Slide 1

Am I stressed and can We cope? A social identity perspective on stress at work Rolf van Dick Goethe Universitt, Frankfurt Department of Psychology and Center for Leadership and Behavior in Organizations Associazione Italiana di Psicologia Palermo 18 September 2015 Slide 2 in collaboration with and a big Thank You to. Alex Haslam, Jrgen Wegge, Andreas Mojzisch, Jan Husser, Ullrich Baumann, Oliver Christ, Jolanda Jetten, Tom Postmes, Anne OBrien, Steve Reicher, Sebastian Schuh, Sonja Rohrmann, Mona Wolf, Johannes Ullrich, Lorenzo Avanzi, Jordi Escartin, Nik Steffens, Johanna Frisch, Ulrich Wagner, Ulrich Baumann, Imeke Goldschmidt Slide 3 3 Overview The social identity approach Social identity, well-being and mental health Hypotheses Preliminary evidence Further evidence Experimental data: In the lab Longitudinal data: Stress in a theatre production Case study data: Well-being in the BBC Prison Study Meta-analytical evidence Conclusions Slide 4 4 The social identity approach Incorporating principles from social identity theory (Tajfel & Turner, 1979, 1986) and self-categorization theory (Turner, 1985; Turner et al., 1987, 1994). Argues that there is a qualitative difference between behaviour that is based on personal identity (I) and that based on social identity (we). Slide 5 When a social identity (a social self-category; e.g., us psychologists, us academics) becomes salient, this has a number of important implications for social and organizational behaviour (Haslam, 2004; van Dick, 2004). In particular, those who share this identity are more likely to: perceive themselves as relatively interchangeable engage in mutual influence (because they perceive themselves to have a common perspective on reality) co-ordinate their behaviour with reference to emergent group norms (e.g., those that define the group as positively distinct from other groups) work collaboratively to advance the interests of the group as a whole. soziale Identitt (Academics) The social identity approach Slide 6 6 These ideas have implications for a broad range of social and organizational issues (e.g., Haslam, 2004; Haslam et al., 2003; van Dick, 2001, 2004; Riketta & van Dick, 2005). e.g., leadership, motivation, turnover, productivity (e.g., van Dick et al., 2004a, 2004b, 2007; Ullrich, Christ & van Dick, 2009; Wieseke, Ahearne, Lam, & van Dick, 2009) e.g., citizenship: To the extent that people define themselves in terms of social identity, they should be more willing to engage in behaviour which promotes the interests of the group to which that identity relates even if this involves personal cost. (e.g., Christ, van Dick, Wagner, & Stellmacher, 2003; van Dick et al., 2006) But do they have any implications for well-being and mental health? Slide 7 7 Primary appraisal: Is this stressful? To the extent that social identity is salient, answer will be determined by implications of stressor for ones group rather (just) than ones personal self. Stressful for me? Stressful for us? Personal identity salient Social identity salient primary When a shared social identity is salient, this should impact on both primary and secondary appraisal (Lazarus & Folkman, 1984): Social identity, well-being and mental health: Hypotheses (see Haslam & van Dick, 2011; van Dick & Haslam, 2012) Slide 8 8 When a shared social identity is salient, this should impact on both primary and secondary appraisal (Lazarus & Folkman, 1984): social support Can we cope? Can I cope? secondary Secondary appraisal: Is it possible to cope? To the extent that social identity is salient, answer will be determined by the resources and condition of ones ingroup and the social support it provides rather than personal self. Stressful for me? Stressful for us? Personal identity salient Social identity salient primary Social identity, well-being and mental health: Hypotheses Slide 9 9 Among other things, this is because social identity provides the basis for a shared cognitive framework which means any help will be interpreted in the spirit in which it is intended. This is also associated with identity-based trust (reduced paranoia). It should also mean that individuals have greater potential for collective self-actualization. This should manifest itself in an enhanced sense of collective accomplishment. Social identity, well-being and mental health: Hypotheses To the extent that they define themselves in terms of shared group membership, individuals should provide, receive, and benefit from the social support of fellow ingroup members (Branscombe et al., 1999; Postmes & Branscombe, 2002). Slide 10 10 When someone