity performances (Dawson et al., 2020 ). While we do not seek to draw broader conclusions from the small numbers of young people we worked with, we suggest that the young people’s tendency to perform identity through different modalities can be interpreted as both reflecting and (re)producing dominant gender relations. That is, it was the masculine enactments that seemed to most successfully mobilise the space, time and material resources to perform recognisable tech identities. In turn, such performances were largely absent within data on young women in our study.

Conclusions

In this paper, we reported on a study of young people’s tech identity performances, drawing on conceptual tools of identity performativity (Butler, 1990 , 1993 ) and intra-action (Barad, 2007 ). In our data collection, we encouraged multiple forms of expression, which was especially valuable for young people who found it challenging to engage in verbal performances of identity through interviews and written forms. Including the material in the research process supported performances of young people’s interest and expertise, which we argue was critical for supporting identity negotiations for a more diverse range of young people. Had we only used more traditional language-based and observational research methods without specifically enabling and including the material entities, some of the young people’s identity performances may have been less visible, or misread as ‘thinner’ or invisible.

This paper seeks to make original empirical and methodological contributions to STEM education research. Empirically, the paper contributes to new understandings of identity performativity through exploring the role of materiality in the production of tech identity performances. We interpreted the material-discursive entanglements as performances of tech identity that can be read and interpreted similarly to verbal or written forms and in ways that expanded our understanding. Matter not only served as a mode of expression, but was also intimately connected to how young men like Ginger and Black-and-White performed identity. Moreover, the case studies demonstrated that both digital and physical materiality were integral to how the two young men produced their identities, highlighting the role of both tangible and intangible non-human matter. While STEM education literature has previously, at least to an extent, considered the role played by physical objects in young people’s identity work (e.g. Calabrese Barton et al., 2013 ), our findings suggest that future research would usefully benefit from extending the focus also to non-tangible, digital materiality, especially given the role that technology and the internet play in young people’s lives. We found that Barad’s intra-action shifted and disrupted the direction of analysis by enriching what was possible to articulate and claim.

The research suggested that identity enactments did not always travel easily across spaces. Previous research has similarly found that identity work is often context-specific; identity performances enacted in one space do not necessarily extend to another. Several scholars have, for instance, previously pointed out the disparities of opportunities for science/STEM identity performance between informal learning spaces and the school science classroom (Calabrese Barton et al., 2013 ; Thompson, 2014 ), or even between two different science classrooms (Carlone et al., 2014 ). We found that STEM identity performances across different (physical and digital) spaces were also influenced by the material opportunities available to young people, which we suggest adds a valuable, novel perspective to existing scholarship. The weakened opportunities for tech identity performances at school, such as a lack of equipment and adequate support, posed challenges to the sustainability of identity work. Experiences in informal spaces and online worlds that enabled and supported Ginger’s identity work—and we would say, were crucial for his engagement with technology—were at risk of being ‘cancelled out’ at school, threatening his longer-term identity work and trajectory.

Methodologically, accounting for non-human materiality offers a way to study tech identity performances that goes beyond the focus on what youth say and what researchers observe. We argue that this approach offers more scope to those who might find verbal articulation difficult. As others have proposed, the reliance on language-based articulations tends to privilege particular people/modalities/data (Milne & Scantlebury, 2019 ; Skeggs, 2004 ). As Mazzei and McCoy ( 2010 , p. 515) remind us, to move the research endeavours forward, we need to open up what ‘counts’ as data in order to give voice to diverse participants and diverse forms of expressions. Moreover, this approach led us to consider agency within the specific enactments, not solely as a property of an individual, thereby emphasising the importance of looking at phenomena through human/non-human intra-actions and paying sufficient attention to materiality.

Based on our findings, we would call for more focus on multimodality in STEM identity research. We argue that accounting for various forms of materiality might offer a less deficit, more equitable way to study a wider range of tech identity performances, enabling enactments that are owned and celebrated by young people who might otherwise risk being missed or side-lined. Broadening the approach to studying STEM identity performances can help highlight equity and inclusion challenges facing young people in maintaining, protecting and developing STEM identities while moving between digital and physical spaces of home, school, informal educational spaces and the online. Importantly, our approach enabled us to gain insights that may have been masked otherwise.

We anticipate that increased attention to materiality and multimodality could also be beneficial to informal STEM learning practice, such as for moving endeavours towards greater openness, sensitivity and responsiveness to what is going on with young people. This focus might entail supporting diverse ways of communicating, engaging with STEM and performing identity, along with recognising how the presence and absence of particular ‘matter’ might advance or curtail engagement opportunities. Further, supporting multimodality might facilitate more equitable practice where young people (and particularly those that may have been marginalised within dominant educational spaces) might be able to leverage ways of performing STEM identity—not only through verbal and written performances, but also through their engagement with physical and digital materiality.

Finally, we would suggest that more needs to be done to support the translation of identity resources across spaces, such as through closer collaboration between science education providers and greater recognition of the varied forms of STEM engagement. The lack of scope for Ginger and Black-and-White to perform valued tech identities across different spaces points to disjunctures between identity performances that demonstrate experiences, skills and knowledge, and limited recognition for such performances in the school context. The findings of this study highlight the importance of alternative provision for engaging with STEM. While it might be difficult to effect change within families and the support available to young people within their home environment, the findings of this study suggest that connectivity between formal and informal educational spheres could be improved to better support young people. Existing literature offers insights into identity development in relation to technology, including within digital spaces (Bennett, Maton, & Kervin, 2008 ; Goode, 2010 ); it would be valuable to consider how such endeavours might best be supported in other educational spaces. To conclude, the study presented in this paper provides valuable evidence that more needs to be done to recognise and value diverse young people’s skills, experiences and modes of identity work within dominant spaces, such as within the formal education system, in order to support them in the ongoing development of their identity work in relation to STEM.

Availability of data and materials

For ethical and data protection reasons, we are unable to share any data from this project.

Abbreviations

Science, technology, engineering and mathematics

Anthias, F. (2002). Where do I belong? Narrating collective identity and translocational positionality. Ethnicities , 2 (4), 491–514.

