George W. Bohrnstedt

Robustness in Regression Analysis

This research was funded, in part, by National Institutes of Health grant No. 5-TO1-GM-01526. An earlier version of this paper was presented to a symposium, Methodology in Sociology, jointly sponsored by the Methodology Section of the American Sociological Association and Loyola University, Chicago, Illinois, June 1011, 1970. The authors are especially grateful to G. William Walster for his helpful comments on the earlier version of the paper.

Social Science Methodology: The Past Twenty-Five Years

Social science methodology has virtually exploded in the past 25 years. While the growth in methodological developments has been roughly exponential for the social sciences in general, it is clear from reading the six articles that make up this special issue that the various social science disciplines occupy very different segments along the growth curve. Most would certainly agree that the methodologies employed by economists are more advanced than those of the other social science disciplinespsychology is probably next in terms of its methodological innovations. The two disciplines that appear to have the least developed methodologies are anthropology and history. Sociology and political science fall between. While interest in the development of social science methodology has permeated the various social science disciplines in some cases for well

Evidence for extending the circumplex model of personality trait language to self-reported moods.

We show that self-ratings by 141 undergraduates on 17 items form a circumplex that closely agrees with Conte and Plutchiks (1981) circumplex of personality traits. The order of items on a circumplex, generated by self-ratings following a here-and-now instruction set, scarcely differs from that of a circumplex representing the same ratings averaged over a 2-week period. The configuration becomes more elliptical as one moves from the state-like measures (here-and-now ratings) to pure trait measures (averages), but measurement bias is shown to be the most probable cause of the distortion. Allowing for the limitations of small samples of items and raters, the findings show that the circumplex model represents a valid theoretical construct and does not arise from measurement error or from an illusory implicit personality theory.

Common, Specific, and Error Variance Components of Factor Models Estimation with Longitudinal Data

In the classic factor-analysis model, the total variance of an item is decomposed into common, specific, and random error components. Since with cross-sectional data it is not possible to estimate the specific variance component, specific and random error variance are summed to the items uniqueness. This procedure imposes a downward bias to item reliability estimates, however, and results in correlated item uniqueness in longitudinal models. In this article, we describe a method for estimating common, specific, and random error variance with longitudinal data. An empirical example demonstrates the specification and testing of hypotheses regarding the temporal invariance of common, specific, and error components, and the practical utility of our approach. This example makes clear the limitations of combining specific and random error components. The assumption that all unique variance is random error is shown to be untenable.

Are the good beautiful or the beautiful good? The relationship between children's perceptions of ability and perceptions of physical attractiveness.

A catalytic composite comprising a refractory inorganic oxide-supported Group VIB and Group VIII metal component is disclosed. The catalytic composite is characterized by a method of preparation which comprises coextruding a precursor compound of the Group VIB metal component, a precursor compound of the Group VIII metal component, and a finely divided refractory inorganic oxide admixed with hydrogen peroxide and sufficient water to provide an extrudable dough. The coextruded product is subsequently dried and calcined to yield an improved catalytic composite, particularly with respect to the hydrodesulfurization of residual fuel oils.

The Effects of Recalled Childhood and Adolescent Relationships Compared to Current Role Performances on Young Adults' Affective Functioning

Addressing the issue of continuity and change in affective development, we undertake an evaluation of the effects of childhood and adolescent peer and parental relationships compared with the effects of current role performances on depressed affect and self-esteem in young adults. We formulate five hypotheses which (a) relate depressed affect and self-esteem to current role performances; (b) predict a significant long-term effect of poorly formed interpersonal relationships with parents and peers on depressed affect; and (c) suggest that an authoritative-reciprocalfamily environment fosters positive affective development, but that an authoritarian-autocratic environment retards affective development. An empirical illustration using the Beck Depression Inventory and the Rosenberg Self-Esteem Scale scores of 103 undergraduates suggests that (a) current role performances affect current levels of depressed affect and self-esteem; (b) the number of close friendships and parenting styles in adolescence have strong direct effects on later levels of depressed affect; and (c) parenting styles during childhood have significant indirect effects on later levels of depressed affect. Thus self-esteem appears to be perfectly plastic, alterable by the quality of current role performances; whereas depressed affect can be long lasting, reflecting the quality of relationships with parents and peers formed during childhood and adolescence.

External Validity and Evaluation Research

This paper delimits and explicates threats to external validity particularly problematic in evaluation research. Five categories of factors are discussed: selection effects, measurement effects, confounded treatment effects, situational effects, and effects due to differential mortality. The paper focuses on pointing up specific ways in which each of the factors threaten generalizability and possible solutions to the methodological problems presented.

Some Limitations On Generalizability From Social Psychological Experiments

The experimental model has severe limitations that often are ignored by experimenters. The limitations are likely to account for the low level of external validity of experiments. Placing the model in the context of requirements for measurement emphasizes the limitations and suggests alternatives for improvements in the future.

THE WORK COMPONENTS STUDY (WCS): A REVISED SET OF MEASURES FOR WORK MOTIVATION

The present study, through reanalysis of earlier data and the analysis of data from a new sample, reports on the refinement of the Work Components Study (WCS). An earlier study yielded six scores but the present one suggests that one of them should be broken into two, yielding seven scores. It is shown that the internal consistency of these scores is satisfactory and that the scores are relatively independent of each other. Finally, the utility of the scores is evaluated by showing their relationships with measures of personality, ability, and aspirations.

Student-Teacher Racial Match and Its Association With Black Student Achievement: An Exploration Using Multilevel Structural Equation Modeling:

This study examines student-teacher “racial match” for its association with Black student achievement. Multilevel structural equation modeling was used to analyze 2013 National Assessment for Educational Progress (NAEP) Grade 4 Reading Assessment data to examine interactions of teacher race and student race in their associations with student achievement (N = 165,410 students, 23,710 teachers). Effects on Level 1 random slope parameters suggest that a match of Black male students with Black teachers is associated with higher reading scores for this group, as is a match of Black female students with a Black or Hispanic teacher. Level 2 effects suggest that greater classroom composition of Black males is associated with lower reading achievement. Moderation by gender and implications for educational policy are discussed.