praises my organization, it feels like a personal compliment. When someone criticizes my organization, it feels like a personal insult. I am very interested in what others think about my organization. When I talk about my organization, I usually say we rather than they. My organizations successes are my successes. Social identity, stress and mental health: Measurement (e.g., Mael & Ashforth, 1992; van Dick et al., 2004; Doosje et al., 2005) Slide 11 11 Van Dick & Wagner, 2002, 2 studies among German schoolteachers Study 1 (N=201): organizational identification physical symptoms: r=-.30 Study 2 (N=283): occupational identification physical symptoms: r=-.41; team identification physical symptoms: r=-.28 Van Dick at al. (2004; N=459): Employees after an organizational merger showed most negative emotions when pre-merger identification was high and post-merger identification low Wegge & van Dick (2006; N=161 call center employees): Org. identification burnout ( EE: r=-.28; PA: r=.48; DP: r=-.40) Social identity, well-being and mental health: First (correlational) evidence Slide 12 12 Survey data: Employees reports of stress and well-being (Haslam et al., BJSP, 2005) Is social identification associated with increased social support and does this protect individuals from adverse effects of stress? Survey of bar workers and bomb disposal experts (N = 40). Self-report measures Social identification with workteam (2 items, =.81) Social support (3 items, =.81) Work stress (5 items, =.68) Perceived stress of bomb disposal work (3 items, =.95) Perceived stress of bar work (2 items, =.71) Slide 13 13 Bar workers find handling bombs more stressful than bar work, but bomb handlers report the opposite. Group membership provides a basis for interpreting and normalizing stress Perceived stressfulness of work Survey data: Employees reports of stress and well-being (Haslam et al., BJSP, 2005) Slide 14 14 In both groups, social identification is associated with (a) receipt of more social support (b) less stress Relationship between social identification and well-being is mediated by social support. social identification Stress -.47** social support.55** -.56** -.33* Survey data: Employees reports of stress and well-being (Haslam et al., BJSP, 2005) Slide 15 15 Social identification is associated with (a) receipt of more social support (b) less stress social identification Stress -.33** social support.61** -.51**.04 Survey data: Patients reports of stress and well-being post-surgery (Haslam et al., BJSP, 2005) Relationship between social identification and well-being is mediated by social support. Slide 16 16 Organizational identification is associated with (a) receipt of more social support from colleagues (b) feelings of higher collective efficacy (c) lower burnout Survey data: 192 Italian high school teachers (Avanzi, Schuh, Fraccaroli, & Van Dick, 2015) Slide 17 17 Social identification (with ones family) reported by relatives relates to (a)Social support (reported by relatives) (b) Childrens medical parameters 1. better kidney functions (Kreatinin, r=-.34*) 2. less damaged liver cells (GPT, r=-.32*) 3. better liver performance (Quick, r=-.36*) Family identity and children with liver diseases Survey of 45 relatives who donated parts of their livers and medical data from receiving children (Baumann, van Dick et al., in prep.) Slide 18 18 The above studies provide data which is consistent with, and helps flesh out, a social identity approach to stress-related issues. However, like most other research in the field this data is (a) correlational (b) cross-sectional, and (c) based on self-report. As a result, it doesnt allow us to disentangle cause and effect, or look at stress, support and well-being as they evolve in the context of developing group dynamics. Survey data: Reports of stress, support and well-being Slide 19 19 Experimental data 1: Students appraisal of stress and well-being (Haslam et al., S&H, 2004) Does social identification affect appraisal? Modification of classic Lazarus (e.g., 1966) study. Participants exposed to message that an arithmetic task is challenging or stressful. But message emanates from person described as either ingroup member (a fellow student), or outgroup member (a stress sufferer). Outcome measures: Self-reported stress while performing task (22 items, =.92). Slide 20 20 Does social identification affect appraisal? Reported stress Yes. Appraisal (and associated reaction to a potential stressor) only responds to informational support if this emanates from an ingroup source. Experimental data 1: Students appraisal of stress and