Google Scholar  

Anthias, F. (2008). Thinking through the lens of translocational positionality: An intersectionality frame for understanding identity and belonging. Translocations: Migration and Social Change , 4 (1), 5–20.

Archer, L., Dawson, E., DeWitt, J., Godec, S., King, H., Mau, A., … Seakins, A. (2017). Killing curiosity? An analysis of celebrated identity performances among teachers and students in nine London secondary science classrooms. Science Education , 101 (5), 741–764.

Archer, L., Dawson, E., Seakins, A., DeWitt, J., Godec, S., & Whitby, C. (2016). “I’m being a man here”: Urban boys’ performances of masculinity and engagement with science during a science museum visit. Journal of the Learning Sciences , 25 (3), 438–485.

Archer, L., DeWitt, J., Osborne, J., Dillon, J., Willis, B., & Wong, B. (2013). ‘Not girly, not sexy, not glamorous’: Primary school girls’ and parents’ constructions of science aspirations. Pedagogy, Culture & Society , 21 (1), 171–194.

Barad, K. (2003). Posthuman performativity: Toward an understanding of how matter comes to matter. Journal of Women in Culture and Society , 28 (3), 801–831.

Barad, K. (2007). Meeting the universe halfway: Quantum physics and the entanglement of matter and meaning . Durham: Duke University Press.

Bennett, J. (2010). Vibrant matter: A political ecology of things . Durham: Duke University Press.

Bennett, S., Maton, K., & Kervin, L. (2008). The ‘digital natives’ debate: A critical review of the evidence. British Journal of Educational Technology , 39 (5), 775–786.

Butler, J. (1990). Gender trouble: Feminism and the subversion of identity . London: Routledge.

Butler, J. (1993). Bodies that matter: On the discursive limits of sex . London: Routledge.

Calabrese Barton, A., Kang, H., Tan, E., O’Neill, T. B., Bautista-Guerra, J., & Brecklin, C. (2013). Crafting a future in science tracing middle school girls’ identity work over time and space. American Educational Research Journal , 50 (1), 37–75.

Carlone, H. B., & Johnson, A. (2007). Understanding the science experiences of successful women of color: Science identity as an analytic lens. Journal of Research in Science Teaching , 44 (8), 1187–1218.

Carlone, H. B., Scott, C. M., & Lowder, C. (2014). Becoming (less) scientific: A longitudinal study of students’ identity work from elementary to middle school science. Journal of Research in Science Teaching , 51 (7), 836–869.

Charles, M., & Thébaud, S. (Eds.) (2018). Gender and STEM: Understanding segregation in science, technology, engineering and mathematics . MDPI.

Cover, R. (2012). Performing and undoing identity online: Social networking, identity theories and the incompatibility of online profiles and friendship regimes. Convergence: The International Journal of Research into New Media Technologies , 18 (2), 177–193.

Danielsson, A. T. (2011). Exploring woman university physics students ‘doing gender’ and ‘doing physics’. Gender and Education , 24 (1), 25–39.

Dawson, E., Archer, L., Seakins, A., Godec, S., DeWitt, J., King, H., … Nomikou, E. (2020). Selfies at the science museum: Exploring girls’ identity performances in a science learning space. Gender and Education , 32 (5), 664–681.

de Freitas, E., & Curinga, M. X. (2015). New materialist approaches to the study of language and identity: Assembling the posthuman subject. Curriculum Inquiry , 45 (3), 249–265.

Foucault, M. (1988). Technologies of the Self. In L. Martin, H. Gutman, & P. Hutton (Eds.), Technologies of the Self: A Seminar with Michel Foucault , (pp. 16–49). Amherst: University of Massachusetts Press.

Francis, B. (2000). The gendered subject: Students’ subject preferences and discussions of gender and subject ability. Oxford Review of Education , 26 (1), 35–48.

Godec, S. (2018). Sciencey girls: Discourses supporting working-class girls’ to identify with science. Education Sciences , 8 (19), 1–17.

Gonsalves, A. (2014). “Physics and the girly girl – there is a contradiction somewhere”: Doctoral students’ positioning around discourses of gender and competence in physics. Cultural Studies of Science Education , 9 (2), 503–521.

Gonsalves, A., & Danielsson, A. (2017). Identity, masculinity and materiality: Mapping out new terrain in physics education research . Dublin: Presentation at the European Science Education Research Association conference.

Gonsalves, A. J., Silfver, E., Danielsson, A., & Berge, M. (2019). “It’s not my dream, actually”: Students’ identity work across figured worlds of construction engineering in Sweden. International Journal of STEM Education , 6 (1), 1–17.

Goode, J. (2010). The digital identity divide: How technology knowledge impacts college students. New Media & Society , 12 (3), 497–513.

Hall, S., & Du Gay, P. (Eds.) (1996). Questions of cultural identity. Los Angeles, London, New Delhi, Singapore, Washington, DC: Sage.

Kim, A. Y., Sinatra, G. M., & Seyranian, V. (2018). Developing a STEM identity among young women: A social identity perspective. Review of Educational Research , 88 (4), 589–625.

Latour, B. (2005). Reassembling the social: An introduction to actor-network-theory . Oxford: Oxford University Press.

Law, J., & Mol, A. (2002). Complexities: Social studies of knowledge practices . Durham: Duke University Press.

Leander, K. M. (2002). Locating Latanya: The situated production of identity artifacts in classroom interaction. Research in the Teaching of English , 37 (2), 198–250.

Leonardi, P. M. (2010). Digital materiality? How artifacts without matter, matter. First Monday , 15 (6).

Mac An Ghaill, M. (1994). The making of men . Buckingham: Open University Press.

MacLure, M. (2013). Classification or wonder? Coding as an analytic practice in qualitative research. In B. Coleman, & J. Ringrose (Eds.), Deleuze and research methodologies , (pp. 164–183). Edinburgh: Edinburgh University Press.

Mazzei, L. A., & McCoy, K. (2010). Thinking with Deleuze in qualitative research. Journal of Qualitative Studies in Education , 23 (5), 503–509.

McLuhan, M., & Fiore, Q. (1967). The medium is the massage: An inventory of effects . New York: Bantam books.

Mcphie, J. (2018). I knock at the stone’s front door: Performative pedagogies beyond the human story. Parallax , 24 (3), 306–323.

