W. Todd Abraham

Advances in testing the statistical significance of mediation effects

P. A. Frazier, A. P. Tix, and K. E. Barron (2004) highlighted a normal theory method popularized by R. M. Baron and D. A. Kenny (1986) for testing the statistical significance of indirect effects (i.e., mediator variables) in multiple regression contexts. However, simulation studies suggest that this method lacks statistical power relative to some other approaches. The authors describe an alternative developed by P. E. Shrout and N. Bolger (2002) based on bootstrap resampling methods. An example and step-by-step guide for performing bootstrap mediation analyses are provided. The test of joint significance is also briefly described as an alternative to both the normal theory and bootstrap methods. The relative advantages and disadvantages of each approach in terms of precision in estimating confidence intervals of indirect effects, Type I error, and Type II error are discussed.

Maladaptive Perfectionism as a Mediator and Moderator Between Adult Attachment and Depressive Mood

This study examined maladaptive perfectionism (concern over mistakes, doubts about one’s ability to accomplish tasks, and failure to meet high standards) as both a mediator and a moderator between adult attachment (anxiety and avoidance) and depressive mood (depression and hopelessness). Survey data were collected from 310 undergraduates and analyzed using structural equation modeling (SEM) methods. Results indicated that maladaptive perfectionism partially mediated the relationship between attachment anxiety and depressive mood and fully mediated the relationship between attachment avoidance and depressive mood. Bootstrap methods were used to assess the magnitude of the indirect effects. Significant moderator effects were also found with SEM methods. The association between attachment anxiety and depressive mood was stronger as perfectionism increased. Perfectionism was not a significant moderator for attachment avoidance and depressive mood.

Neighborhood context and financial strain as predictors of marital interaction and marital quality in African American couples

Demographic characteristics, family financial strain, neighborhood-level economic disadvantage, and state of residence were tested as predictors of observed warmth, hostility, and self-reported marital quality. Participants were 202 married African American couples who resided in a range of neighborhood contexts. Neighborhood-level economic disadvantage predicted lower warmth during marital interactions, as did residence in the rural south. Consistent with the family stress model (e.g., Conger & Elder, 1994), family financial strain predicted lower perceived marital quality. Unexpectedly, neighborhood-level economic disadvantage predicted higher marital quality. Social comparison processes and degree of exposure to racially based discrimination are considered as explanations for this unexpected result. The importance of context in relationship outcomes is highlighted.

Missing data: a review of current methods and applications in epidemiological research

Purpose of review Researchers inevitably confront missing data. In cross-sectional studies, nonresponse to specific items causes item-level missing data. Longitudinal studies pose a greater likelihood of item nonresponse and introduce unit nonresponse when data for an individual are missing because that person was not available for assessment. The need to adequately deal with missing data remains, regardless of whether missing data result from item nonresponse, participant attrition, or sporadic availability of respondents. The wealth of missing data techniques available to researchers often produces uncertainty regarding which to use. Our purpose is to discuss the applicability of general methods for dealing with missing data and to review current advances associated with specific missing data techniques. Recent findings Traditional missing data methods such as complete case analysis often produce bias and inaccurate conclusions. Similar problems extend to single imputation techniques commonly thought of as improvements over complete case methods. Research demonstrates that procedures such as multiple imputation, which incorporate uncertainty into estimates for missing data, often provide significant improvements over traditional methods. Summary Recent work suggests that multiple imputation and specific modeling techniques offer general methods for dealing with missing data that perform well across many types of missing data situations. In addition, advances in desktop computers and the development of user-friendly software make these techniques accessible to researchers in all fields. Future research will undoubtedly result in further refinements and extensions of these techniques, making them applicable to difficult but common situations in which missing data arise.

Gendered trajectories of work control and health outcomes in the middle years: a perspective from the rural Midwest.

Objectives: The objective was to investigate whether increasing health heterogeneity during the middle years is attributed, at least in part, to the influence of varying levels of, and changes in, work control of the current midlife cohort. Method: The study used four waves of data collected from 372 employed rural women and 320 employed men of the midlife cohort during a 10-year period. Variables included self-reported work control, stressful nonwork life events, and mental and physical health. The analyses used latent growth curve modeling. Results: The results partially supported the hypothesized pathways. For middle-aged men, work control directly influences health outcomes, whereas for middle-aged women work control indirectly influences health outcomes through the occurrence of stressful life events. Discussion: Through understanding these processes, work can be better designed to promote positive health outcomes, minimize negative health outcomes, and allow for better formulation and more effective implementation of health promotional programs.

