On the fit of models to covariances and methodology to the Bulletin.

It is noted that 7 of the 10 top-cited articles in the Psychological Bulletin deal with methodological topics. One of these is the Bentler-Bonett (1980) article on the assessment of fit in covariance structure models. Some context is provided on the popularity of this article. In addition, a citation study of methodology articles appearing in the Bulletin since 1978 was carried out. It verified that publications in design, evaluation, measurement, and statistics continue to be important to psychological research. Some thoughts are offered on the role of the journal in making developments in these areas more accessible to psychologists.

Comparative fit indexes in structural models.

Normed and nonnormed fit indexes are frequently used as adjuncts to chi-square statistics for evaluating the fit of a structural model. A drawback of existing indexes is that they estimate no known population parameters. A new coefficient is proposed to summarize the relative reduction in the noncentrality parameters of two nested models. Two estimators of the coefficient yield new normed (CFI) and nonnormed (FI) fit indexes. CFI avoids the underestimation of fit often noted in small samples for Bentler and Bonett's (1980) normed fit index (NFI). FI is a linear function of Bentler and Bonett's non-normed fit index (NNFI) that avoids the extreme underestimation and overestimation often found in NNFI. Asymptotically, CFI, FI, NFI, and a new index developed by Bollen are equivalent measures of comparative fit, whereas NNFI measures relative fit by comparing noncentrality per degree of freedom. All of the indexes are generalized to permit use of Wald and Lagrange multiplier statistics. An example illustrates the behavior of these indexes under conditions of correct specification and misspecification. The new fit indexes perform very well at all sample sizes.

Choice of structural model via parsimony: a rationale based on precision.

It is shown that, in large samples, the more parsimonious of two competing nested models yields an estimator of the common parameters that has smaller sampling variance. The use of parsimony as a criterion for choice between two otherwise acceptable models can thus be rationalized on the basis of precision of estimation.

Can test statistics in covariance structure analysis be trusted?

Covariance structure analysis uses chi 2 goodness-of-fit test statistics whose adequacy is not known. Scientific conclusions based on models may be distorted when researchers violate sample size, variate independence, and distributional assumptions. The behavior of 6 test statistics is evaluated with a Monte Carlo confirmatory factor analysis study. The tests performed dramatically differently under 7 distributional conditions at 6 sample sizes. Two normal-theory tests worked well under some conditions but completely broke down under other conditions. A test that permits homogeneous nonzero kurtoses performed variably. A test that permits heterogeneous marginal kurtoses performed better. A distribution-free test performed spectacularly badly in all conditions at all but the largest sample sizes. The Satorra-Bentler scaled test statistic performed best overall.

Use of structural equation modeling to test the construct validity of the SF-36 Health Survey in ten countries: results from the IQOLA Project. International Quality of Life Assessment.

A crucial prerequisite to the use of the SF-36 Health Survey in multinational studies is the reproduction of the conceptual model underlying its scoring and interpretation. Structural equation modeling (SEM) was used to test these aspects of the construct validity of the SF-36 in ten IQOLA countries: Denmark, France, Germany, Italy, the Netherlands, Norway, Spain, Sweden, the United Kingdom, and the United States. Data came from general population surveys fielded to gather normative data. Measurement and structural models developed in the United States were cross-validated in random halves of the sample in each country. SEM analyses supported the eight first-order factor model of health that underlies the scoring of SF-36 scales and two second-order factors that are the basis for summary physical and mental health measures. A single third-order factor was also observed in support of the hypothesis that all responses to the SF-36 are generated by a single, underlying construct--health. In addition, a third second-order factors, interpreted as general well-being, was shown to improve the fit of the model. This model (including eight first-order factors, three second-order factors, and one third-order factor) was cross-validated using a holdout sample within the United States and in each of the nine other countries. These results confirm the hypothesized relationships between SF-36 items and scales and justify their scoring in each country using standard algorithms. Results also suggest that SF-36 scales and summary physical and mental health measures will have similar interpretations across countries. The practical implications of a third second-order SF-36 factor (general well-being) warrant further study.

Cognition and the corpus callosum: verbal fluency, visuospatial ability, and language lateralization related to midsagittal surface areas of callosal subregions.

