John K. Butler

A behavioral decision theory approach to modeling dyadic trust in superiors and subordinates

The relative importances of five determinants of dyadic trust (integrity, competence, consistency, loyalty, and openness) were investigated from two perspectives: trust in superiors and trust in subordinates. Two experiments were designed from research on behavioral decision theory. The participants, 78 undergraduate management students, responded to cues that described hypothetical superiors (Exp. 1) and subordinates (Exp. 2). Responses indicated the amount of trust held in each of 32 superiors and 32 subordinates. Integrity, competence, and consistency were stronger than loyalty or openness as determinants of trust in ones superiors and in ones subordinates. There were no differences in the importance of any of the determinants of trust in ones subordinates versus ones superiors. Method considerations are discussed.

Trust Expectations, Information Sharing, Climate of Trust, and Negotiation Effectiveness and Efficiency:

Trust expectations, information quantity shared, the climate of trust, and negotiated outcomes were investigated. Initial trust expectation was manipulated, and participants provided data about the quantity of information shared, the climate of trust, and the outcomes of the negotiations. The 324 participants were practicing managers who conducted a negotiating role play with win-win potential. Information sharing followed from initial trust expectations, and information sharing did not fully mediate the relationship between expectations and climate of trust. Negotiating effectiveness was associated with the quantity of information shared but not with trust. Procedural inefficiency (time to discover the solution) decreased as information quantity increased, but procedural inefficiency was not related to trust. One measure of outcome inefficiency (complexity of the agreement) decreased as trust increased, but another measure of outcome inefficiency (monetary cost of the agreement) was not related to trust. Unexpectedly, buyers’ monetary cost increased as information increased, suggesting that sellers might profit from information exchange and, indirectly, from trust. Contributions and limitations of the study are discussed in addition to implications of trust in negotiations by practicing managers.

Measuring the relative importances of social responsibility components: A decision modeling approach

In this study, a decision modeling approach is used to measure the relative importances of four social responsibility components. When given information concerning the economic, legal, ethical and philanthropic activities of 16 hypothetical organizations, 159 junior and senior management students judged the social responsibility of these firms. The study used two types of analysis: first, a within-subject regression, then a between-subject ANOVA. Results showed ethical behavior to be most important in judging social responsibility; legal behavior was second, discretionary behavior third, and economic behavior was least important. In addition, all but one rater consistently applied the social responsibility components. The implications of these results and suggestions for future research are discussed.

Extrinsic Reward Valences and Productivity of Business Faculty: A within-and between-Subjects Decision Modeling Experiment

The valences of six extrinsic rewards that can accrue to business faculty members were investigated. A decision modeling experiment was designed with 16 different hypothetical reward situations to elicit effort decisions from 52 respondents. Using instrumentality theory, valences were represented by within-subject regression coefficients, obtained by regressing effort decisions on a set of orthogonal cues. Exploratory analyses yielded significant effects of reward type for the valences within all rank and tenure categories and for the valences of tenure and recognition across rank and tenure categories. Significant correlations were found between research productivity and the valences of mobility and promotion. Interactions with rank were found for the correlations of research productivity with the valences of money and mobility. The findings were consistent with need theories and support conceptual connections between valences and needs.

Validity of Students' Teaching Evaluation Scores: The Wimberly-Faulkner-Moxley Questionnaire

We present evidence for the validity of scores on a published instrument for student evaluations of teaching performance. We replicated the instruments factor structure very closely, yielding five factors virtually identical to those of a previous study that used two different populations. For construct validation, we developed a preliminary theoretical model that predicts different effects of three sets of variables on student ratings of teachers. Student expectations about teachers had noteworthy effects on all dimensions of teaching effectiveness. Class size and teacher experience had much weaker effects on the ratings.

MALE NEGOTIATORS: CHIVALRY OR MACHISMO OR BOTH? '

This study investigated the behavior of male negotiators as a function of the sex of their opponents and observers having no stake in the outcomes. Participants were 409 graduate and undergraduate students who engaged in role played negotiation. Analyses were conducted for only the all-male dyads. These males were rated as more dominating and offered to pay more when observed by women (n = 29) than by men (n = 64). Previously, dominating behavior had been interpreted as “chivalry” when male negotiators had stake-holding constituents. Since our observers were disinterested, our results (stated above) suggest that “machismo” might be a better interpretation than “chivalry” for such behavior. Although we found no evidence of obliging behavior by men with female opponents (which could be defined as chivalry), we did find dominating behavior by men with female constituents (which could also be defined as chivalry), but only when these men had female opponents.

A Vector Model for Describing and Comparing Profiles

This paper develops and tests an algorithm for computing indices for scoring and comparing multidimensional psychometric profiles. A Euclidean distance, d, derived from the Pythagorean theorem, is generally preferred over other indices of profile similarity. However, d assumes orthogonal profile dimensions. A model, which relaxes the orthogonality assumption, is offered. The model calculates a distance vector that comprises elevation, scatter and shape to describe a profile. It also operationalizes a new concept, the angular orientation of a profile. In addition, it calculates two indices of profile similarity: a distance vector and a measure of angular alignment. A brief example follows the model.

Validation of the Job Desirability Exercise: A Decision-Modeling Experiment for Measuring the Need for Socialized Power

A new behavioral measure of the need for socialized power was developed and tested. No previous attempt has been made to measure this construct. Participants were 267 graduate and undergraduate business majors. The new instrument was found to possess adequate internal consistency and test-retest reliability, and to avoid social desirability response bias. Correlations with other variables and confirmation of six hypotheses supported construct validity. These results are discussed in light of further validation of the instrument.

A Program for Computing an Index of Scale Unidimensionality

After briefly explaining an index of homogeneity for the items of a measurement scale, this paper describes a program that computes the index. Written in SAS, the program converts an output data set created by a factor analysis to two matrices: (a) the absolute values of the interitem correlations and (b) the item commonalities. These matrices are used to calculate the index.

EFFECTS OF CUE ORDER IN A DECISION-MODELING INSTRUMENT

Two experiments, using two separate samples and two methods of analysis (Hotellings T2 and hierarchial regression), ruled out the cue-order threat to the internal validity of a decision-modeling instrument. Implications for measurement are discussed. Caution is suggested in generalizing to other populations and other instruments, despite the apparent robustness of the method of decision modeling.