MEASUREMENT ERROR IN RESEARCH ON HUMAN RESOURCES AND FIRM PERFORMANCE: ADDITIONAL DATA AND SUGGESTIONS FOR FUTURE RESEARCH

Gerhart and colleagues (2000) and Huselid and Becker (2000) recently debated the presence and implications of measurement error in measures of human resource practices. This paper presents data from 3 more studies, 1 of large organizations from different industries at the corporate level, 1 from commercial banks, and the other of autonomous business units at the level of the job. Results of all 3 studies provide additional evidence that single respondent measures of HR practices contain large amounts of measurement error. Implications for future research into the HR firm performance relationship are discussed.     

UNEQUAL ATTENDANCE: THE RELATIONSHIPS BETWEEN RACE, ORGANIZATIONAL DIVERSITY CUES, AND ABSENTEEISM

Although prior evidence has demonstrated racial differences in employee absenteeism, no existing research explains this phenomenon. The present study examined the roles of 2 diversity cues related to workplace support—perceived organizational value of diversity and supervisor–subordinate racial/ethnic similarity—in explicating this demographic difference among 659 Black, White, and Hispanic employees of U.S. companies. Blacks reported significantly more absences than their White counterparts, but this difference was significantly more pronounced when employees believed their organizations placed little value on diversity. Moreover, in a form of expectancy violation, the Black–White difference was significant only when employees had racially similar supervisors (and thus would expect their companies to value diversity) and perceived that the organization placed little value on diversity.     

Changes in Religious Giving Reflect Changes in Involvement: Age and Cohort Effects in Religious Giving, Secular Giving, and Attendance

We present two patterns over time in religious giving, secular giving, and religious service attendance. The first pattern describes the prewar cohort (born 1924–1938) as they aged between middle adulthood (ages 35–49) and their senior years (ages 62–76). The second pattern compares the baby boom cohort (born 1951–1965) in middle adulthood to the middle adulthood of the prewar cohort. We present patterns for all families as well as separately for Catholic and Protestant families using data from three sources. The prewar cohort increased their religious giving and attendance as they aged, but—compared to the prewar cohort in middle adulthood—baby boomers give less than expected to religion and attend less. Baby boomer giving is noticeably less than expected and attendance noticeably lower among Catholic boomers, but less so among Protestant boomers. We argue that together these patterns are evidence that changes in religious giving reflect changes in religious involvement.     

Religion and Substance Abuse Treatment: Individual and Program Effects

The relationship between personal religiousness and substance abuse treatment outcomes has emerged as an important issue in the public health arena. Using the “moral community” perspective, a conceptual framework developed by Stark, Kent, and Doyle (1982) to analyze the contextual effects of religion, we explore the degree to which religion influences two drug treatment outcome measures—critical retention and commitment to treatment. The data are derived from the Drug Abuse Treatment Outcome Studies (DATOS), a national study of 10,010 clients enrolled in 70 drug treatment programs. Three research questions were addressed: (1) What is the relationship between an individual's level of religiosity and retention in treatment and commitment to treatment? (2) How does the ecological context of treatment programs shape the individual‐level relationships? (3) To what extent are program practices and characteristics directly linked to outcome level? The findings are supportive of the literature that shows a weak to moderate relationship between religiosity and treatment outcomes. However, the findings did not show strong support for the “moral community” hypothesis. Although there was a wide variation in the size of the individual‐level religiosity–treatment correlations, the variation could not be conclusively attributed to the overall religious emphasis of the programs. The findings suggest that further research is needed in order to understand fully the role of religion in substance abuse treatment.     

Conflicting Intuitions about Causality1

In this article I examine five kinds of conflicting intuitions about the nature of causality. The viewpoint is that of a probabilistic theory of causality, which I think is the right general framework for examining causal questions. It is not the purpose of this article to defend the general thesis in any depth but many of the particular points I make are meant to offer new lines of defense of such a probabilistic theory. To provide a conceptual framework for the analysis, I review briefly the more systematic aspects of the sort of probabilistic theory of causality I advocate. I first define the three notions of prima facie cause, spurious cause, and genuine cause. The technical details are worked out in an earlier monograph (Suppes 1970) and are not repeated.     

Obtaining results through modeling

This article presents practical applications of the theoretical notion of modeling. Use of live and symbolic models is emphasized, along with the modeling techniques of role reversal and role identification.     

A Dilemma of Late Life Memory

ABSTRACT Unexpected but vivid and compelling memories are a wide-spread experience in late life. The experience has often been described in literature, and in recent years has been the object of extensive gerontological research under the label ‘life review’. Such memories often include a reversal of judgment about a past act, relationship, event, etc. What earlier was judged to be so is, in the retrospect of late life, judged not to have been so after all. This presents a question: which judgment—the earlier or the later—has better epistemological credentials in such cases? Some obvious possible answers are considered and rejected. It would seem that the issue is not resolvable on epistemological grounds. A parallel dilemma seems to appear in other dimensions of experience. An example from aesthetic experience is briefly considered.     

THE MORAL REQUIREMENT IN THEISTIC AND SECULAR ETHICS

One of the central tasks of meta‐ethical inquiry is to accommodate the common‐sense assumptions deeply embedded in our moral discourse. A comparison of the potential of secular and theistic ethics shows that, in the end, theists have a greater facility in achieving this accommodation task; it is easier to appreciate the action‐guiding authority and binding nature of morality in a theistic rather than in a secular context. Theistic ethics has a further advantage in being able to accommodate not only this essential conceptual feature of morality, but also the existence of moral requirements and their source of normativity.