Early family adversity, youth depressive symptom trajectories, and young adult socioeconomic attainment: A latent trajectory class analysis

Abstract Using prospective data from 485 adolescents over a 10-year period, this chapter applies group-based latent class trajectory analysis in the area of life course stress. This analysis included several stages. First, group-based trajectory class analyses identified heterogeneous developmental patterns of depressive symptoms for groups of youth. Depressive symptom trajectories for the majority of youth followed normative patterns. However, depressive symptom trajectories for other youth showed two non-normative patterns characterized by recovery from extremely high symptom levels in adolescence and continued escalation of depressive symptoms throughout adolescence and young adulthood. Second, the analysis examined antecedents and consequences of depressive symptom trajectories during adolescence and early adulthood. The findings showed that having highly educated parents protects youth from non-normative development characterized by continued symptom escalation, whereas experiences with negative family life events and parental rejection increase the vulnerability of youth falling into a non-normative group characterized by extremely high initial levels of depressive symptoms that decline over time. Finally, linking identified group membership to subsequent young adult outcomes, the analysis provided information about the continuity of group-based processes into young adulthood. Young adult social status attainments of youth in escalating and recovering groups were lower than those in the normative group. Identification of members of these heterogeneous groups and antecedents, concurrent processes, and consequences linked to group membership provide a potentially useful prognostic tool for early intervention efforts and policy formation.

Experiential self-focus writing as a facilitator of processing an interpersonal hurt.

OBJECTIVES This study examined the effects of experiential self-focus writing on changes in psychological outcomes (i.e., unforgiveness and negative affect) after an interpersonal hurt and the buffering effects of experiential self-focus writing on the association between anger rumination and these psychological outcomes. DESIGN A sample of 182 college students who had experienced interpersonal hurt were randomly assigned to either the experiential self-focus writing condition, in which participants wrote about their feelings and experiences related to the hurt, or to a control writing condition in which they wrote about a recent neutral event. RESULTS Latent growth curve analyses indicated that changes in unforgiveness over time did not differ between the experiential self-focus writing and the control writing conditions. However, relative to the control writing condition, negative affect decreased faster during writing and increased more slowly at follow-ups in the experiential self-focus writing condition. CONCLUSIONS The results supported the hypothesis that negative affect resulting from an interpersonal hurt would significantly decrease over time among participants in the experiential self-focus writing group compared with the control group. Implications of experiential self-focus writing for interpersonal hurt and directions for future studies are discussed.

Neighborhood racial discrimination and the development of major depression.

This study examined the impact of neighborhood racial discrimination on the development of major depressive disorder (MDD) in a sample of African American women. Participants were 499 women from Georgia and Iowa with no history of MDD who were followed for 9 to 11 years. Several neighborhood characteristics (community social disorder, community cohesion, and community racism) and individual characteristics (negative life events, financial strain, personal outlook, religious involvement, relationship quality, negative affectivity, and individual experiences of racism) were employed as predictors of whether or not the women met criteria for MDD during this period of time. In a multilevel logistic regression analysis, neighborhood-level discrimination as well as individual-level variables including the number of negative life events, financial strain, and negative affectivity were found to be significant predictors of developing MDD. Analyses of cross-level interactions indicated that the effects of neighborhood-level discrimination were moderated by the quality of individuals’ relationships, such that better relationships with others served to lessen the effect of neighborhood discrimination on depression. Implications of these findings for understanding the negative effects of racial discrimination are discussed.

Family health and income: A two-sample replication.

The current study examined psychological and family health predictors of change over time in household income, using data from longitudinal studies of African American (N = 889, 93.5% female) and Mexican origin (N = 674, 100% female) families. Participants self-reported their household income, as well as their emotional, personality, and cognitive resources. Participant behavioral and physical resources were coded from observed family interactions. Although income did not predict change in any personal resources, all five classes of personal resources (i.e., emotional, personality, cognitive, behavioral, physical) predicted change in income across a 10-year span (Study 1) and a 6-year span (Study 2). Income is potentially caused by these personal resources, or both income and these personal resources share a common cause. The dominant approach of assuming income causes personal and family health needs stronger support.

Intergenerational continuity in attitudes: A latent variable family fixed-effects approach.

Attitudes are associated with behavior. Adolescents raised by parents who endorse particular attitudes are relatively more likely to endorse those same attitudes. The present study addresses conditions that would moderate intergenerational continuity in attitudes across 6 domains: authoritative parenting, conventional life goals, gender egalitarianism, deviancy, abortion, and sexual permissiveness. Hypothesized moderators included the attitudes of the other parent, and adolescent sex. Data come from a 2-generation study of a cohort of 451 adolescents (52% female), a close-aged sibling, and their parents. After employing a novel specification in which family fixed-effect models partitioned out variation at the between-family level, hypotheses were tested on the within-family variance. Unlike typical family fixed-effect models, this specification accounted for measurement error. Intergenerational continuity was not significant (deviancy), negative (sexual permissiveness), and conditional on the attitudes of the coparent (authoritative parenting, conventional life goals, and gender egalitarianism). Adolescent age, sex, and conscientiousness were accounted for in all analyses.