535 British Journal of Psychology (2011), 102, 535–545 C  2011 The British Psychological Society The British Psychological Society www.wileyonlinelibrary.com Testing multiple means of self-affirmation Christopher J. Armitage ∗ and Richard Rowe University of Sheffield, UK The effects of self-affirmation manipulations are not yet well understood because several different affirmation manipulations have been tested using multiple dependent variables. The aim of the present research was to establish whether global self-feelings and self- esteem, or interpersonal positive feelings are affected by affirming the self. Experiment 1 showed that completing a kindness questionnaire led to adolescent girls reporting more positive interpersonal feelings, but no greater self-feelings or self-esteem than completing an opinions (control) questionnaire. Experiment 2 showed again that self- affirmation did not affect self-esteem, and that there were few differences between writing an essay about a cherished value, writing an essay about kindness, or completing a questionnaire about kindness in boosting positive interpersonal feelings. The findings are discussed in relation to self-affirmation theory and the possible practical implications of self-affirmation for aiding the delivery of social and health messages. According to self-affirmation theory, fundamentally, people are motivated to maintain self-integrity by preserving a positive, moral, and adaptive self-image (Steele, 1988). The implication is that any threat to the self (e.g., a health risk message) will undermine self-worth unless a person’s self-image can be bolstered. According to Steele’s (1988) theory, self-image can be maintained by self-affirming in one domain (e.g., recalling one’s acts of kindness) even if one is being threatened in another domain (e.g., health) because people can defend their global sense of self-worth, rather than (for example) the threat directly attributable to health risks. There is a growing body of evidence showing that self-affirmation manipulations are effective in changing people’s attitudes and intentions, increasing receptiveness to threatening information and ultimately changing behaviour (e.g., Harris & Epton, 2009; McQueen & Klein, 2006). For example, Sherman, Nelson, and Steele (2000, Experiment 2) found that affirmed students were more likely to buy condoms at the end of an experimental session; and Armitage (2010) found that, over the course of a month, self-affirmation reduced alcohol consumption by more than 1 unit 1 per day compared with a 0.03 unit per day decrease in the control group ( p < .01). ∗ Correspondence should be addressed to Chris Armitage, Centre for Research in Social Attitudes, Department of Psychology, University of Sheffield, Western Bank, Sheffield S10 2TP, UK (e-mail: c.j.armitage@sheffield.ac.uk). 1 A unit is defined as 8 g/10 ml of pure alcohol, which is equivalent to half a pint/300 ml of ordinary strength beer, a 125-ml glass of wine at 9% strength or one measure/25 ml of spirits. DOI:10.1111/j.2044-8295.2010.02014.x