Mendick, H. (2005). A beautiful myth? The gendering of being/doing ‘good at maths’. Gender and Education , 17 (2), 203–219.

Miller, D. (2010). Stuff . Cambridge: Polity Press.

Milne, C., & Scantlebury, K. (Eds.) (2019). Material practice and materiality: Too long ignored in science education . New York: Springer.

Murris, K. (2016). The posthuman child: Educational transformation through philosophy with picturebooks . Oxon: Routledge.

Palmer, A. (2011). “How many sums can I do?”: Performative strategies and diffractive thinking as methodological tools for rethinking mathematical subjectivity. Reconceptualizing Educational Research Methodology , 1 (1), 3–18.

Rainey, K., Dancy, M., Mickelson, R., Stearns, E., & Moller, S. (2018). Race and gender differences in how sense of belonging influences decisions to major in STEM. International Journal of STEM Education , 5 (10), 1–14.

Ringrose, J., & Rawlings, V. (2015). Posthuman performativity, gender and ‘school bullying’: Exploring the material-discursive intra-actions of skirts, hair, sluts, and poofs. Confero , 3 (2), 1–37.

Simpson, A., & Bouhafa, Y. (2020). Youths’ and adults’ identity in STEM: A systematic literature review. Journal for STEM Education Research. https://doi.org/10.1007/s41979-020-00034-y .

Skeggs, B. (2004). Class, Self, Culture . London: Routledge.

Talafian, H., Moy, M. K., Woodard, M. A., & Foster, A. N. (2019). STEM identity exploration through an immersive learning environment. Journal for STEM Education Research , 2 (2), 105–127.

Taylor, C. A. (2013). Objects, bodies and space: Gender and embodied practices of mattering in the classroom. Gender and Education , 25 (6), 688–703.

Thomson, P. (2008). Doing visual research with children and young people . New York; London: Routledge.

Thompson, J. (2014). Engaging girls' sociohistorical identities in science. Journal of the Learning Sciences, 23 (3), 392–446.

Varelas, M. (Ed.) (2012). Identity construction and science education research: Learning, teaching, and being in multiple contexts . Rotterdam: Sense.

Varma, R. (2007). Women in computing: The role of geek culture. Science as Culture , 16 (4), 359–376.

Varma, R. (2010). Why so few women enroll in computing? Gender and ethnic differences in students’ perception. Computer Science Education , 20 (4), 301–316.

Vincent-Ruz, P., & Schunn, C. D. (2018). The nature of science identity and its role as the driver of student choices. International Journal of STEM Education , 5 (48), 1–12.

Wong, B. (2016). ‘I’m good, but not that good’: Digitally-skilled young people’s identity in computing. Computer Science Education , 26 (4), 299–317.

Download references

Acknowledgements

Youth Equity + STEM (YESTEM) is a 4-year Science Learning+ Partnership project (2017–2021). The UK principal investigator is Louise Archer and the co-investigator is Emily Dawson; the US principal investigator is Angela Calabrese Barton and the co-investigators are Carmen Turner and Lynn Dierking.

We presented aspects of this paper at two international conferences, European Conference on Educational Research 2018 in Bolzano and European Science Education Research Association conference 2019 in Bologna. We thank the conference participants for engaging with our work and helping us further our thinking. We would especially like to thank the participants of this study, the young people and the practitioners working within informal STEM learning organisations, without whom this publication would not be possible.

The paper is based upon work supported under a collaboration between the National Science Foundation (NSF), Wellcome and the Economic and Social Research Council (ESRC) via a grant from the NSF (NSF grant no. 1647033) and a grant from Wellcome with ESRC (Wellcome Trust grant no. 206258/Z/17/A). Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the view of NSF, Wellcome or ESRC.

Author information

Authors and affiliations.

Department of Education, Practice and Society, UCL Institute of Education, University College London, 20 Bedford Way, London, WC1H 0AL, UK

Spela Godec, Uma Patel & Louise Archer

UCL Department of Science and Technology Studies, University College London, 22 Gordon Square, London, WC1H 0AW, UK

Emily Dawson

You can also search for this author in PubMed   Google Scholar

Contributions

All authors contributed to the data collection, data analysis and drafting of the paper. SG and UP led the interpretation of the data, writing and revising of the paper. The author(s) read and approved the final manuscript.

Corresponding author

Correspondence to Spela Godec .

Ethics declarations

Competing interests.

The authors declare that they have no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Godec, S., Patel, U., Archer, L. et al. Young people’s tech identity performances: why materiality matters. IJ STEM Ed 7 , 51 (2020). https://doi.org/10.1186/s40594-020-00249-w

Download citation

Received : 17 March 2020

Accepted : 06 September 2020

Published : 06 October 2020

DOI : https://doi.org/10.1186/s40594-020-00249-w

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Identity performativity
• Materiality
• Intra-action

• Open access
• Published: 21 May 2024

The mediating effects of self-efficacy and study engagement on the relationship between specialty identity and career maturity of Chinese nursing students: a cross-sectional study

• Yanjia Liu 1 , 2 ,
• Mei Chan Chong 2 ,
• Yanhong Han 1 ,
• Hui Wang 3 &
• Lijuan Xiong 1  

BMC Nursing volume  23 , Article number:  339 ( 2024 ) Cite this article

197 Accesses

Metrics details

Career maturity is a crucial indicator of career preparedness and unpreparedness can cause the turnover of new nurses. Considerable empirical work demonstrates the potential associations between specialty identity, self-efficacy, study engagement, and career maturity. This study aimed to explore the mediation role of self-efficacy and study engagement on the relationships between specialty identity and career maturity among Chinese nursing students.

Four hundred twenty-six Chinese nursing students were recruited between September 11 and October 30, 2022. The online survey was conducted following the CHERRIES checklist. Electronic questionnaires assessed their perceived specialty identity, self-efficacy, study engagement, and career maturity. The descriptive analysis, Harman single-factor analysis, Pearson correlation tests, structural equation modeling, and the bootstrap method were employed in data analysis.

Bivariate correlation analysis identified a positive correlation between specialty identity, self-efficacy, study engagement, and career maturity ( r  = 0.276–0.440, P  < 0.001). Self-efficacy and study engagement partially mediated the relationship between specialty identity and career maturity. Self-efficacy and study engagement played a chain mediating role between specialty identity and career maturity.