well-being (Haslam et al., S&H, 2004) Slide 21 21 Experimental data 2: Call Center Simulation (Wegge, Schuh & van Dick, S&H, 2012) Does identification buffer against stress? N=96 experienced call-center agents were invited to the laboratory, completed a questionnaire on the organizational identification and were then subjected to faked customers in a polite versus unpolite condition DV: Changes in Immunoglobulin A levels Slide 22 22 Experimental data 2: Call Center Simulation (Wegge, Schuh & van Dick, S&H, 2012) Does identification buffer against stress? Slide 23 23 Experimental data 3: Identity as stress buffer (Husser, Kattenstroth, van Dick & Mojzisch, JESP, 2012) Does shared identity buffer against stress? Students were assigned to one of two conditions in groups with high versus low shared identity Next, they completed the Trier Social Stress Test: Giving a job talk to a jury Counting backwards in steps of 17, starting from 2043 DV: Changes in Cortisol levels Slide 24 24 Does shared identity buffer against stress? personal Identity social Identity s alivary cortisol (nmol/l) -15 min + 5 min + 20 min + 30 min+ 40 min Experimental data 3: Identity as stress buffer (Husser, Kattenstroth, van Dick & Mojzisch, JESP, 2012) Slide 25 25 Does shared identity helps utilize social support? Experimental data 4: Identity as stress buffer (Frisch, Husser, van Dick & Mojzisch, 2014) non-supportive supportive Slide 26 26 Experimental data 4: Identity as stress buffer (Frisch, Husser, van Dick & Mojzisch, 2014) non-supportive supportive Does shared identity helps utilize social support? Slide 27 27 Longitudinal data: Stress in a theatre production (Haslam et al., S&H, 2009) Is social identification associated with reduced burnout and does this contribute to the long-term impact of stress? Five-phase study of theatre production staff (N = 30). Self-report measures: Social identification with production (3 items, mean =.86) Burnout (5 items, mean =.62) Organizational citizenship (3 items, mean =.82) Slide 28 28 Self-report measures Examined as a function of social identification (high, low) on five occasions: (1) after audition (2) mid-rehearsal (3) after dress rehearsal (4) after final production (5) two weeks after the final production Longitudinal data: Stress in a theatre production (Haslam et al., S&H, 2009) Slide 29 29 Social identification is relatively enduring Longitudinal data: Stress in a theatre production (Haslam et al., S&H, 2009) Slide 30 30 Low identifiers experience more burnout during critical phases of production (dress rehearsal, first performance). Longitudinal data: Stress in a theatre production (Haslam et al., S&H, 2009) Slide 31 31 Social identification is predictive of long-term citizenship Longitudinal data: Stress in a theatre production (Haslam et al., S&H, 2009) Slide 32 32 Longitudinal design allows us to look at unfolding relationship between social identification, burnout and long- term well-being Social identification has positive impact on long-term citizenship because it protects group members from burnout during critical phases of group activity social identification (T1) citizenship (T5).39** burnout (T3,4) .39** .40*.24 Longitudinal data: Stress in a theatre production (Haslam et al., S&H, 2009) Slide 33 33 Case study data: Well-being in the BBC Prison Study (Haslam & Reicher, JAP, 2006) One study with unique potential was conducted in 2001 with the BBC: The BBC Prison Study This was designed as an intensive examination of group dynamics, and involved studying the behaviour of normal healthy volunteers after they had been randomly assigned to be either Prisoners or Guards in a simulated prison. Provides opportunity for an integrated analysis of social, organizational and clinical issues (Haslam & Reicher, SciAmMind, 2005, BJSP, 2006, JAP, 2006; Reicher & Haslam, Psychologist, 2006a, BJSP, 2006b; Reicher, Haslam, Hopkins, LQ, 2005). Slide 34 34 More specifically, the study was designed to test predictions concerning (a) the conditions under which people would define themselves in terms of social identity (b) the consequences of this for social, clinical and organizational functioning Case study data: Well-being in the BBC Prison Study (Haslam & Reicher, JAP, 2006) The study lasted 8 days. Observational data (800+ hrs of film footage), physiological and psychological data related to 60+ key social, clinical and organizational states. Slide 35 35 Over the course of the study Haslam et al. manipulated factors that SIT predicts should encourage members of low-status groups (Prisoners) to act in terms of social identity and resist inequality (Tajfel & Turner, 1979) 1 Day 1 boundaries permeable (possible to be promoted from Prisoner to Guard). Prisoners should act as individuals and maintain status quo. 3 Day 5 exposure to cognitive alternatives (introduce trade unionist as new Prisoner). Prisoners should act more as a group and challenge status quo. Day 3 boundaries impermeable (not possible to be promoted from Prisoner to Guard). Prisoners should act more as a group. 