LOG-LINEAR MODELING IN PERSONNEL RESEARCH

Log-linear modeling is a discrete multivariate statistical technique that is designed specifically for analyzing data when both the independent and dependent variables are categorical or nominal. The purpose of this paper is to demonstrate the utility of this technique in personnel research. The paper (a) discusses behavioral areas of application, (b) compares log-linear modeling with chi-square and regression analysis, (c) presents the basic principles and hypotheses of log-linear modeling, and (d) shows how the technique is used.     

On the Nature of Marital Relationships: Measurable Differences in Spontaneous Agreement

The Spontaneous Agreement (SA) scores on a lengthened form of the Ferreira-Winter Questionnaire of 242 normal and 177 abnormal couples were compared in a cross-sectional study, taking into account the duration of their marriage. For those couples married a short time, SA did not differ for the two groups. The overall SA was higher for normals, and the longer the normal couples were married, the higher their SA. SA did not increase with duration of marriage for abnormal couples. Both groups had higher SA scores than synthetically matched couples.     

INTERVIEW NOTES AND VALIDITY

Three studies were conducted investigating the effects of notetaking behavior and the content of notes on validity in a selection interview. Overall, results indicate that when notetaking was voluntary, notetakers made more valid ratings than non-notetakers. But, when note-taking was manipulated non-notetakers made ratings that were just as valid as raters told to take behavioral notes, and actually more valid than those raters instructed to take general notes. With regard to the content of notes, behavioral-type notes were found to contribute to greater validity. In addition, in the first and third study, procedural type notes resulted in lower validity. These results may have important implications for interviewer training for notetaking.     

Equality ratio equity,general linear equity and framing effects in judgments of allocation divisions

It is argued here that there is not necessarily a contradiction between the general linear and equal ratio approaches to equity, and the two may be integrated to make more precise predictions. It was hypothesized that, (a) equity is best represented by a limited equal ratio rule; (b) the more a linear distribution systematically deviates from equal input/outcome ratios, the more unfair it will be judged, but (c) subjects will prefer a distribution that accords with a general linear formula in certain cases where an equal ratio distribution of available outcomes would actually infringe a preeminent or established equal ratio distribution. Four studies are presented, the results of which seem broadly to support these ideas. However, the results also suggest that equity judgments can be strongly influenced by framing effects.