Conclusions

The underlying mechanism can explain the relationships between specialty identity and career maturity: a direct predictor and an indirect effect through self-efficacy and study engagement. Policymakers and educators should emphasize the importance of specialty identity and provide tailored strategies for improving care maturity depending on nursing students’ specialty identity, self-efficacy, study engagement in the early stages of career development.

Peer Review reports

Introduction

The media visibility obtained by nursing during the COVID-19 pandemic has made the public aware of nurses’ role in promoting and maintaining health [ 1 ]. As the social environment becomes more conducive to nursing career development, adequate awareness and preparedness for nursing careers are driving nursing students to adapt to and be satisfied with their careers [ 2 , 3 ].

Career maturity is a crucial indicator of career preparedness [ 4 ], which is defined as the readiness to make age-appropriate career decisions with adequate information and accomplish career development-related tasks [ 5 ]. Unpreparedness and difficulties in taking on the nurse’s role were the main reasons newly graduated nursing students left nursing in their first years [ 6 ]. The turnover rate for new nurses in their first year of employment can reach as high as 69%, with a range of 12.10–69% [ 6 , 7 , 8 , 9 ]. In addition, new nurses who experienced higher levels of career maturity were also less likely to leave the profession [ 10 ]. Therefore, more research focusing on career maturity should re-engage nursing educators and managers and support the development of customized programs in the early stage of career development.

Specialty identity as a predictor to career maturity

Super’s theory emphasizes that career development is a lifelong activity closely related to individual maturity and experiences [ 11 ]. It encompasses the development of behaviors and professional identity [ 12 ]. Work values, including professional identity, are crucial for career development and can influence career maturity [ 13 ]. professional identity significantly correlates with high school students’ career maturity [ 14 ]. Additionally, specialty identity appears as a part of professional identity in studies worldwide [ 15 ]. To clarify this concept in student groups, specialty identity is defined as the emotional acceptance and recognition of learners based on their understanding of the specialty being studied, accompanied by positive external behaviors and an inner sense of satisfaction [ 15 ]. Therefore, this study proposes Hypothesis 1: specialty identity significantly predicts career maturity among Chinese nursing students.

The mediating effect of self-efficacy between specialty identity and career maturity

Self-efficacy and career maturity are positively related [ 16 , 17 ]. According to social cognitive theory, self-efficacy is a belief in a person’s ability to achieve their goal [ 18 ]. Regarding career maturity, self-efficacy could be an internal driver for students to dedicate themselves to the fields they have chosen [ 16 ]. Students with high self-efficacy can improve their professionalism and self-confidence, thereby achieving high degrees of career maturity [ 16 ]. Further, professional identity is found to be significantly correlated with self-efficacy [ 19 , 20 ]. Yao et al. [ 21 ] found that self-efficacy mediated between professional identity and self-reported competence among nursing students. Thus, this study poses Hypothesis 2: Self-efficacy is the mediating variable affecting specialty identity and career maturity among Chinese nursing students.

The mediating effect of study engagement between specialty identity and career maturity

Study engagement is a vital variable related to academic performance, achievement, persistence, and retention, which refers to a positive psychological process including attention, energy and effort in learning [ 22 , 23 ]. Astin’s theory of student involvement emphasizes that the significant environmental factors that can influence their engagement entail students’ backgrounds, such as residence, experiences, and academic involvement [ 24 ]. A significant correlation exists between study engagement and career maturity [ 25 ]. Moreover, Liu et al. [ 26 ] report that professional identity is positively correlated with study engagement, and the mediating role of study engagement in professional identity and career adaptability is significant. Based on the above evidence, this study posits Hypothesis 3: study engagement is the mediating variable affecting specialty identity and career maturity among Chinese nursing students.

The chain mediating effect of self-efficacy and study engagement between specialty identity and career maturity

Based on the aforementioned information, self-efficacy and study engagement may play a single mediating role between specialty identity and career maturity. However, the relationship between self-efficacy and study engagement remains to be clarified. In addition, whether these variables play a chain mediating effect between specialty identity and career maturity must be explored. Previous research has shown that self-efficacy positively correlates with study engagement [ 27 , 28 ]. The relationship between specialty identity and career maturity may be influenced by self-efficacy in the first place and by study engagement in the second. Therefore, this study proposes Hypothesis 4: Chain mediation describes the relationship among the four variables.

Overall, this study explores the relationship between specialty identity and career maturity. It also examines the potential mediation model of specialty identity, self-efficacy, and career maturity, the potential mediation model of specialty identity, study engagement, and career maturity, and the potential chain mediation of the four variables using mediation analysis.

This cross-sectional online survey was conducted among nursing students between September 11 and October 30, 2022. This online survey was designed, disseminated and conducted following the Checklist for Reporting Results of Internet E-Surveys (CHERRIES) [ 29 ] (see in the Supplement File 1 ). The online questionnaire entailed demographic sheet and four instruments with different question styles (single choice and Likert scales).

Variables and data collection instruments

Sociodemographic variables.

Sex (female, male), Higher education institution type (university, college), and Degree (diploma, bachelor’s).

• Specialty identity

The College Student Specialty Identity Scale (CSSIS) developed by Qin [ 15 ], was used to measure medical students’ specialty identity [ 30 ]. It is a 23-item scale with four subscales (cognitive, emotional, behavioral, and appropriateness). Students scored each item on a five-point Likert scale (1–5: strongly disagree to strongly agree). In Qin’s study, it had good reliability (α = 0.955) [ 15 ]. In the present study, Cronbach’s α was 0.949.

• Study engagement

Study engagement was assessed using the Utrecht Work Engagement Scale-Student (UWES-S) [ 31 ]. Schaufeli et al. [ 32 ] developed the UWES and revised its items to measure students’ study engagement. Li & Huang [ 31 ] introduced UWES-S, translated it into Chinese, and validated it among undergraduate students. The UWES-S comprises 17 items grouped into three subscales (vigor, dedication, and absorption) and uses a 7-point scale (0 = never, to 7 = always). The cumulative scores range from 0 to 102, with higher scores indicating greater study engagement. The internal consistency using Cronbach’s alpha was 0.919 [ 31 ]. In this study, Cronbach’s α = 0.956.