2 Case study data: Well-being in the BBC Prison Study (Haslam & Reicher, JAP, 2006) Slide 36 36 As predicted, these manipulations led to an increase in the Prisoners social identity over time. Social identification Case study data: Well-being in the BBC Prison Study (Haslam & Reicher, JAP, 2006) Slide 37 37 As predicted, these manipulations did lead to an increase in the Prisoners social identity over time. Moreover, as their sense of shared social identity increased, the Prisoners started to resist the Guards authority (seen on measures of compliance and organizational citizenship). Their resistance ultimately culminated in a breakout which made the Guards regime collapse. Case study data: Well-being in the BBC Prison Study (Haslam & Reicher, JAP, 2006) Slide 38 38 In the face of the Prisoners increasing power, the Guards became apprehensive about their own power and authority. This, combined with the Prisoners insurrection, contributed to a steady decline in their sense of shared social identity. Social identification Case study data: Well-being in the BBC Prison Study (Haslam & Reicher, JAP, 2006) Slide 39 39 What were the consequences of these emergent dynamics for participants well-being and mental health? Clinical Depression Depression of prisoners reduces as social identity increases Depression of guards increases as social identity declines Case study data: Well-being in the BBC Prison Study (Haslam & Reicher, JAP, 2006) Slide 40 40 Cortisol in saliva ( g/10ml) Stress of guards increases as social identity declines Case study data: Well-being in the BBC Prison Study (Haslam & Reicher, JAP, 2006) Slide 41 41 Burnout Burnout of guards (esp. callousness and sense of lack of accomplishment) increases as social identity declines and group fails Case study data: Well-being in the BBC Prison Study (Haslam & Reicher, JAP, 2006) Slide 42 Organizational Identification stress, health, well-being, burnout: 42 independent samples, 63 effect sizes, N = 16.297 42 Meta-analytical evidence (Steffens, Schuh, Haslam, Jetten, & van Dick, in prep.) Slide 43 Slide 44 OI: r =.23, 95%CIs: [.18,.27] - Fail safe N= 222 44 Meta-analytical evidence (Steffens, Schuh, Haslam, Jetten, & van Dick, in prep.) Slide 45 45 Conclusions Considered together, these studies provide diverse and vivid evidence that social identity processes are central to the dynamics of well-being in social and organizational contexts. Shared social identity provides a basis for group members to give and receive social support and to derive feelings of collective accomplishment from working collectively to overcome and remove stressors. On the other hand, lack or loss of social identity increases the likelihood of paranoia (reflecting lack of trust and support), disrespectful treatment (bullying), negative reaction to stressors, burnout and depression. Slide 46 46 If groups fail to achieve ingroup-defining goals, this may also be particularly stressful for those who identify strongly with them. and the well-being of others who dont embody group norms (e.g. bullying of outgroup and ingroup members (see: Ullrich, Escartin, Schlter, Zapf, & van Dick, EJWOP, 2013 ). When social identity is salient, individuals desire to contribute to the group and to live up to, and enforce, group norms may mean they are willing to compromise personal well-being (e.g., leading to exhaustion; see Avanzi, van Dick, Fraccaroli & Sarchielli, W&S, 2012). Potential downsides Slide 47 47 Conclusions It is worth noting that our analyses go against dominant psychological approaches which are profoundly individualistic. For example, these see stress as a personal response to a particular stimulus situation, and hence best predicted by personality (hence massive literature on Type A; hardiness, coping styles, etc.) and best treated by individual-level responses (e.g., cognitive restructuring, relaxation therapy, stress counselling). We believe, however, that social identity-based group viability is central to well-being it plays a key role in determining whether stressors change us (through avoidance or denial) or we try to change them (through resistance). Slide 48 Molte grazie!