The present research employed a correlational-design laboratory study to examine the initialinteraction component of a developmental model of Graen's leader-member exchange theory of leadership proposed by Dienesch and Liden (1986). The influence of ability, liking, and the three upward-influence tactics of ingratiation, assertiveness, and rationality on the quality of leadermember exchange was examined both from the perspective of group leaders and members. The 188 subjects were assigned to groups of 4 (a leader and 3 members). From the leader's perspective, liking for members and ability of members were the variables most consistently related to quality of leader-member exchange. From the member's perspective, all variables except self-assessed ability were related to quality of leader-member exchange. The differences in perspective between leaders and members are discussed.

LaPiere's (1934) seminal research into the attitude-behavior relationship has been commonly misinterpreted as pointing to a discrepancy between attitudes and behavior. In fact, the actual discrepancy uncovered was between true attitudes -the tendency to act in a certain way -and that which is measured by an attitude questionnaire. LaPiere's primary concern was to point out the danger of assuming that questionnaire-assessed "attitudes" lead to actual behavior in specific situations. In the present paper we theoretically and methodologically critique LaPiere's classic study. , as w ell as interpretations by selected succeeding scholars, are examined, and several general conclusions about the contribution of the study to current attitude-behavior research are offered.LaPiere's initial interest in the relationship between attitudes and behavior was sparked by his earlier research into t he relationship between symbolic (speech) and non-symbolic (action) behavior. From this research, LaPiere (1928) concluded that there is no "necessary" association between what people say they will do and how they actually behave in a specific situation. LaPiere (1932) noted, however, "to prove that there is no necessary relationship does not prove that such a relationship may not exist" (p. 231). His curiosity about this relationship provided the impetus for his classic study.

No abstract