• Self-efficacy

This study used the Chinese version of the General Self-Efficacy Scale to assess self-efficacy [ 33 ]. Schwarzer et al. [ 34 ] developed the original version. It was adapted to the context of China and validated by Wang et al. [ 33 ]. It is a 10-item scale with a 4-point Likert scale (from completely incorrect to completely correct). Cronbach’s alpha was 0.871 in Wang et al.’s study [ 33 ]. In this study, Cronbach’s α = 0.899.

• Career maturity

Career maturity was measured using the validated Chinese version of the career maturity scale [ 35 ]. The original version developed by Lee [ 36 ] was translated into Chinese by Zhang et al. [ 35 ]. The instrument entails 34 items broken down into six subscales: career decisiveness (CD), career confidence (CC), career independence (CI), career value (CV), relational dependence (RD), and career reference (CR). A 5-point Likert scale, ranging from strongly disagree to strongly agree, was adopted. The range of total scores was 34–170, with higher values indicating higher levels of career maturity. Its reliability coefficient was 0.86, as measured using Cronbach’s alpha [ 35 ]. In this sample, Cronbach’s α = 0.900.

Participants and data collection procedure

This study was conducted at five higher education institutions in Hubei province, China. The target population, full-time nursing students, were surveyed using convenience sampling. The suggested minimum sample size based on Monte Carlo simulations studies was adopted [ 37 ], and the minimum and maximum sample sizes for structural equation models were 200 and 460 respectively [ 38 ]. The final sample size in the design stage was 250–575, accommodating a possible dropout rate of 20%.

For data collection, we uploaded the integrated questionnaires on Wenjuanxing ( https://www.wjx.cn/ , Acquired NO.168,902,709). This website offers the most popular and convenient tool for anonymous data collection and collecting data anonymously extraction. One investigator from each university or college was invited to collect the data. All investigators held master’s degrees and understood the critical points for questionnaire collection well. Simple training was conducted before questionnaire distribution. Each nursing student can review and change their answers if necessary, but they were provided only one chance to submit the online questionnaires. The individual IP address can be yielded after submissions and provided for verification. Finally, 560 questionnaires were administered. After double-checking and eliminating invalid questionnaires, 426 valid questionnaires were extracted, yielding an effective response rate of 76.07%.

Data analysis

After completing the descriptive analysis, Harman single-factor analysis was performed to assess the common method bias, and Pearson correlations were calculated in SPSS Version 26.0. Subsequently, structural equation modeling was validated, and the chain-mediation effect was examined using the bootstrap method in AMOS Version 23.0 with 5000 samples. The significance level was set at 0.05.

Ethical considerations

This study was approved by the Ethics Committee of Jingmen No. 2 People’s Hospital, affiliated to Jingchu University of Technology (Approval No.2020002-1). All students provided verbal consent to participate in the study and voluntarily completed and submitted the questionnaire.

Harman single-factor analysis

The self-reported nature of the data meant the possibility of common method bias [ 39 ]. The Harman single-factor analysis showed that the eigenvalues of the five common factors were greater than 1. The first common factor explained 35.50% of the variance, which is lower than the recommended threshold of 50% [ 40 ]. Therefore, no common method bias was detected.

Descriptive statistics and correlation analysis

The sociodemographic variables were as follows: female ( n  = 370, 86.85%), male ( n  = 56, 13.15%); university ( n  = 322, 75.59%), college ( n  = 104, 24.41%); freshmen ( n  = 101, 23.71%), sophomores ( n  = 166, 38.97%), juniors ( n  = 127, 29.81%), seniors ( n  = 32, 7.51%); Urban areas ( n  = 151, 35.45%), Rural areas ( n  = 275, 64.55%). The age of nursing students range from 18 to 25 (mean = 19.89, SD=1.27 ) (see Table  1 ).

Table  2 shows the mean scores of the four key variables were 80.40 ± 21.66, 57.56 ± 16.04, 26.23 ± 9.37, and 113.42 ± 31.61. Positive correlations were found between the key variables: specialty identity and study engagement ( r  = 0.276), specialty identity and self-efficacy ( r  = 0.319), specialty identity and career maturity ( r  = 0.300), study engagement and self-efficacy ( r  = 0.420), study engagement and career maturity ( r  = 0.319), and self-efficacy and career maturity ( r  = 0.440) (each p  < 0.001).

The chain-mediation effect analysis

A chain-mediation structural model was constructed with specialty identity as the independent variable, career maturity as the dependent variable, and self-efficacy and study engagement as the mediating variables. The model fitting results showed that χ 2 /df = 2.965, the comparative fit index = 0.963, the Tucker-Lewis index = 0.958, and the root mean square error of approximation = 0.068, indicating good model fit. The chain-mediation effect model diagram of specialty identity, self-efficacy, study engagement, and career maturity of nursing students is shown in Fig.  1 .

The chain mediation effect model diagram of specialty identity, self-efficacy, study engagement, and career maturity of nursing students

The bootstrapping method found that the 95% confidence interval (95% CI) of the chain-mediation path from specialty identity to career maturity was [0.002, 0.054], which did not include 0, indicating significance. Thus, two possible mediation effects were detected: the mediating roles of self-efficacy in the relationship between specialty identity and career maturity and of study engagement in the relationship between specialty identity and career maturity, both of which were significant with a 95% CI (see Table  3 ).

This study explored the relationships between specialty identity, self-efficacy, study engagement, and career maturity and demonstrated the mediation models in Chinese nursing students. The finding identified a positive correlation between specialty identity and career maturity, and specialty identity can influence career maturity in three ways: self-efficacy, study engagement, and self-efficacy → study engagement, supporting the four Hypotheses. Despite this study did not validate the potential confounders such as career resilience [ 20 ], career adaptability [ 26 ], and resource management [ 29 ], the findings may improve our understanding of the underlying mechanism of these four variables and provide meaningful ideas for taking measures to improve nursing students’ career maturity.

