A Multi‐Source,Multi‐Study Investigation of Job Performance Prediction by Political Skill

Political skill is a social effectiveness construct with a demonstrated capacity to predict job performance. However, because performance prediction research in this area to date has made exclusive use of self‐reports of political skill, and due to frequent distrust of self‐ratings of constructs in important personnel decisions, there is a need to investigate how multiple alternative sources of political skill and job performance measures relate, thus raising both theoretical and methodological issues. In three studies, employing a triadic data collection methodology, and utilising both cross‐sectional and longitudinal designs, this research tested the hypotheses that employee political skill, measured from the perspective of employees' assessor A, will positively predict job performance rated by assessor B (i.e. Hypothesis 1a), and vice versa, that employee political skill measured by assessor B will predict job performance ratings measured by assessor A (i.e. Hypothesis 1b).     

Attributions restore consistency in bargaining with liked/disliked partners

The experiment deals with the impact of self-esteem and liking for the partner on the attribution of agreement and deadlock in bargaining. Fifty-eight male and 70 female students played the Harsanyi-Selten bargaining game with incomplete information eight times, allegedly each time with a randomly selected partner. In fact in four games a computer program simulated the partner. Combining an experimental variation of liking (liking—disliking), own costs (low, high), partner costs (low, high) the experiment followed a 2×2×2 repeated measures design. As predicted by a path model from balance theory (a) failure (deadlock) was attributed more to the partner and less to self than success (agreement), (b) success was attributed more to the liked than the disliked partner, whereas failure was attributed more to the disliked than the liked partner.     

Genetic Counseling Practices in Germany: A Comparison Between East German and West German Geneticists

As part of an international study of ethics and genetics, we present a comparison between survey responses of 43 East German and 212 West German geneticists to anonymous questionnaires. Both groups indicated that the experience of the Third Reich has impacted the genetics profession in Germany today. East German geneticists reported more directive counseling practices after prenatal diagnosis for 10 of 26 conditions than those from West Germany. When asked to give their personal opinions about pregnancy termination, East Germans were more accepting of abortion than their West German colleagues for 7 of 24 fetal indications. In addition, there were significant differences between the two sample groups for 8 questions on the perception of disability and society. Discussions with German geneticists suggest that, while both groups were affected by Germany's experience of Nazism, different abortion laws, political systems, and ideas about the doctor-patient relationship in former East and West Germany may account for discrepancies in reported genetic counseling practices and in attitudes toward abortion and disability.     

Formal paraphasias: A single case study

A case study is reported of an aphasic patient, RB, who showed frequent form-related whole-word substitutions in oral naming, writing to dictation, and reading aloud. In both written language tasks, the abstractness of the targets influenced the number of formal errors. In oral naming, a high proportion of formal paraphasias was related to the intended words in both form and meaning. A comparison between targets and formal paraphasias indicated a high agreement both in word class, number of syllables, stress pattern, and in basic (stressed) vowels. The agreement in consonants (including word-initial consonants), however, was low. It is argued that RB's formal substitutions are not caused solely by errors of lexical selection but that semantic, lexical, and segmental factors contribute to the error outcome.     

Pseudo maximum likelihood estimation and a test for misspecification in mean and covariance structure models

Using the theory of pseudo maximum likelihood estimation the asymptotic covariance matrix of maximum likelihood estimates for mean and covariance structure models is given for the case where the variables are not multivariate normal. This asymptotic covariance matrix is consistently estimated without the computation of the empirical fourth order moment matrix. Using quasi-maximum likelihood theory a Hausman misspecification test is developed. This test is sensitive to misspecification caused by errors that are correlated with the independent variables. This misspecification cannot be detected by the test statistics currently used in covariance structure analysis.For helpful comments on a previous draft of the paper we are indebted to Kenneth A. Bollen, Ulrich L. Küsters, Michael E. Sobel and the anonymous reviewers of Psychometrika. For partial research support, the first author wishes to thank the Department of Sociology at the University of Arizona, where he was a visiting professor during the fall semester 1987.