690 Journal of Applied Social Psychology, 2006, 36, 3, pp. 690–707. r 2006 Copyright the Authors Journal compilation r 2006 Blackwell Publishing, Inc. Helping Behavior as a Subtle Measure of Discrimination Against Lesbians and Gay Men: German Data and a Comparison Across Countries 1 UTE GABRIEL 2 University of Bern Bern, Switzerland RAINER BANSE University of York York, Great Britain To unobtrusively assess attitudes toward lesbians and gay men, the wrong-number technique was used in a field experiment in Germany. The results are compared to studies using the same paradigm in Switzerland, Great Britain, and the United States. This approach gives a realistic picture of intercultural differences in social behavior against lesbians and gay men. Across studies, the results indicated that homosexuals are less likely to receive help than are heterosexuals. The variation of this effect between countries closely corresponded to the ranking of attitudes to- ward homosexuality assessed in survey studies. Contrary to survey studies, however, women showed only marginally less negative attitudes toward gay per- sons than men, when actual helping behavior was used as an attitude index. There is a growing body of sociological and psychological literature dealing with the attitudes of heterosexuals toward gay persons or homo- sexual behavior (Baker & Fishbein, 1998; Fernald, 1995; Herek, 2000; Kite & Whitley, 1998; LaMar & Kite, 1998; Schope & Eliason, 2000). Survey as well as laboratory research has been conducted to study explicit opinions and evaluations, and their relation to other variables. One important de- terminant of attitudes toward lesbians and gay men has been identified in personality variables such as authoritarianism, religiosity, and sex stereo- types. 1 The authors are grateful to the German Telecom for their generous help and technical support for this research. We thank Meike Herting, Sebastian Korinth, Ina Parrhysius, Nor- man Radtke, as well as Susanne Schulze for running the German field experiment; Melanie C. Steffens and Ulrich Wagner for their helpful comments; and Iain Glen for critical comments and stylistic corrections. 2 Correspondence concerning this article should be addressed to Ute Gabriel, Universita¨t Bern, Institut fu¨ r Psychologie, Unitobler, Muesmattstr. 45, CH-3012 Bern, Switzerland. E-mail: ute.gabriel@psy.unibe.ch