In this study, the findings revealed a positive correlation between specialty identity and career maturity, indicating that specialty identity could significantly predict career maturity in nursing students, which is consistent with previous studies [ 14 , 41 ]. However, most nursing students enrolled in nursing school with insufficient specialty identity owing to poor nursing image and a lack of acknowledgment of career growth [ 42 ], meaning their unpreparedness for learning nursing and career development. Specialty identity is an emotional foundation of career maturity, and it can serve as a powerful psychological adjustment when it comes to nursing students’ specialty or job selection. Therefore, the importance of specialty identity on career maturity should be valued by nurse educators and clinical mentors, and further studies should specifically develop and conduct the education program to verify the roles of specialty identity in the early stage of career development. For example, an innovative course about the power of nursing including embracing the healer’s art course, seed talk and reflection exercises was found to connect the nursing students’ values to their specialty identity, and facilitate their professional formation and the development of nursing practice [ 43 ].

The first pathway confirmed was the mediating role of self-efficacy in the relationship between specialty identity and career maturity, aligning with its mediating effect on professional identity and career maturity in a previous study [ 41 ]. When nursing students perceive higher levels of specialty identity, they may have a stronger sense of self-efficacy and achieve greater career maturity. This finding is consistent with the Knowledge–Attitude–Belief–Practice model [ 44 ]. For nursing students, specialty identity and self-efficacy can support attitudes and beliefs about learning nursing specialties [ 21 ] and play the role of internal driving strength in the chase for a feasible professional study plan and career plan. As a result, career maturity could be a feedback indicator for learning behaviors and career preparedness.

This study also verified the mediating effect of study engagement on specialty identity and career maturity. This finding is consistent with a mediation analysis confirming the mediating role of study engagement between professional identity and career maturity among pre-service kindergarten teachers [ 25 ]. This result also supports the predictive impact of study engagement on the beneficial development of careers [ 26 , 45 ]. The mediating effect of study engagement revealed that if nursing students perceive high levels of specialty identity, they might have greater study engagement, achieve more knowledge and skills related to the nursing specialty, and possess high degrees of career maturity to adapt to the nursing profession. However, this study identified the study engagement had a limited mediating effect with a low effect size. The possible reason is that nursing is a specialized and complex discipline, which requires lifelong learning as health needs change and medical technology advances. In a short period, study engagement can improve nursing knowledge and skills, which is conducive to career preparedness, but high levels of career maturity are the result of long-term study engagement especially since this career needs continued education or continued career development [ 46 ].

Additionally, these findings supported the assertion that the chain relationship between self-efficacy and study engagement mediates the relationship between specialty identity and career maturity. The indirect effect of the pathway, including self-efficacy, was greater than that of the chain pathway and the pathway, including a single study engagement. Higher specialty identity could yield higher self-efficacy [ 19 , 20 ], and higher self-efficacy is related to greater study engagement [ 27 , 28 ]. Thus, nursing students with higher specialty identity might have higher self-efficacy and greater study engagement, which leads to higher career maturity. This model also revealed that increased self-efficacy might contribute to nursing students’ high study engagement levels. When nursing students have a sense of high self-efficacy, their learning behaviors become more effective. They are more willing to devote themselves to learning, thus producing higher study engagement. Despite some studies have demonstrated the effect of interventions such as career planning group counseling [ 47 ] and self-reflection-focused career course [ 48 ] on nursing students’ career maturity, what we found in this study provide theoretical foundation for the development and implementation of multifaceted interventions to improve nursing students career maturity and career development. Furthermore, further attention should be given to the interdisciplinary collaborations, such as positive psychology and nursing education, that can be contribute to explore novel perspectives and approaches to studying career maturity.

This study had some limitations. First, the cross-sectional design without a longitudinal method fails to explore the changes in psychological variables over time, which might restrict the temporal and causal inference. Therefore, scholars should focus on exploring the trajectory changes of these variables, notably the mutability of these psychological features, in the further studies, and the longitudinal and sustained interventions like tutor systems and peer learning should be strongly encouraged. Second, the nursing students were selected from five schools in Hubei province, China, which might limit the generalizability to all Chinese nursing students. As there are disparities in the curriculum systems of different schools, the results could be impacted by cognitive errors caused by teaching philosophy and training purposes. Therefore, the potential influencing factors should be considered and other mediators excluding self-efficacy and study engagement also should be explored in further studies. Third, selection bias may arise from the application of convenience sampling. Therefore, scholars could employ probability sampling methods like random stratified sampling to recruit nursing students. Finally, since all instruments were self-reported, the true feelings of these nursing students were not captured or tracked. From this, the research designs to deepen the understanding of the mechanisms underlying nursing students’ career development, such as mixed-method study and qualitative study, should be considered.

A correlational and mediation analysis was used to examine the relationships between four variables. Specialty identity could be a predictive factor for nursing students’ career maturity. Most importantly, specialty identity can indirectly influence career maturity among nursing students through the mediating effect of self-efficacy, study engagement, and the chain mediating effect of self-efficacy and study engagement, supporting career-related theories. Policymakers and educators should focus on the value of specialty identity to promote nursing students’ career development. Specialty identity may be conducive to stimulating students with a strong sense of self-efficacy and robust study engagement. Nursing students with high self-efficacy and study engagement may perceive greater career maturity. Thus, scholars and educators should be encouraged to provide tailored career guidance programs and practical interventions to enhance nursing students’ career maturity in the early stage of career development.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Fawaz M, Anshasi H, Samaha A. Nurses at the Front line of COVID-19: roles,responsibilities, risks, and rights. Am J Trop Med Hyg. 2020;103(4):1341–2. https://doi.org/10.4269/ajtmh.20-0650 .

Article   CAS   PubMed   PubMed Central   Google Scholar  

Kim J, Shin S. Development of the nursing practice readiness scale for new graduate nurses: a methodological study. Nurse Educ Pract. 2022;59:103298. https://doi.org/10.1016/j.nepr.2022.103298 .

Article   PubMed   Google Scholar  

Zhang J, Shields L, Ma B, et al. The clinical learning environment, supervision and future intention to work as a nurse in nursing students: a cross-sectional and descriptive study. BMC Med Educ. 2022;22(1):548. https://doi.org/10.1186/s12909-022-03609-y .

