Effects of Salient Multiple Identities on Women's Performance Under Mathematics Stereotype Threat

Previous research on affective extremity and social identity complexity suggested that women's mathematics stereotype threat might be alleviated by reminding individual women of their multiple roles and identities, most of which would presumably be unrelated and thus impervious to negative stereotypes regarding math performance. To test this hypothesis, we primed the relevant stereotype and then asked men and women college students to draw self-concept maps with many or few nodes. When they drew no maps or maps with few nodes, highly math-identified women scored significantly worse than highly math-identified men on a subsequent Graduate Record Examination-like math test, but when they drew maps with many nodes, they scored as well as those men. Theoretical and practical implications of the results are discussed.     

When organizational justice and the self-concept meet: Consequences for the organization and its members

In two studies, we examined the joint effects of employed participants’ self-concept levels and perceptions of fairness on organizational attitudes and citizenship behavior intentions. We examined the effects of chronic self-concept activation in Study 1, whereas we primed the working self-concept in Study 2. Combining the results of both studies, we found support for our hypotheses that particular self-concept levels and organizational justice dimensions interact to predict various work-related outcomes. Specifically, we observed interactions between the relational self-concept and interactional justice, and between the collective self-concept and procedural justice, such that the justice–outcome relationships were stronger for those experiencing higher activation on the relevant self-concept level. Thus, as hypothesized, justice information is weighted differently depending on the particular level of self-concept that is active. In addition, interesting direct effects of employees’ self-concepts were also observed. We discuss the implications of these findings and the importance of considering the self-concept in conjunction with organizational justice.     

An Exploration of Cognitive Demands in Group Interaction as a Moderator of Information Processing Variables in Perceptions of Leadership

In the application of information processing (IP) models in organizational settings, potential boundary or moderator variables are sometimes overlooked. We investigated whether the impact of important IP variables in the leadership perception literature was affected by a potentially important boundary variable: cognitive demands extraneous to impression formation. In contrast to past research, both quantity and quality (prototypicality) of behavior affected leadership perceptions in both low and high information load conditions. This result implies that prototype-related processing may be automatic enough to influence perceptions of leadership in actual organizational settings where cognitive demand is often high. Further, quantity of verbal behavior had a significant impact on causal attributions for level of group task performance and on perceptions of control of the groups'activities, suggesting that this variable may have important implications for inferences about a person's influence on work group processes and outcomes. The significance of these findings for the issues of leader influence and for the measurement of leader behavior is discussed.     

Monthly fluctuation in task concentration in female college students.

50 university women with consistent and rhythmic menstrual cycles were given Golden's version of the Stroop color-word test to evaluate their concentration accuracy. 50 male undergraduates were also tested as a control sample. The participants took the Stroop test every seven days for four consecutive weeks to allow examination of whether change in the color and word naming activity occurred. A split-plot analysis of the variance of the scores gave a significant difference for the sequence in which the tests were taken, the measures recited during the testing, and the sex of the respondent. When the values for the women were ordered to correspond to stages of their menstrual cycle, the over-all mean accuracy rose slowly to midcycle, then dropped as menses approached. This supports the hypothesis that the women's concentration was at the lowest level during the latter part of the monthly cycle. Women scored significantly higher on the color and word naming task on the Stroop test than men.     

Estimation of latent ability and item parameters when there are omitted responses

Omitted items cannot properly be treated as wrong when estimating ability and item parameters. A convenient method for utilizing the information provided by omissions is presented. Theoretical and empirical justifications are presented for the estimates obtained by the new method.Research reported in this paper has been supported by grant GB-32781X from National Science Foundation.     

Effects of participative vs assigned goals and feedback in a multitrial task

A dynamic multitrial perspective on goal setting was adopted in order to investigate the effects of both goals (assigned vs participative) and goal-discrepant performance feedback on subsequent goal commitment and performance. Eighty subjects were initially assigned to either an assigned or participative goal condition, and performed a multitrial task with all subjects receiving goal discrepant performance feedback following the first task trial. Assigned rather than participative goal setting led to higher goal commitment, and large goal/feedback discrepancies led to greater reductions in subsequent goals. Because of these goal changes, the ability of initial goals to predict performance decreased over trials. However, when goal commitment, which reflects revised goals, was also considered, the ability to predict performance actually increased over trials. A moderating effect for need achievement was obtained such that higher-need achievers were more goal committed and performed better under participative goal setting than low-need achievers. Discussion focused on the practical and theoretical importance of a dynamic goal-setting perspective in explaining and predicting responses to goals and feedback systems.     

Sampling fluctuations resulting from the sampling of test items

Sampling fluctuations resulting from the sampling of test items rather than of examinees are discussed. It is shown that the Kuder-Richardson reliability coefficients actually are measures of this type of sampling fluctuation. Formulas for certain standard errors are derived; in particular, a simple formula is given for the standard error of measurement of an individual examinee's score. A common misapplication of the Wilks-Votaw criterion for parallel tests is pointed out. It is shown that the Kuder-Richardson formula-21 reliability coefficient should be used instead of the formula-20 coefficient in certain common practical situations.Most of the work reported here was carried out under contract with the Office of Naval Research. The writer is indebted to Professor S. S. Wilks, who has checked over certain critical portions of a draft of this paper.     

Reversing roles: a cognitive strategy for undoing memory deficits associated with token status

Tested whether having tokens (Ts) adopt the role of judge reduces cognitive deficits; examined several hypotheses to explain these deficits. In 3 experiments, Ss were asked to remember as many as possible of opinions exchanged in a group interaction with 3 actors. Experiment 1 demonstrated that judging majority members helped gender Ts improve their memory and ruled out self-denigration as a mediator of token deficits. Experiment 2 indicated that judging others was effective regardless of whether the others were said to know about it or not, ruling out insulation from evaluative scrutiny as a viable mediator for the judge role. Experiment 3 suggested the judge role restores completely the Ts, cognitive capacities and ruled out heightened responsibility as an explanation for the improved memory of judges. This work suggests that Ts may perform better if they can restructure cognitively their social environments.     

Estimating the imputed social cost of errors of measurement

If a loss function is available specifying the social cost of an error of measurement in the score on a unidimensional test, an asymptotic method, based on item response theory, is developed for optimal test design for a specified target population of examinees. Since in the real world such loss functions are not available, it is more useful to reverse this process; thus a method is developed for finding the loss function for which a given test is an optimally designed test for the target population. An illustrative application is presented for one operational test.This work was supported in part by contract N00014-80-C-0402, project designation NR 150-453 between the Office of Naval Research and Educational Testing Service. Reproduction in whole or in part is permitted for any purpose of the United States Government.