A contribution to the study of the metric and Euclidean structures of dissimilarities

This paper is concerned with the geometric properties of dissimilarity coefficients defined on finite sets and especially with their Euclidean nature. We present several particular transformations which preserve Euclideanarity and we complete, through the study of a one-parameter family, the current knowledge of the metric and Euclidean structure of coefficients based on binary data. These results are directly deduced from two theorems which prove the positive semi-definite status of some quadratic forms which play a large role in some definitions of dissimilarity commonly used.The authors wish to thank B. Fichet for his helpful suggestions, the associate Editor and an anonymous reviewer for comments and highly constructive criticisms on earlier drafts of the paper.     

A multiple-level examination of job activities and employee outcomes

Prior research has investigated relationships between job activities and employee outcomes, but multiple levels of analysis considerations often have been overlooked. In a study of a police organization, three theoretical perspectives were examined, and multiple levels of analysis were considered conceptually and empirically, to better specify and test associations between activities and outcomes. Results from Within and Between Analysis (WABA) suggested that relationships between law enforcement and resource control job activitiesand employee outcomes of attention, latitude, and involvement operated at multiple levels of analysis. The associations were based on differences between individuals, work groups, and job-based collectives (job families). Implications for future research and practice are discussed.     

Initial examination of a multidimensional model of trauma-related guilt: Applications to combat veterans and battered women

Guilt is conceptualized as a multidimensional construct consisting of negative affect and a set of interrelated cognitions. Guilt magnitude is thought to be a function of the magnitudes of six variables posited as primary components of guilt: a negative event, distress, perceptions of responsibility, lack of justification, wrongdoing, and false beliefs about preoutcome knowledge. The model was tested with samples of Vietnam veterans and battered women. Participants rated their reactions to and perceived roles in trauma-related events. Among Vietnam veterans, distress ratings were highly correlated with guilt severity. Cognitive guilt-component variables were significantly correlated with guilt in both groups. In multiple regression, guilt components accounted for 61% of variance in veterans' guilt and 44% of variance in women's guilt. Among veterans, distress ratings were highly correlated with measures of PTSD and depression. In both groups, cognitive guilt-component variables were positively correlated with psychopathology. Results support the view that beliefs about one's role in trauma are important factors in posttrauma adjustment.     

An examination of linkages between personal characteristics and dimensions of transformational leadership

Recent research in organizational behavior has found that transformational leadership can engender improved employee work outcomes in many types of organizations. Personal characteristics of transformational leaders, however, have not been thoroughly investigated. This paper reports the findings of a study that explored relationships between personal characteristics and dimensions of transformational leadership using sales managers as the population of interest. Findings suggest that a reconceptualization of the proposed linkages may be necessary. Potential explanations for the results obtained are presented along with managerial and research implications.The authors gratefully acknowledge the host company for its participation in the study.     

The analytical solution of the additive constant problem

Ifd is a measure of dissimilarity on a finite set withn elements, the smallest positive constantc* such thatd +c has an euclidean representation for allc ≥c* is shown to be the largest eigen-value of a matrix of size 2n.     

Unmasking affiliative behavior among juvenile Midas cichlids (Cichlasoma citrinellum)

Adults of the Midas cichlid (Cichlasoma citrinellum) are predominantly normal-colored (N) morphs, although 8% in nature are gold (G) morphs. They tend to mate assortatively by color; that behavior may be influenced by prior experience with parents or siblings or by their own color. A previous experiment indicated that juveniles respond according to a rule that they avoid and behave defensively toward G juveniles and that they consort with and behave aggressively toward N juveniles (Barlow & Siri, 1987). The fear-provoking effect of G may have interfered with more subtle aspects of choice because the 4 treatment fish in each group reacted through the glass barrier to the approach of the subject. In the present experiment such feedback was eliminated by using one-way mirrors. N juveniles reared by normal parents (N X N) differentiated between N and G treatment fish only in that they attacked significantly more toward G; thus their behavior was contrary to that predicted by the fear-provoking effect of gold. The notable effect among juveniles reared by G parents (G X G) was that N spent more time with N and avoided G treatment fish, in keeping with the fear-provoking effect of gold. The results are discussed in terms of the differences in experiences between juveniles reared N X N, all of which were themselves N, and those reared G X G, consisting of both color morphs. Despite the differences in the two experiments, the main conclusion remains in place: Affiliative responses of juveniles, as measured in these tests, do not predict choice of mate among adults.     

The relation of field dependence to signal detection while imaging

Twelve field-dependent (FD) and 12 field-independent (FI) males were administered 260 trials of a signal detection task in four phases. Signal type (auditory, visual, or no signal) was crossed with four sets of imaging treatments (pure visual representations, pure sounds without visual counterparts, and two sets representing sounds with visual counterparts). Detection of a signal in isolation was 90% accurate, as was the detection of no signal while imaging. Cross-modality imaging interfered with detection of the stimulus but to a significantly lesser extent than did same-modality imaging. In all detection/imaging treatments, FIs were significantly more sensitive to the signal than were FDs; however, FDs and FIs did not differ in accuracy of detecting isolated signals. The results are discussed in terms of the commonalities shared by perception and imaging as they appear in the Perky effect and cognitive style.