Article   PubMed   PubMed Central   Google Scholar  

Lee I, Heok JW, Rojewski, Roger B. Hill. Classifying Korean adolescents’ career preparedness. Int J Educ Vocat Guid. 2013;13:25–45. https://doi.org/10.1007/s10775-012-9236-5 .

Article   Google Scholar  

Savickas ML. Career maturity: the construct and its measurement. Vocat Guid Q. 1984;32(4):222–31. https://doi.org/10.1002/j.2164-585X.1984.tb01585.x .

Casey K, Fink R, Jaynes C, Campbell L, Cook P, Wilson V. Readiness for practice: the senior practicum experience. J Nurs Educ. 2011;50(11):646–52. https://doi.org/10.3928/01484834-20110817-03 .

Chen SF, Fang YW, Wang MH, Wang TF. Effects of an Adaptive Education Program on the Learning, Mental Health and Work intentions of New Graduate nurses. Int J Environ Res Public Health. 2021;18(11):5891. https://doi.org/10.3390/ijerph181158915 . Published 2021 May 31.

Kovner CT, Brewer CS, Fatehi F, Jun J. What does nurse turnover rate mean and what is the rate? [published correction appears in Policy Polit Nurs Pract. 2017;18(4):216–217]. Policy Polit Nurs Pract. 2014;15(3–4):64–71. https://doi.org/10.1177/1527154414547953 .

Zhang YP, Huang X, Xu SY, Xu CJ, Feng XQ, Jin JF. Can a one-on-one mentorship program reduce the turnover rate of new graduate nurses in China? A longitudinal study. Nurse Educ Pract. 2019;40:102616. https://doi.org/10.1016/j.nepr.2019.08.010 .

Kawai K, Yamazaki Y. The effects of pre-entry career maturity and support networks in workplace on newcomers’ mental health. J Occup Health. 2006;48(6):451–61. https://doi.org/10.1539/joh.48.451 .

Super DE. A life-span, life-space approach to career development, edited by D Brown, L Brooks & Associates. Career choice and development: Applying contemporary theories to practice.1990.2nd edition. San Francisco: Jossey-Bass.

Schreuder AMG, Coetzee M, Careers. An organisational perspective. 2011. 4th edition. Cape Town: Juta.

Langley R, Du Toit R, Herbst DL. Manual for the career development questionnaire. Pretoria: Human Sciences Research Council. 1992; 157:373–422.

Dipeolu A, Deutch S, Hargrave S, Storlie CA. Developmentally relevant Career constructs: response patterns of youth with ADHD and LDs. Can J Career Dev. 2019;18(1):45–55.

Google Scholar  

Qin P. B.The characteristics and correlation study of college students’ specialty identity [Chinese]. Master thesis: Southwest University.2009.

Purwandika R, Ayriza Y. The influence of self-efficacy on career maturity of high school students in pacitan regency. In2nd International Seminar on Guidance and Counseling 2019 (ISGC 2019) 2020; (pp. 93–7). Atlantis.

Singh PK, Shukla RP. Relationship between career maturity and self-efficacy among male and female senior secondary students. MIER J Educational Stud Trends Practices. 2015 Nov;10:164–79. https://doi.org/10.52634/mier/2015/v5/i2/1486 .

Bandura A. On the functional properties of perceived self-efficacy revisited. J Manag. 2012;38(1):9–44.

Mei XX, Wang HY, Wu XN, Wu JY, Lu YZ, Ye ZJ. Self-Efficacy and Professional Identity among freshmen nursing students: a latent Profile and Moderated Mediation Analysis. Front Psychol. 2022;13:779986. https://doi.org/10.3389/fpsyg.2022.779986 .

Zhou Y, Chen S, Deng X, Wang S, Shi L. Self-efficacy and career resilience: the mediating role of professional identity and work passion in kindergarten teachers. J Psychol Afr. 2023;33(2):165–70. https://doi.org/10.1080/14330237.2023.2207052 .

Yao X, Yu L, Shen Y, Kang Z, Wang X. The role of self-efficacy in mediating between professional identity and self-reported competence among nursing students in the internship period: a quantitative study. Nurse Educ Pract. 2021;57:103252. https://doi.org/10.1016/j.nepr.2021.103252 .

Bond M, Buntins K, Bedenlier S, Zawacki-Richter O, Kerres M. Mapping research in student engagement and educational technology in higher education: a systematic evidence map. Int J Educational Technol High Educ. 2020;17(1):1–30. https://doi.org/10.1186/s41239-019-0176-8 .

Schaufeli WB, Martinez IM, Pinto AM, Salanova M, Bakker AB. Burnout and engagement in university students: a cross-national study. J Cross-Cult Psychol. 2002;33(5):464–81.

Astin AW. Student involvement: a developmental theory for higher education. InCollege student development and academic life, 2014 (pp. 251–62). Routledge.

Zhang L, Chen M, Zeng X, Wang X. The relationship between professional identity and career maturity among pre-service kindergarten teachers: the mediating effect of learning engagement. Open J Social Sci. 2018;6(6):167–86. https://doi.org/10.4236/jss.2018.66016 .

Liu X, Ji X, Zhang Y, Gao W. Professional Identity and Career adaptability among Chinese Engineering students: the Mediating Role of Learning Engagement. Behav Sci (Basel). 2023;13(6):480. https://doi.org/10.3390/bs13060480 .

Wu H, Li S, Zheng J, Guo J. Medical students’ motivation and academic performance: the mediating roles of self-efficacy and learning engagement. Med Educ Online. 2020;25(1):1742964. https://doi.org/10.1080/10872981.2020.1742964 .

Eysenbach G. Improving the quality of web surveys: the Checklist for reporting results of internet E-Surveys (CHERRIES). J Med Internet Res. 2004;6(3):e34. https://doi.org/10.2196/jmir.6.3.e34 .

Heo H, Bonk CJ, Doo MY. Influences of depression, self-efficacy, and resource management on learning engagement in blended learning during COVID-19. Internet High Educ. 2022;54:100856. https://doi.org/10.1016/j.iheduc.2022.100856 .

