Effect of social identity salience on healthy eating intentions and behaviour

Abstract

BACKGROUND: Self-categorisation theory and the identity-based motivation perspective suggest that people’s motivation to engage in a particular behaviour is stronger when that behaviour is congruent with their salient social identity. In situations where a certain social identity is made salient, or where people identify strongly with a particular group, the social norm associated with that group may have a strong effect on individual behaviour. This perspective can be used to enhance the understanding of health-related intentions and behaviour. The aim of this thesis is to investigate the usefulness of adding concepts related to social identity to existing social cognitive models of healthy eating. The prediction being made is that members of groups that value healthy eating might be more likely to engage in healthy eating when their membership in that particular group is made salient.

Five experimental studies tested the effect of social identity salience and group identification on healthy eating intentions and behaviour. Both intentions and behaviour were measured in each of the five studies, to allow for investigating the existence and potential causes of the intention-behaviour gap for healthy eating.

METHODS AND RESULTS: All five studies included random assignment of participants to conditions, and an experimental manipulation of social identity salience or social image healthiness. In Study 1 (n = 149), conducted among female university students, participants’ female, family, or personal identity was made salient. The results showed that increasing the salience of female or family identity led to stronger healthy eating intentions, but did not increase the likelihood of picking a healthy snack over an unhealthy one. Study 2 (n = 115) did not include a successful manipulation of salient social identity, but it showed a positive association between female identification, measured as a trait, and healthy eating intentions, even after controlling for attitude, subjective norm and perceived behavioural control. Study 3 (n = 156) included a manipulation of social identity salience (female or student) and a manipulation of social image healthiness (images presenting in-group members engaging in either healthy or unhealthy behaviour). The results corroborated the earlier finding that female identification is positively correlated with healthy eating intentions. Also, the results indicated that when participants were shown social images of their in-group members engaging in healthy or unhealthy behaviour, they expressed intentions in line with the social images only if they did not express strong identification with the in-group. Study 4 (n = 87) was conducted in the context of Australian identity and included a manipulation of social images healthiness. The findings provided evidence for the existence of a vicarious licensing effect for healthy eating. Namely, for participants who highly identified with their social group, exposure to pictures of other in-group members engaging in healthy behaviour resulted in choosing less healthy food items from a restaurant menu. Study 5 (n = 117) demonstrated the existence of a vicarious licensing effect in the context of female identity, where participants’ food intake during a taste test was predicted by the interaction of the social image healthiness and their group identification.

CONCLUSIONS: By examining the predictors of both healthy eating intention and behaviour, the research presented in this thesis sheds light on some of the phenomena potentially underlying the intention-behaviour gap for healthy eating, particularly among women. It appears that the healthy eating norm is internalised by women and translated into healthy eating intentions, to the extent that women who identify more highly with their gender group, and those whose female identity is made temporarily salient, also express stronger healthy eating intentions. The association between female identification and healthy eating behaviour, however, appears to be much less consistent, and in most studies the correlation between healthy eating intentions and eating behaviour was poor, even though a variety of measures of behaviour was used.

These findings suggest that actual eating is often not predicted by intentions, but depends on contextual factors, such as being given an opportunity to reinforce the healthy eating goal, or the availability of information about in-group members’ eating behaviour. The results also have implications for health-psychological interventions, in suggesting that people’s response to health-related content (such as social images that may be used in health promotion interventions) may be different depending on their level of group identification. In line with the vicarious licensing effect, individuals who report high levels of group identification might be less likely to respond to interventions aimed at their specific social groups.