Normal volunteers (28 women), 20-45 years old, completed tests of visuospatial ability, verbal fluency, and language lateralization, and the midsagittal surface areas of the splenium, isthmus, midregion, and genu of the corpus callosum were measured from inversion recovery magnetic resonance images. Multivariate statistics were used to analyze patterns of correlations. Verbal fluency correlated positively with the area of the splenium and with the area of a posterior callosal factor defined largely by the splenium. The posterior callosum, particularly the splenium, also correlated negatively with language lateralization. There were no other consistent brain-behavior relationships. These results are relevant to understanding factors involved in the development of cognitive characteristics that show sex differences and to understanding the neural basis of language lateralization and verbal abilities.

Psychosocial correlates and predictors of AIDS risk behaviors, abortion, and drug use among a community sample, of young adult women.

Relations among latent constructs of Social Conformity, Sensation Seeking, Polydrug Use, Sexual Experience, Abortion, and Risky AIDS Behaviors were examined among a community sample of women (N = 438, mean age = 25.5 years) using confirmatory factor analysis (CFA) and predictive structural equation models (SEM). In the CFA, Risky AIDS Behavior was strongly related to more Polydrug Use and less Social Conformity and modestly related to Sexual Experience and Abortions. In SEMs, Social Conformity significantly predicted less Risky AIDS Behavior and less Polydrug Use but did not predict Abortions. Prior Sexual Experience predicted more Polydrug Use and Abortions. We conclude that the same psychological processes and predispositions that relate low social conformity to drug use and other unhealthy behaviors also influence AIDS-risk behaviors, even among a community sample of women.

Attitudes and health behavior in diverse populations: drunk driving. Alcohol use, binge eating, marijuana use, and cigarette use.

Five different health behaviors (cigarette use, alcohol use, binge eating, illicit drug use, and drunk driving) were studied prospectively in 5 different groups of subjects. Associations between attitudes toward these behaviors and the behaviors themselves were investigated over at least 2 waves of measurement. Findings revealed that attitudes predicted behavior nonspuriously in 2 instances: alcohol use and marijuana use. Attitudes did not predict drunk driving, binge eating, or smoking behaviors. Past behavior predicted attitude in the domains of binge eating and smoking, but not in the domains of alcohol use, drunk driving, or marijuana use. The results are discussed in terms of several alternative approaches that have implications for interventions that attempt to influence health behavior through attitude change.

Structural equation analyses of clinical subpopulation differences and comparative treatment outcomes: characterizing the daily lives of drug addicts.

The use of structural equation modeling (SEM) is illustrated for comparative treatment outcome research conducted with heterogeneous clinical subpopulations within large multimodality treatment settings. All analyses are accomplished with SEM analogs of more familiar classical multivariate techniques. The effect of the early period of treatment on the daily lives of 486 clients in two drug abuse treatment modalities (methadone maintenance and outpatient counseling) is evaluated. Structured means analysis is used to assess initial differences between modalities on the latent means of 6 latent constructs reflecting daily life. The effect of treatment modality and attrition from the program on daily life latent constructs is evaluated while initial selection differences are statistically controlled. Effect sizes are computed on the basis of SEM parameter estimates. The advantage of SEM over classic multivariate approaches for correcting for selection bias when assessing comparative outcomes is explained.

Cognitive motivation and drug use: a 9-year longitudinal study.

The predictive precedence of expectancy constructs, operationally defined as cognitive motivations, and drug use was investigated over a 9-year period from adolescence to adulthood. Alternative predictions from three different classes of theories of expectancy-behavior relations, including expectancy theory, a Skinnerian approach, and a reciprocal determinism perspective, were evaluated. The results are most consistent with the notion based in expectancy theory that cognitive motivations are nonspurious and possibly functionally autonomous influences on the use and abuse of drugs. More limited support is found for the view that drug use leads to cognitive motivations, as postulated in other theoretical perspectives. Other findings reveal the presence of expectancy generalization processes consistent with Rotter's (1954) expectancy theory, as well as the unique status of cognitive motivations for alcohol as an independent predictor of problem drug use.

On the usefulness of latent variable causal modeling in testing theories of naturally occurring events (including adolescent drug use): a rejoinder to Martin.

