Patrick A. Knight

The Impact of Perceived Group Success-Failure on Motivational Beliefs and Attitudes: A Causal Model

Two models describing the theoretical relationship among the constructs of perceived group success-failure, individual and collective levels of job-specific efficacy and outcome expectancy, satisfaction, and organizational commitment were developed. Model 1 proposed that the causal relationship between group success-failure and subsequent attitudinal variables (satisfaction and organizational commitment) is completely mediated by efficacy and outcome expectancy variables. The second model proposed a direct causal link from perceived group success-failure to the attitudinal variables and omitted questionable causal paths from personal efficacy and collective outcome expectancy to the attitudinal variables. The competing models were subjected to structural equation path model analysis. The second model produced an adequate fit between theory and data. Implications for organizations are discussed.

How do people with different attachment styles balance work and family? A personality perspective on work-family linkage.

This study explored whether different models of work-family relationship were possible for individuals with different attachment styles. A mail survey was conducted using employees (N = 481) at a midwestern university in the United States. Results suggested that (a) individuals with a preoccupied attachment pattern were more likely to experience negative spillover from the family/home to the work domain than those with a secure or dismissing style, (b) securely attached individuals experienced positive spillover in both work and family domains more than those in the other groups, and (c) preoccupied individuals were much less likely to use a segmentation strategy than the other 3 attachment groups. However, when the conventional job satisfaction life satisfaction relationship was examined, the data provided unique support for the spillover model. Implications of the findings for both attachment and work family relationship literatures are discussed.

Humility revisited: Self-esteem, information search, and policy consistency

Abstract A computer simulation of a managerial task was used to study the relationship between self-esteem and the information search and policy consistency of subjects. As predicted, there were significant negative relationships between self-esteem and both experimenting (number of policy changes), and information search (requests for performance feedback and other information). The implications of these results for the practice of experimenting decision making are discussed.

Assimilation and contrast effects in performance ratings: Effects of rating the previous performance on rating subsequent performance.

This study examined the effects of rating previous performance on ratings of subsequent performance. Participants (N = 214) reviewed a performance script depicting either good or poor performance and then reviewed and rated another script depicting average subsequent performance. Approximately half of the participants rated the previous performance immediately after reviewing it, and the other half did not rate the previous performance. As expected, a Performance Level x Existence of Ratings interaction was found. When they had not rated the previous performance, raters were biased in favor of recalling behaviors consistent with their previous impressions, suggesting an assimilation effect. However, when they had rated the previous performance, raters provided ratings inconsistent with their previous ratings, suggesting a contrast effect. These effects were stronger for ratings from a behavioral expectation scale than for those from an ability rating scale.

Effects of gender differences and selection agent expertise on leader influence and performance evaluations

Abstract The effects of selection agent expertise and various gender factors upon the influence and perceived expertise of male and female leaders, and upon group cohesiveness, were examined. Leaders (male or female) were selected by an agent (male or female, expert or nonexpert) to lead groups working on either a masculine or a feminine task. While the influence of the leaders was unaffected by the manipulations, ratings of expertise and group cohesiveness were affected. Leaders selected by experts were given higher expertise ratings than were those selected by nonexperts, as were leaders working on masculine as opposed to feminine tasks. Also, in the feminine task condition, perceived expertise was higher for female leaders and for leaders chosen by female agents. Cohesiveness ratings were highest for groups with leaders chosen by experts, except for groups with male leaders and masculine tasks, where cohesiveness ratings were higher when the agent was a nonexpert. The implications of these results for the effects of selection agent and task characteristics in leader selection, and issues of criterion selection in leadership research, are discussed.

Heroism versus competence: Competing explanations for the effects of experimenting and consistent management☆

Abstract B. M. Staw and J. Ross ( Journal of Applied Psychology , 1980 , 65, 249–260) present data which suggest that one barrier to experimenting management is an implicit theory of heroic leadership favoring managers who are consistent in the face of failure, The currant research tests this theory of heroism against an implicit theory of competent management. Subjects read a case study describing a manager who was either experimenting or consistent, and either immediately successful or successful only after initial failure (i.e., ultimately successful). The results supported the implicit theory of competent management with the consistent managers, whose first policies were effective, rated significantly higher than the ultimately successful-experimenting manager, whose first two policies failed. The immediately successful-experimenting manager was unexpectedly rated significantly higher than all other managers. The results show that there is not a general bias against experimentation in favor of consistency, and that reactions to managers are based more upon evidence of competence than upon experimenting and consistency per se.

Facing the challenges of the year 2000: Learning to manage multidimensional workers

The “organization man” of the 1950s was characterized by an all-encompassing commitment to the work organization. Today, workers are multidimensional and seek a balance between work and nonwork commitments. Research to date has addressed the capacity for multiple commitments simplistically by examining two simultaneously held commitments, such as work organization--family, work organization--union, and work organization--church. Research is needed that examines peoples simultaneous and multiple commitments to all of the relevant organizations in their lives. We have proceeded from the one-dimensional “organization man” to the dually-committed person; we now need to consider the person with multiple commitments. A model of multiple organizational commitment is presented, and implications and research directions are discussed.