Accurate tests of statistical significance for r(WG) and average deviation interrater agreement indexes

The authors demonstrated that the most common statistical significance test used with r(WG)-type interrater agreement indexes in applied psychology, based on the chi-square distribution, is flawed and inaccurate. The chi-square test is shown to be extremely conservative even for modest, standard significance levels (e.g., .05). The authors present an alternative statistical significance test, based on Monte Carlo procedures, that produces the equivalent of an approximate randomization test for the null hypothesis that the actual distribution of responding is rectangular and demonstrate its superiority to the chi-square test. Finally, the authors provide tables of critical values and offer downloadable software to implement the approximate randomization test for r(WG)-type and for average deviation (AD)-type interrater agreement indexes. The implications of these results for studying a broad range of interrater agreement problems in applied psychology are discussed.     

Empirical support for a Type A belief system

Much of the research on Type A behavior has focused on overt manifestations of the pattern; consequently, the underlying psychological dimensions are poorly understood (Matthews, 1982). Price (1982) has proposed an elaborate, but empirically unvalidated, model suggesting that a specific set of beliefs underlies the Type A behavior pattern. A series of experiments was conducted to test the validity of Price's model and to develop a device for assessing these beliefs. The internal consistency and test-retest reliability of this measure were .94 and .84, respectively. Significant positive correlations were found with traditional measures of Type A behavior as well as various facets of anger and anxiety. Finally, subjects who endorsed a high degree of Type A beliefs set significantly higher performance standards and were less likely to achieve these standards than their low-scoring counterparts. Results are discussed with regard to the theoretical construct of Type A as well as their implications for future research and treatment.This research is an extension of a study presented at the 92nd meeting of the American Psychological Association, Toronto, Canada, August 1984Special thanks to to Drs. James A. Blumenthal, Martin T. Gipson, Abby C. King, and Ellie T. Sturgis for their helpful comments in the preparation of this article.     

Selection of Public Housing Tenants: On the Feasibility of Using an Objective Screening Procedure1

Problems facing public housing authorities can be addressed through two basic alternative perspectives. One assumes that difficulties result from the behavior of certain “problem” tenants and recommends careful screening of applicants. This perspective influenced a housing authority in a small city, which desired a means to select tenants who would be less likely to exhibit problems. Information in the records of the housing authority was analyzed to identify tenant characteristics predictive of problem tenancy. The findings indicated that these problems were not as widespread as believed by management and that tenant characteristics were not strongly predictive of problems. Application of this approach would probably not constitute a significant improvement over present practices and might allow unfair discrimination to occur unwittingly. Instead, the most feasible strategy to reduce the impact of problem tenancy, even within a single locale, seems to be to focus on management practices and to address the more fundamental causes of tenant difficulties.     

Sex differences in psychological burnout in teachers

Sex differences in levels of experienced psychological burnout, and in antecedents and consequences, were examined among teachers and department heads. Male teachers reported greater burnout and less job satisfaction than did female teachers. Although male department heads scored significantly higher on psychological burnout, there were no sex differences on measures of satisfaction and emotional well-being. The findings are explained in terms of sex differences in levels of social support.     

Negative emotional responses to motherhood-related support receipt during pregnancy predict postpartum depressive symptoms

Background and Objectives: Prenatal stress increases risk for postpartum depression. While social support availability may attenuate this risk, little research has examined support receipt during pregnancy, which has been linked to increased distress in other domains. This study assesses the implications of motherhood-related and motherhood-unrelated support receipt for daily distress during pregnancy and tests whether negative responsiveness to motherhood-related support predicts postpartum depression risk. Design and Method: Thirty-one pregnant women were recruited from the community for a 3-wave study (beginning at approximately 26 weeks gestation, 34 weeks gestation, and 4 weeks postpartum). Each wave included a survey of general characteristics (e.g. depressive symptoms) and a two-week diary period measuring mood, stress, and support. Results: A multilevel model analysis suggested that motherhood-related support predicted greater increases in daily distress than motherhood-unrelated support. Follow-up regression analysis showed that those who responded more negatively to motherhood-related support reported higher postpartum depressive symptoms. Conclusions: Although preliminary due to the sample size, the results suggest that how women interpret and respond to support may contribute to postpartum depression risk. Future research should focus on how mothers can more positively construe the support they receive in addition to how to increase the support available to them.     

Evaluation of an Air Force Child Physical Abuse Prevention Project Using the Reliable Change Index

Three hundred and seventy-nine parents at 25 Air Force bases participated in a collaborative child physical abuse prevention project designed by the United States Air Force Family Advocacy Program and Father Flanagan's Boys' Home (Boys Town). These participants were active duty or civilian employee parents and their spouses who completed the Boys Town Common Sense Parenting ® program as part of a comprehensive prevention effort. We hypothesized that participants would report decreased child behavior problems, improved relationships with family members, and reduced risk for child physical abuse after attending parenting sessions. Results supported this hypothesis. The data were also analyzed to determine the frequencies of individual parents who made statistically reliable changes on the dependent measures using Jacobson's Reliable Change Index. Nineteen to 30% of the participants reported significant improvement in the areas assessed. Parents who improved in abuse risk were at a higher level of risk and were less satisfied with family relationships prior to their participation than those who did not change. Implications for prevention of child physical abuse and future directions for research are discussed.     

Workaholism in Organizations: Gender Differences

Although the concept of workaholism inorganizations has received considerable attention, ourunderstanding of it based on research evidence islimited. This results from the absence of both suitabledefinitions and measures of the concept. This study, usingmeasures developed by Spence and Robbins (1992),examines gender differences in a predominantly whitemanagerial sample-in three workaholism components and workaholic job behaviors among managers andprofessionals. Although females and males were found todiffer on many personal and situational demographiccharacters, they were similar on the three workaholism components: work involvement, feeling driven towork, and work enjoyment. Females reported higher levelsof particular workaholic job behaviors (e.g.,perfectionism, job stress) likely to be associated with lower levels of satisfaction andwell-being.     

Parent vs. Self-Report: Contributions Toward Diagnosis of Adolescent Psychopathology

Assessment of psychopathology in adolescents can be complicated due to uncertainties about who should be considered the primary informant. While a multimethod, multiinformant approach to assessment allows for a thorough assessment, it can also result in contradictory findings. The purpose of this study was to use a sample of 121 adolescents to investigate issues of parent-adolescent agreement on behavior rating scales and a structured diagnostic interview. Additionally, this study evaluated whether adolescent report offers unique information to the assessment beyond the parent's report. The sample included both normal controls and adolescents who met criteria for diagnosis of an Internalizing or Externalizing disorder based on the DSM-IV criteria. Agreement between parents and adolescents on the Diagnostic Interview for Children and Adolescents—Revised (DICA-R) was low for ADHD (39.4%) and Externalizing Disorders (41.1%) and moderate for Internalizing Disorders (56.6%). A significant difference was found between the number of ADHD symptoms reported, with parents reporting more symptoms (M = 9.42) than adolescents (M = 8.34). Regression analysis indicated that the adolescent completed Youth Self-report (YSR) contributed unique information beyond the parent completed measures in predicting both the adolescent and the parent interviews for Internalizing Disorders. This finding emphasizes the need to continue to collect self-report information, especially when assessing Internalizing disorders, which inherently contain symptoms indicating high-risk behaviors such as suicide ideation.