What we say versus why we do: a basic two-factor model of helping decisions in hypothetical and actual helping situations

Abstract

Social psychology’s long history of studying helping behavior has yielded longstanding theories of helping based on field research using classic confederate protocols (e.g., Gaertner & Dovidio, 1977; Dovidio & Gaertner, 1981; Gross, Wallston, & PIlliavin, 1975; Emswiller et al., 1971; West & Brown, 1975). Evolutionary and cognitive psychological theory have added to these findings by introducing ideas such as likelihood of reciprocation (Trivers, 1971) and cost/benefit analysis into the decision to provide help. In recent years, however, the shift toward utilizing survey methods to measure behavior has raised questions as to how accurately modern social psychological research captures human behavior (Baumeister et al., 2007). The current study collected field research on helping behaviors, of varied costs and benefits to both the helper and recipient, followed by an identical hypothetical survey utilizing the same variables on the same college population. Additionally, trials in both studies included researchers dressed to cue either ingroup or outgroup university affiliation as a manipulation of expectations of reciprocation. In both the field and hypothetical studies, there was little to no evidence of an effect of reciprocation likelihood and mixed evidence for high cost behaviors eliciting lower probabilities of helping. More pressingly, the qualitative pattern of helping differed between the hypothetical and field studies, indicating a difference in decision making processes in each type of survey. This finding has significant implications for the use of hypothetical survey methods to capture behavior in social psychological studies of helping.