Synthesized clustering: A method for amalgamating alternative clustering bases with differential weighting of variables

In the application of clustering methods to real world data sets, two problems frequently arise: (a) how can the various contributory variables in a specific battery be weighted so as to enhance some cluster structure that may be present, and (b) how can various alternative batteries be combined to produce a single clustering that best incorporates each contributory set. A new method is proposed (SYNCLUS, SYNthesizedCLUStering) for dealing with these two problems.We wish to thank Anne Freeny and Deborah Art for their computer assistance, and Ed Fowlkes for his helpful technical discussion. We would also like to acknowledge the insightful and helpful comments from the editor and reviewers.     

Mood and the mundane: relations between daily life events and self-reported mood

Daily mood ratings and corresponding diary entries were studied to determine relations between common events and two independent mood factors--Positive Affect (PA) and Negative Affect (NA)--in a sample of 18 young adults over a 3-month period. In an extension of findings from earlier interindividual studies, PA (enthusiastic, delighted vs. sluggish, drowsy) was found to be associated with a wide range of daily events, whereas fewer correlations were found between these events and NA (distressed, nervous, angry vs. calm, relaxed). The relation between high PA and reported social interactions (particularly physically active social events) was especially robust, and its effects were noted repeatedly; NA was unrelated to social activity. As hypothesized, high NA was associated with physical problems; contrary to expectations, low PA also tended to be correlated with health complaints. Overall, the results reaffirm the importance of assessing NA and PA independently and suggest that PA is an interesting and important dimension that deserves more research attention. Theoretical considerations and clinical implications are discussed.     

Social categorization in minority influence: The case of homosexuality

Three experiments were conducted in order to compare the influence of ingroup and outgroup minorities and to assess the role of Zeitgeist perception in minority influence. The results confirmed that ingroup minorities are more influential than outgroup minorities. This overall finding was observed in two different experimental paradigms, using either a small group setting in which subjects interacted with the minority or the simultaneous social influence paradigm in which both influence sources impinge simultaneously (via written information) upon the subjects. These results were supportive of Tajfel's social identity theory while contradicting Kelley's augmenting/discounting principle. Finally, subjects' perception of the Zeitgeist was unrelated to the magnitude of minority influence.     

CHANGES IN THE FACIAL PROMINENCE OF WOMEN AND MEN OVER THE LAST DECADE

Two studies measured the facial prominence of women and men in popular images over the last decade. The first study used 1,200 photo-graphs from Time and Newsweek for the years 1970 and 1980. The results indicate that overall, men have received greater facial prominence than women, and facial prominence has increased over the decade. Only for Newsweek did women show a significantly greater increase in facial prominence than men. The second study used 640 photographs from Good Housekeeping and Ms. for the years 1974 and 1984. The results show that overall, men have received greater facial prominence than women, and facial prominence has risen over the decade, although not by much. In Ms. , women but not men gained; in Good Housekeeping , women gained slightly, but not as much as men. The results are consistent with earlier reports of an asymmetry in representation of men's and women's faces and suggest that this asymmetry is slowly diminishing.     

Providing help and desired relationship type as determinants of changes in moods and self-evaluations

Observed in 2 initial studies was converging evidence that helping improves the helpers' moods and self-evaluations. In these studies Ss induced to help showed improved moods and self-evaluations relative to Ss not given an opportunity to help. A 3rd study examined the moderating effects of desired relationship type on reactions to having helped. In this study Ss were led to desire either a communal or an exchange relationship with another. They then helped the other or were not allowed to help. Among Ss led to desire a communal relationship, but not among those led to desire an exchange relationship, helping was associated with greater improvements in moods than not helping. Helping tended to improve self-evaluations regardless of desired relationship type. However, this effect reached statistical significance only among subjects led to desire a communal relationship.     

What is the mechanism by which suicide attempts predispose to later suicide attempts? A mathematical model

A state dependence model of serial behavior suggests that each occurrence increases the subsequent likelihood of that behavior being repeated. A heterogeneity model, by contrast, suggests that the likelihood of a behavior occurring is predetermined, and uninfluenced by intervening occurrences. We have applied the random-effects probit model of Gibbons and Bock (1987) to examine the fit of the state dependence and heterogeneity models to longitudinal data on suicide attempts by 928 patients with affective disorder. Heterogeneity but not state dependence was required to model these data. The findings suggest that when considering patients with moderate to severe major affective disorder, the clinician should not interpret the absence of any recent suicide attempts to mean that the patient is at relatively low risk for attempting suicide in the future. An implication of the heterogeneity model is that suicide attempts made many years ago may have equal value to recent attempts when estimating an individual's "predisposition" to nonlethal attempts in the future.