This paper considers the arguments raised by Martin (1982) against the methodologies used by Huba, Wingard, and Bentler (1981) and their subsequent conclusions. Several of Martin's criticisms are the result of a misreading of our paper and selective citations, whereas other criticisms were discussed in the original paper and resolved through alternate forms of data analysis. Further analyses are presented to address issues raised by Martin. Martin's arguments against latent variable models are refuted.

Dimensions of subjective female orgasmic responsiveness.

Results of recent psychological research have been at odds with the major physiological evidence on the nature of female orgasmic responsiveness. Although female orgasm appears to be a single process physiologically, subjective reports by women indicate that individual differences in their experience is multidimensional. The basis for the discrepant findings is analyzed via two studies of reported orgasmic experiences in women. One sample of 115 sexually experienced women yielded data that replicated the finding that coital and masturbatory orgasmic responsiveness can be statistically distinguished and independently assessed, thus indicating that sample selection associated with prior research could not explain the result. A separate sample of 101 sexually experienced women were given an instrument that assessed coital responsiveness but differentiated masturbatory responsiveness into dimensions associated with masturbation when alone and with a partner without intercourse. A confirmatory factor analysis substantiated the existence of the three dimensions of female orgasm. However, female orgasmic responsiveness was highly correlated across the three dimensions so that it was possible to isolate a large second-order factor of general orgasmic responsiveness that is consistent with most of the physiological evidence on the unitary nature of orgasm. Correlates of the orgasm scales with background data revealed some discriminant validity for the three dimensions of responsiveness.

An 8-year study of multiple influences on drug use and drug use consequences.

Multiple long-term influences on young adult drug use and abuse were tested within an interactionist perspective using latent-variable causal models. Intrapersonal influences included early drug use and social conformity. Proximal interpersonal influences were captured by perceived peer drug use, perceived adult drug use, and family disruption. More distal influences included perceptions of community approval or disapproval for drug use. Outcome measures included not only use of drugs but also disruptive drug use (getting high at work or school) and self-admitted problems with drugs. Data were obtained from 654 participants at three equally spaced time points during an 8-year longitudinal study that began when the subjects were in junior high school. Confirmatory factor analyses were used to test the adequacy of the hypothesized measurement model. Subsequently, a structural equation, or path model, was used to examine the across-time relations among the latent and manifest variables. Stability effects were found for all repeatedly measured latent variables across the 4-year spans. Social conformity strongly influenced other latent variables across time. Early drug use and perceived adult drug use were prominent predictors of young adult drug use, whereas prior drug use predicted disruptive drug use and a lack of social conformity predicted problems with drug use. Early adult alcohol use predicted later disruptive drug use and problems with drug use. Perceived adult drug use generally influenced later alcohol use, whereas peer drug use influenced later cannabis and hard-drug use. Implications for prevention and treatment using a multicausal interactionist perspective are discussed.

A comparison of two latent variable causal models for adolescent drug use.

The competing simplex (involvement) and common factor models for youthful drug use are statistically compared using causal modeling with latent variables methods in a sample of 1,634 young adolescents. Latent variables of alcohol, marijuana, and other illicit drug use were confirmed and causally interconnected in a set of states. The confirmatory factor and the simplex stage models were found to be acceptable representations of the observed data according to both statistical and psychometric criteria. The results are discussed in terms of their implications for theories of emerging lifestyles including drug use, methodology, and policy about psychoactive substances.

Personality, problem drinking, and drunk driving: mediating, moderating, and direct-effect models.

Three different general explanations of the effect of personality on problems from drinking alcohol were investigated. One general explanation involved mediating effects. The 2nd explanation involved direct effects of personality. The 3rd general personality process held that alcohol consumption and personality interact as moderating effects on drinking problems. Results provided support for each of the 3 general explanations of personality effects, although certain effects were found primarily for only 2 of the 6 personality constructs investigated (sensation seeking and cognitive motivation). These findings helped delimit the personality processes associated with drinking problems and demonstrated the viability of several specific processes that go beyond traditional assumptions about personality and problem drinking.

Determinants of sexual and dating behaviors among adolescents.

