226 Personality and Social Psychology Review 18(3) Table 4.Measures of Social ldentification Scale name Scale items Reference Single-tem Social I identity with [In-group]. Postmes.Haslam and Jans (2013) Four-ltem Social ldentification 1.lidentify with [-group]. measure 3 am glad to be[ Oakes. and Koomen.(1998)and Leach et al (008) Group Mem ed in the activities Note.All items are measured on a 7-point Likert-type scale from strongly disagree (1)to strongly agree (7). not account for any unique variance after social identifica Evidence for H5:Social identification will tion was accounted for determin the impact of th earoussocialfactor (1999), easured three components depression affect.All three comp the SIA give tively related to depr ssion.However,the strongest rela tionship was with the in-group ties component of socia edness.However,this is not to say that other aspects of grou cation( d by ag s social support (instrumental and emot belong).This adds to the evidence that it is the perceived nda belonging dimension of social connectedness(Category3) group membership that arise from,and are facilitated by,a outcom inally,in sense of shared social identity e Consist ent with this claim,the ial cal-psychol gical litera that this intervention had the effect of decreasing depres sion symptoms most among those participants example.manipulations that increase perceived shared grour ed a strong sense of belonging with the social group with H4 all the above studies sugg st that &Re200 002 1999),and social support (emotional,intellectual,and mate at measure et al. 2005;S.A.Haslam,Reicher,& job of capturing the psychological substrates of dep e h little if k has been focused or sion than other more objective indicators.Critically to )it issues pertaining directly to depression.in two studies that ins wh here is a difference b psych most closely address H5,social support was found to medi e he negative ship be tio why individuals can feel most alone in a crowd,or why,as imental studies have shown that it is onb when a shared group membership is psychologically salien the same year own that potentially supportive actions are perce ed as support or Bangkok."one can have the ex rience of "sittin ve.For example,stu under the same glass bell jar,stewing in my own sour air" than a threat when this encour ement came from an in. (p.185). group member(S.A.Haslam,Reicher,&Levine,2012).226 Personality and Social Psychology Review 18(3) not account for any unique variance after social identifica￾tion was accounted for. H4 is also consistent with evidence reported by Cameron (1999), who measured three components of social identifi￾cation: identity centrality, in-group ties, and in-group affect. All three components were significantly and nega￾tively related to depression. However, the strongest rela￾tionship was with the in-group ties component of social identification (as assessed by agreement with items such as “In a group of Mt Allison students, I really feel like I belong”). This adds to the evidence that it is the perceived belonging dimension of social connectedness (Category 3) that is most pertinent to health outcomes. Finally, in a study of disadvantaged people who were facilitated to join social groups (Category 2), Cruwys et al. (in press) found that this intervention had the effect of decreasing depres￾sion symptoms most among those participants who reported a strong sense of belonging with the social group (Category 3). Consistent with H4, all the above studies suggest that Category 3 measures of social connectedness are better predictors of depression than Category 1 or 2 measures. In other words, it appears that measures of individuals’ own sense of their social connection to others do a rather better job of capturing the psychological substrates of depres￾sion than other more objective indicators. Critically too, it also explains why there is a difference between psycho￾logical states such as loneliness and the material reality of being alone. Among other things, this is needed to explain why individuals can feel most alone in a crowd, or why, as Sylvia Plath put it more poetically in The Bell Jar (written in 1963, the same year that she took her own life), how even when “on the deck of a ship or at a street café in Paris or Bangkok,” one can have the experience of “sitting under the same glass bell jar, stewing in my own sour air” (p. 185). Evidence for H5: Social identification will determine the impact of the various social factors (e.g., social support) that are implicated in depression As the previous section suggests, the SIA gives primacy to social identification as the key ingredient of social connect￾edness. However, this is not to say that other aspects of group life, such as social support (instrumental and emotional), trust, and helping are not important—they clearly are. Rather, these are conceptualized as “secondary benefits” of social group membership that arise from, and are facilitated by, a sense of shared social identity.2 Consistent with this claim, the social-psychological litera￾ture provides copious experimental evidence of these “sec￾ondary benefits” flowing from social identification. For example, manipulations that increase perceived shared group membership have been shown to promote trust (Foddy, Platow, & Yamagishi, 2009; Tanis & Postmes, 2005), help￾ing behavior (Levine, Cassidy, Brazier, & Reicher, 2002; M. Levine, Prosser, Evans, & Reicher, 2005; Platow et al., 1999), and social support (emotional, intellectual, and mate￾rial; S. A. Haslam et al., 2005; S. A. Haslam, Reicher, & Levine, 2012; M. Levine & Thompson, 2004). Although little, if any, of this work has been focused on issues pertaining directly to depression, in two studies that most closely address H5, social support was found to medi￾ate the negative relationship between social identification and both stress and well-being (S. A. Haslam et al., 2005). Furthermore, experimental studies have shown that it is only when a shared group membership is psychologically salient that potentially supportive actions are perceived as support￾ive. For example, students preparing for a challenging exam were only encouraged to construe this as a challenge rather than a threat when this encouragement came from an in￾group member (S. A. Haslam, Reicher, & Levine, 2012). Table 4. Measures of Social Identification. Scale name Scale items Reference Single-Item Social Identification measure I identity with [In-group]. Postmes, Haslam and Jans (2013) Four-Item Social Identification measure 1. I identify with [In-group]. 2. I feel committed to [In-group]. 3. I am glad to be [In-group]. 4. Being [In-group] is an important part of how I see myself. Adapted from Doosje, Spears and Ellemers (1995). See also Doosje, Haslam, Spears, Oakes, and Koomen. (1998) and Leach et al. (2008) Multiple Group Memberships Scale 1. I belong to lots of different groups. 2. I am involved in the activities of lots of different groups. 3. I have friends who are in lots of different groups. 4. I have strong ties with lots of different groups. Adapted from the Exeter Identity Transition Scale, see C. Haslam et al. (2008) Note. All items are measured on a 7-point Likert-type scale from strongly disagree (1) to strongly agree (7). Downloaded from psr.sagepub.com at Remen University of China on September 6, 2015