Lisa L. Harlow

The Transtheoretical Model of Change and HIV Prevention: A Review

The transtheoretical model of health behavior change is described and supporting empirical work is presented that reviews the central constructs of the model: the stages of change, processes of change, decisional balance, confidence, and temptation. Model-based applications to a broad range of problem behaviors are summarized. Applications to human immunodeficiency virus (HIV) prevention behavior changes are highlighted for each variable. Finally, several questions about the area of sexual behavior change to reduce risk of HIV exposure are explored and future research ideas are described within the context of this model.

The comparative efficacy of imputation methods for missing data in structural equation modeling

Abstract Missing data is a problem that permeates much of the research being done today. Traditional techniques for replacing missing values may have serious limitations. Recent developments in computing allow more sophisticated techniques to be used. This paper compares the efficacy of five current, and promising, methods that can be used to deal with missing data. This efficacy will be judged by examining the percent of bias in estimating parameters. The focus of this paper is on structural equation modeling (SEM), a popular statistical technique, which subsumes many of the traditional statistical procedures. To make the comparison, this paper examines a full structural equation model that is generated by simulation in accord with previous research. The five techniques used for comparison are expectation maximization (EM), full information maximum likelihood (FIML), mean substitution (Mean), multiple imputation (MI), and regression imputation (Regression). All of these techniques, other than FIML, impute missing data and result in a complete dataset that can be used by researchers for other research. FIML, on the other hand, can still estimate the parameters of the model. The study involves two levels of sample size (100 and 500) and seven levels of incomplete data (2%, 4%, 8%, 12%, 16%, 24%, and 32% missing completely at random). After extensive bootstrapping and simulation, the results indicate that FIML is a superior method in the estimation of most different types of parameters in a SEM format. Furthermore, MI is found to be superior in the estimation of standard errors. Multiple imputation (MI) also is an excellent estimator, with the exception of datasets with over 24% missing information. Considering the fact that FIML is a direct method and does not actually impute the missing data, whereas MI does, and can yield a complete set of data for the researcher to analyze, we conclude that MI, because of its theoretical and distributional underpinnings, is probably most promising for future applications in this field.

An Illustration of a Longitudinal Cross-Lagged Design for Larger Structural Equation Models

A model of HIV behavior risk is tested, using a fully cross-lagged, longitudinal design, to illustrate the analysis of larger (i.e., 3 or more variables across 3 or more time points) structural equation models. The constructs in this integrated model were (a) Perceived Risk for AIDS, (b) Decisional Balance (relative weighing of benefits and barriers of condom use), (c) Self-Efficacy for Condom Use, (d) Lifetime Number of Sex Partners, and (d) Behavior Risk. Data were analyzed from 527 women living in a New England community who completed lifestyle surveys at 3 time points. Results of the cross-lagged analysis indicated a reciprocal, positive relation between Self-Efficacy for Condom Use and Decisional Balance. Behavior Risk for HIV was associated with higher Perceived Risk, lower Self-Efficacy for Condom Use, and lower Decisional Balance. The model showed excellent fit to the data and explained substantial variance in Time 3 measures of model constructs. Interpretations, as well as strengths and limitations of the analysis, are presented.

Causation issues in structural equation modeling research

As the use of structural equation modeling (SEM) has increased, confusion has grown concerning the correct use of and the conclusions that can be legitimately drawn from these methodologies. It appears that much of the controversy surrounding SEM is related to the degree of certainty with which causal statements can be drawn from these procedures. SEM is discussed in relation to the conditions necessary for providing causal evidence. Both the weaknesses and the strengths of SEM are examined. Although structural modeling cannot ensure that necessary causal conditions have been met, it is argued that SEM methods may offer the potential for tentative causal inferences to be drawn when used with carefully specified and controlled designs. Keeping in mind that no statistical methodology can in and of itself determine causality, specific guidelines are suggested to help researchers approach a potential for providing causal evidence with SEM procedures.

A structural model of the relationships among stage of disease, age, coping, and psychological adjustment in women with breast cancer

The present study used structural equation modeling to examine the relationships among disease stage (i.e. Stage II versus Stage IV), age, coping style, and psychological adjustment in 100 women diagnosed with breast cancer. Five separate models were examined: a full model, a mediational model, a demographic‐disease model, a coping style model, and a regression model The analyses revealed that the present data best fit the mediational model in which age and stage of disease were not directly associated with psychological adjustment but, instead, were mediated by coping style (χ2(25)=45.776, AASR=0.05, CFI=0.94). The mediational model accounted for 56% of the variance in psychological adjustment. In particular, the model showed that younger women and women with an earlier disease stage used greater levels of the coping strategy characterized as a fighting spirit and lower levels of the coping strategies characterized as hopelessness/helplessness, anxious preoccupation, and fatalism which, in turn, were related to better psychological adjustment. Overall, these findings may offer an explanation for the conflicting findings regarding the relationship between age, stage of disease, and psychological adjustment to breast cancer by illustrating that coping strategies may be an essential mediating factor; in turn, a mediating model of psychological adaptation may offer useful information for clinicians as they implement interventions designed to improve patients coping efforts.