The study of adolescent sexual development has typically focused on the incidence and age of onset of particular behaviors. Recent developments in the analysis of covariance structures permit more powerful simultaneous tests of the determinants of adolescent sexual behavior. These new methods allow the assessment and evaluation of the interrelatedness between unobserved, latent constructs. Data were obtained from 183 males and 193 females ranging in age from 12 to 18 years old, selected from three birth cohorts using a stratified random sample. Eight latent constructs--Sexual and Dating Involvement, Self-Acceptance, Feminist Sex Role Attitudes, Deviant Social Network, Importance of Dating, Lack of Heterosexual Competence, Sexually Active Social Network, and Stressful Change Events--were hypothesized to explain the variance in 26 observed variables. Several causal models were tested that specified structural relationships among the latent constructs. A theoretically meaningful model was arrived at that linked the eight, error-free factors.

Substance use and abuse among children and teenagers.

During the past several years, there has been a renewed national concern about drug abuse, culminating in the current "war on drugs." In this review, we emphasize that even though child or teenage drug use is an individual behavior, it is embedded in a sociocultural context that strongly determines its character and manifestations. Our focus is on psychoactive substances both licit (cigarettes and alcohol) and illicit (e.g., cannabis and cocaine). We feel that it is critical to draw a distinction between use and abuse of drugs and to do so from a multidimensional perspective that includes aspects of the stimulus (drug), organism (individual), response, and consequences. Our selective review of substance use and abuse among children and adolescents covers epidemiology (patterns and extent of drug use), etiology (what generates substance use), prevention (how to limit drug use), treatment (interventions with drug users), and consequences (effects and outcomes of youthful drug use).

The impact of late adolescent substance use on young adult health status and utilization of health services: a structural-equation model over four years.

This study examined the impact of adolescent substance use on young adult health status and health service utilization, while controlling for earlier subjective and objective health problems and seeking medical care. Data were obtained from 654 participants when they were in late adolescence and 4 years later when they were young adults. Latent-variable models were used to determine what effect, if any, adolescent drug use had on later health. A second-order factor of general drug use had a small, but significant, influence on decreasing subjective physical hardiness over the 4-year period. However, general drug use was not directly related to later physical symptomatology, subjective health problems, or health service utilization. General cigarette use and amount of cigarette smoking were significantly related to a range of health outcomes including increased psychosomatic symptoms, respiratory symptoms, reporting trouble with health, nights spent in the hospital, and health service utilization, as well as decreased subjective physical hardiness. Early cannabis use decreased later subjective physical hardiness and increased health problems experienced over the 4 years. Hard drug use was associated with increased emergency physician visits during adolescence and increased perceived trouble with health as young adults. Due to the fact that adolescent general drug use decreased subjective physical hardiness over 4 years, it is possible that continued substance use might lead to more serious objective health consequences later in life (i.e. physical symptoms).

Structural equation models in medical research.

Structural equation modelling (SEM) is a modern statistical method that allows one to evaluate causal hypotheses on a set of intercorrelated nonexperimental data. The sample variances and covariances, and possibly the means, are compared to those predicted by a theory-based hypothetical model after optimal estimation of the parameters of the model. The goodness-of-fit of the empirical data to the hypothesized model is evaluated statistically. This review describes the underlying statistical theory and rationale of SEM. Both confirmatory factor analysis and latent variable path models are discussed. The applicability of SEM to assessment of reliability and validity is noted. A detailed example is provided, and several examples from the medical literature are briefly reviewed. Cautions regarding the possible misuse or misinterpretation of the technique are also mentioned. Possible future directions for the use of SEM in medical research are suggested. Two appendices provide more technical details.

Interactive and higher-order effects of social influences on drug use.

The study of moderators and higher-order effects of social influences on drug use has many implications for theories of health behavior. In the present study, we investigated the longitudinal predictive effects of some of the prominent moderator variables that represent forms of susceptibility toward social influence in teenage drug use. We also studied the possibility that social influence may predict drug use in nonlinear (quadratic) forms, consistent with theories proposing that threshold or decelerating effects may occur in social influences on normatively sanctioned behaviors. Results showed that several of the interactive and quadratic predictive effects were significant. The findings supported the views that certain moderator variables act as buffers, which either protect the individual from social pressures to use drugs, or make the individual more susceptible to such pressures. In addition, two of the obtained quadratic effects of social influence lent support to the application of social impact theory to drug use. Overall, our findings suggest that interactive and nonlinear approaches to social influences on drug use provide a unique and viable theoretical perspective from which to construe this problem health behavior.