Zhang X, Sun B, Tian Z, et al. Relationship between honesty-credit, specialty identity, career identity, and willingness to fulfill the contract among rural-oriented tuition-waived medical students of China: a cross-sectional study. Front Public Health. 2023;11:1089625. https://doi.org/10.3389/fpubh.2023.1089625 .

Li X, Huang R. A revise of the UWES-S of Chinese college samples. Psychol Res. 2010;3(1):84–8.

Schaufeli WB, Salanova M, González-Romá V, Bakker AB. The measurement of engagement and burnout: a two sample confirmatory factor analytic approach. J Happiness Stud. 2002;3:71–92.

Wang CK, Hu ZF, Liu Y. Reliability and validity of General self-efficacy scale. Chin J Appl Psychol. 2001;7(1):37–40. https://doi.org/10.396 9/j.issn.1006-6020.2001.01.007. (in Chinese).

Schwarzer R, Mueller J, Greenglass E. Assessment of perceived general self-efficacy on the internet: data collection in cyberspace. Anxiety Stress Coping. 1999;12(2):145–61.

Zhang ZY, Rong Y, Guan YJ. Reliability and validity of Career Maturity Scale for Chinese College Students. J Southwest Normal Univ (Humanities Social Sci Edition). 2006;32(5):1–6. (in Chinese).

Lee KH. A cross-cultural study of the career maturity of Korean and United States high school students. J Career Dev. 2001;28(1):43–57. https://doi.org/10.1023/A:1011189931409 .

Kyriazos TA. Applied psychometrics: sample size and sample power considerations in factor analysis (EFA, CFA) and SEM in general. Psychology. 2018;9(08):2207. https://doi.org/10.4236/psych.2018.98126 .

Wolf EJ, Harrington KM, Clark SL, Miller MW. Sample size requirements for structural equation models: an evaluation of power, bias, and solution propriety. Educ Psychol Meas. 2013;73(6):913–34.

Jordan PJ, Troth AC. Common method bias in applied settings: the dilemma of researching in organizations. Aust J Manage. 2020;45(1):3–14. https://doi.org/10.1177/0312896219871976 .

Duong CD. Psychological distress related to Covid-19 in healthy public (CORPD): a statistical method for assessing the validation of scale. MethodsX. 2022;9:101645. https://doi.org/10.1016/j.mex.2022.101645 .

Kornspan AS, Etzel EF. The relationship of demographic and Psychological Variables to Career Maturity of Junior College Student-Athletes. J Coll Student Dev. 2001;42(2):122–32.

Kandil F, El Seesy N, Banakhar M. Factors affecting students’ preference for nursing education and their intent to leave: a cross-sectional study. Open Nurs J. 2021;15(1):1–8. https://doi.org/10.2174/187443460211501000 .

WHO. Knowledge, Attitudes, and Practices (KAP) Surveys during Cholera Vaccination Campaigns: Guidance for Oral Cholera Vaccine Stockpile Campaigns. [(accessed on 15 July 2023)]. Available online: https://www.who.int/publications/m/item/knowledge-attitudes-and-practices-(kap)-surveys-during-cholera-vaccination-campaigns-guidance-for-oral-cholera-vaccine-stockpile-campaigns .

Peng MY, Yue X. Enhancing Career decision status of socioeconomically disadvantaged students through Learning Engagement: perspective of SOR Model. Front Psychol. 2022;13:778928. https://doi.org/10.3389/fpsyg.2022.778928 .

Day L, Ziehm SR, Jessup MA, Amedro P, Dawson-Rose C, Derouin A, Kennedy BB, Manahan S, Parish AL, Remen RN. The power of nursing: an innovative course in values clarification and self-discovery. J Prof Nurs. 2017;33(4):267–70. https://doi.org/10.1016/j.profnurs.2017.01.005 .

Mlambo M, Silén C, McGrath C. Lifelong learning and nurses’ continuing professional development, a metasynthesis of the literature. BMC Nurs. 2021;20(1):62. https://doi.org/10.1186/s12912-021-00579-2 .

Xie WY, Yang XL, Cai YM, Mo W, Shen ZM, Li YH, Zhou BF, Li YL. Evaluation of career planning group counseling and its effectiveness for intern male nursing students. BMC Med Educ. 2023;17(1):34. https://doi.org/10.1186/s12909-022-03981-9 .

Kim JH, Shin HS. Effects of self-reflection-focused career course on career search efficacy, career maturity, and career adaptability in nursing students: a mixed methods study. J Prof Nurs. 2020;36(5):395–403. https://doi.org/10.1016/j.profnurs.2020.03.003

Download references

Acknowledgements

We greatly appreciate the support of all nursing students from five schools in Hubei province for their voluntary participation in this study.

Not applicable.

Author information

Authors and affiliations.

Department of Nursing, Wuhan Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China

Yanjia Liu, Yanhong Han & Lijuan Xiong

Department of Nursing Science, Faculty of Medicine, Universiti Malaya, Kuala Lumpur, Malaysia

Yanjia Liu & Mei Chan Chong

Department of Medicine, Jingchu University of Technology, Jingmen, Hubei, China

You can also search for this author in PubMed   Google Scholar

Contributions

Study design: XLJ, CMC, HYH; Data collection and analysis: WH, LYJ; Manuscript preparation: LYJ, HYH. All authors reviewed the manuscript.

Corresponding author

Correspondence to Lijuan Xiong .

Ethics declarations

Ethics approval and consent to participate.

This study has been approved by the Ethical Review Committee of Jingmen No.2 People’s Hospital, affiliated with Jingchu University of Technology (Approval No.2022002-1). All methods were performed in accordance with the relevant guidelines and regulations. Informed consent was obtained from all subjects involved in the study.

Consent for publication

Not application.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Liu, Y., Chong, M.C., Han, Y. et al. The mediating effects of self-efficacy and study engagement on the relationship between specialty identity and career maturity of Chinese nursing students: a cross-sectional study. BMC Nurs 23 , 339 (2024). https://doi.org/10.1186/s12912-024-02002-y

Download citation

Received : 13 December 2023

Accepted : 08 May 2024

Published : 21 May 2024

DOI : https://doi.org/10.1186/s12912-024-02002-y

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Nursing students
• Mediating effect

BMC Nursing

ISSN: 1472-6955

• General enquiries: [email protected]