The attitude cube—A three-dimensional model of situational factors in IS adoption and their impact on the attitude–behavior relationship

Since the introduction of the TAM by Davis et al. (Manage. Sci. 35 (1989) 982–1003), user attitudes have been considered an important determinant of IS adoption. Their impact on behavior is rooted in behavioral frameworks such as the theory of reasoned action (Fishbein and Ajzen, Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research, Addison-Wesley, 1975) and the theory of planned behavior (Ajzen, Organ. Behav. Hum. Decis. Processes. 50 (1991) 179–211). However, in IS adoption models, the predictive power of attitudes for behavior as an endogenous variable has proven strong in some circumstances and weak in others (Bagozzi and Yi, Soc. Psychol. Q. 52 (1989) 266–279; Taylor and Todd, MIS Q. 19 (1995) 561–570). These inconsistencies resulted in an exclusion of the construct attitude from further modified versions of the TAM and other models explaining individual IS adoption and use. To better understand the prevalence of inconsistent attitude–behavior relationships in IS adoption models, this research follows social psychological research practices focusing on situational factors prone to result in insignificant attitude–behavior relationships. To test for the impact of these factors on the attitude–behavior relationship, the article raises the level of analysis to a higher level of abstraction. A scientometric literature review and a meta-analysis of 14 top IS journals from 1989 to 2014 finds that three out of four situational factors positively influence the attitude–behavior relationship in the IS fields: voluntariness, technology type, and adoption context. These factors constitute “the attitude cube”, which provides conceptual guidance for researchers in assessing the conditions under which the attitude–behavior relationship is likely to be weak or strong.