Sexual communication in relationships: when words speak louder than actions.

Sexual communication for expressing sexual desires and gathering HIV risk information were examined as interpersonal constructs related to HIV risk reduction. Community women (n = 816) with at least one heterosexual HIV risk factor (79% Euro-American, 86% some college) completed surveys assessing assertive communication with a sexual partner, HIV risk, and demographic, sexual history, interpersonal negative, and cognitive/attitudinal constructs which formed a hierarchical predictor model. In relationship context comparisons, women with a known partner HIV risk responded more negatively on most measures. Multiple regressions suggested communication is part of an overall cognitive/attitudinal approach to HIV risk, although specific predictors differ by type of communication and partner risk level. Implications for interventions considering relational context, interpersonal power, and counteracting silence are discussed.

Condom Negotiation in Heterosexually Active Men and Women: Development and Validation of a Condom Influence Strategy Questionnaire

The ability to negotiate condom use with a partner is a skill that sexually active men and women must have in order to avoid sexually transmitted diseases including HIV. Despite this fact, there is no psychometrically valid instrument in the literature to measure condom influence strategies. This investigation reports on the development and initial validation of the condom influence strategy questionnaire (CISQ). Exploratory and confirmatory analyses revealed and confirmed six influence strategies used by heterosexually active men and women to negotiate condom use. These CISQ subscales accounted for variance in safer sexual variables including sexual assertiveness, self-efficacy, and partner communication. Further, those who endorsed CISQ subscales were more likely to have intentions to use condoms consistently and to use condoms. Gender differences in subscales favoring women as the ones most likely to use influence strategies also emerged. Implications of these results as well as future directions for research are discussed.

Self-Perception Variables That Mediate AIDS-Preventive Behavior in College Students

Three psychological variables--self-efficacy, control and meaning, and perceived risk--were tested in a structural model predicting AIDS-preventive behavior. Results revealed a good model fit, indicating that these psychological variables did play a role in mediating AIDS-preventive behavior in college students. A multivariate analysis of variance and individual analyses of variance conducted for men and women also revealed gender differences on individual items of self-efficacy, perceived risk, and AIDS-preventive behavior. This study underscores the importance of identifying and assessing the psychological determinants of AIDS-preventive behavior.

Correlates of Adjustment Among Cancer Survivors

Abstract This study examined demographic, clinical, and psycho-social correlates of adjustment among a sample of cancer survivors. Analyses concerning demographic and clinical variables indicated that being married, having a high income and level of education, and a positive perception of ones health was related to higher levels of adjustment; female survivors and survivors of breast cancer (versus prostate cancer) also reported higher levels of sexual adjustment. Analyses concerning psychosocial predictors of adjustment indicated that survivors who reported higher levels of social support, optimism, and meaning in life, and lower levels of avoidant-type coping exhibited better adjustment. A prediction model of adjustment indicated strong empirical support for a model depicting higher psychosocial adjustment as a function of higher levels of social support and meaning in life and lower levels of avoidant-type coping behaviors. Overall, the findings offer important information for understanding variables associated with adaptation to a cancer diagnosis and provide support for the usefulness of clinical services for survivors that provide social support, minimize the use of avoidant-type coping, and help them attain a sense of meaning from their illness.

An Introduction to Latent Variable Growth Curve Modeling: Concepts, Issues, and Applications (2nd ed.)

This book approaches latent growth modeling (LGM) from a practical, handson perspective. LGM, a set of techniques for analyzing change over time, is applied within a structural equation modeling (SEM) framework. LGMs are presented as flexible methods that can be applied to a range of longitudinal datasets. The authors’ stated goal is to provide a reference guide or primer. The intended audience ranges from clinical, developmental, and social psychologists to educational, sociological, and quantitative researchers. The book provides a useful and readable resource to researchers interested in modeling change using latent factors that reflect interindividual variability in initial status and growth. Next, we provide short synopses of each chapter followed by some concluding remarks. The text presents several clear examples to provide guidance on how to set up and interpret various LGMs. Many of the examples draw on the authors’ extensive experience of modeling substance abuse data. In this second edition the authors have updated existing chapters, added three chapters on special topics (i.e., growth mixture modeling, piecewise and pooled time series LGMs, and LGMs with categorical variables), and developed a new overall organization. Background and descriptive information about LGMs are explored in the first four chapters. The next three chapters discuss approaches to multiple group techniques. Several chapters that cover areas of particular interest to longitudinal researchers, such as missing data, power estimation, and interaction effects,