SIGNIFICANCE TESTS AND CONFIDENCE INTERVALS FOR THE ADVERSE IMPACT RATIO

The two most common methods for assessing adverse impact, the four-fifths rule and the z -test for independent proportions, often produce discrepant results. These discrepancies are due to the focus on practical versus statistical significance, and on differing operational definitions of adverse impact. In order to provide a more consistent framework for evaluating adverse impact, a new significance test is proposed, which is based on the same effect size as the four-fifths rule. Although this new test was found to have slightly better statistical power under some conditions, both tests have low power under the typical conditions where adverse impact is assessed. An alternative to significance testing would be to report an estimate of the adverse impact ratio along with a confidence interval indicating the degree of precision in the estimate.     

The Education-Contingent Association Between Religiosity and Health: The Differential Effects of Self-Esteem and the Sense of Mastery

Using data from a representative sample of adults in Toronto, Canada, I examine the education-contingent association between religiosity (subjective religiosity and religious attendance) and four health-related outcomes: depression, anxiety, alcohol use, and self-rated health. I also test the extent that two personal resources—the sense of mastery and self-esteem—contribute to those associations. Findings indicate that subjective religiosity and attendance are generally associated with lower levels of depression, anxiety, alcohol use, and poor health. Moreover, although not entirely uniform, subjective religiosity and attendance tend to be associated more negatively with these outcomes among individuals with fewer years of education. While the sense of mastery suppresses the education-contingent influence of religiosity on distress outcomes, self-esteem generally contributes to those patterns. On balance, the suppression effects of mastery are offset by the explanatory effects of self-esteem. These findings elaborate on the well-established association between religiosity and health by illustrating education-contingent effects and potential counterbalancing roles of personal resources in these processes.     

EFFECTS OF ADVERTISED HUMAN RESOURCE MANAGEMENT PRACTICES ON ATTRACTION OF AFRICAN AMERICAN APPLICANTS

Student members of a national organization of African American engineers ( n = 1019) and currently employed African American engineers ( n = 303) responded to a hypothetical job advertisement differing by staffing policy (identity-blind vs. identity-conscious), advertised work characteristics (i.e., individual-based vs. team-based), and compensation system characteristics (pay based on individual performance vs. pay based on work-group performance). Both groups of respondents reported being more likely to apply when the staffing policy was identity conscious (i.e., affirmative action) than when it was identity blind (i.e., equal-employment opportunity). However, only the student sample reported being more likely to apply when the advertisement described team-based work instead of individual-based work. Both groups reacted negatively to the combination of individual-based work and group-performance based pay systems.     

An investigation of relational turbulence and depressive symptoms in newly married women

This study uses the relational turbulence model to examine depressive symptoms in newly married women. It reports data from a pre‐/posttest study in which 152 women completed an online survey during their engagement and 6 months after their wedding. Nearly 12% of the women reported reliable and clinically meaningful increases in depressive symptoms after their wedding. Using panel data analysis, this study found that increases in self uncertainty, partner uncertainty, and relationship uncertainty as well as in partner interference were associated with increases in depressive symptoms. This novel test of the relational turbulence model examined the impact of change in (rather than level of) relational turbulence, and the findings extend previous work by demonstrating how mechanisms of relational turbulence impact mental health.     

Microlevel Structuration in Computer-Supported Group Decision Making

The theory of Structuration focuses on how actions by members of social collectives create the structures that enable and constrain future action. Most previous research on Structuration in groups and organizations used qualitative case studies because Structuration is quite complex. This study introduces a method for the study of Structuration in larger samples of groups or organizations. A category system for the identification of structuring moves is described, along with several methods for characterizing short- and long-term trends in Structuration processes. The method is used to study how groups incorporate a computerized group decision support system (GDSS) into their decision processes. GDSSs are part of a class of new computer and communication technologies designed to help groups improve their meetings and make better decisions. Because GDSSs must be used by a group rather than an individual, social processes are critical in determining their effects on group outcomes. GDSSs influence groups by structuring their activities, and structurational processes are critical mediators of the impacts of GDSSs on group decision making. The study focused on the effects of restrictiveness of GDSS technology on Structuration processes and, in turn, on the relation of Structuration to a key group outcome, consensus change. The results suggest that two major types of structuring processes occur in computer-supported groups and that the nature of structuration is related to degree of consensus change.     

Decision Development in Computer-Assisted Group Decision Making

New computer technologies to aid group communication and decision making are becoming increasingly widespread. This study analyzes how one such technology, a group decision support system (GDSS), affected how group decisions developed over time. The study contrasted decision paths in groups using the GDSS with groups using the same procedural structures incorporated in the GDSS manually and with groups using no procedural structures. A flexible phase mapping method was employed to map group decision paths. The resulting set of seven decision paths varied in both sequence and number of decision phases. An optimal matching procedure was used to compute similarity measures among the 40 paths, and cluster analysis and multidimensional scaling were used to generate an empirical taxonomy of decision paths. Results indicated that the nature of decision paths varied both across the three conditions and within conditions. The decision path types were also related to three outcome variables: consensus change, perceived decision quality, and decision scheme satisfaction. Results indicated that those decision paths that most resembled logical normative sequences had superior outcomes to those that did not.     

The Search for Reliable Generalizations About Messages A Comparison of Research Strategies

Two ways of conducting the search for generalizations about messages are considered: Morley's (this volume) proposal that single-message research designs be used, with subsequent meta-analytic summaries, and Jackson and Jacobs's (1983) proposal that multiple-message designs be used, with messages treated as a random factor in the statistical analysis. Jackson and Jacobs's approach is shown to provide a more dependable, efficient, and practical means for gathering the requisite evidence for dependable generalizations. The charge that multiple-message designs suffer from irreparable problems of experimenter bias is refuted. The treatment of messages as a random factor is defended as statistically appropriate and as clearly preferable to the statistical alternatives.