A Monte Carlo study compared 14 methods to test the statistical significance of the intervening variable effect. An intervening variable (mediator) transmits the effect of an independent variable to a dependent variable. The commonly used R. M. Baron and D. A. Kenny (1986) approach has low statistical power. Two methods based on the distribution of the product and 2 difference-in-coefficients methods have the most accurate Type I error rates and greatest statistical power except in 1 important case in which Type I error rates are too high. The best balance of Type I error and statistical power across all cases is the test of the joint significance of the two effects comprising the intervening variable effect.

/pdf/a-comparison-of-methods-to-test-mediation-and-other-21xho8v2xm.pdf

A comparison of methods to test mediation and other intervening variable effects.

The goals of this article are to (a) describe differences between moderator and mediator effects; (b) provide nontechnical descriptions of how to examine each type of effect, including study design, analysis, and interpretation of results; (c) demonstrate how to analyze each type of effect; and (d) provide suggestions for further reading. The authors focus on the use of multiple regression because it is an accessible data-analytic technique contained in major statistical packages. When appropriate, they also note limitations of using regression to detect moderator and mediator effects and describe alternative procedures, particularly structural equation modeling. Finally, to illustrate areas of confusion in counseling psychology research, they review research testing moderation and mediation that was published in the Journal of Counseling Psychology during 2001.

Testing Moderator and Mediator Effects in Counseling Psychology Research.

In diary studies, people provide frequent reports on the events and experiences of their daily lives. These reports capture the particulars of experience in a way that is not possible using traditional designs. We review the types of research questions that diary methods are best equipped to answer, the main designs that can be used, current technology for obtaining diary reports, and appropriate data analysis strategies. Major recent developments include the use of electronic forms of data collection and multilevel models in data analysis. We identify several areas of research opportunities: 1. in technology, combining electronic diary reports with collateral measures such as ambulatory heart rate; 2. in measurement, switching from measures based on between-person differences to those based on within-person changes; and 3. in research questions, using diaries to (a) explain why people differ in variability rather than mean level, (b) study change processes during major events and transitions, and (c) study interpersonal processes using dyadic and group diary methods.

/pdf/diary-methods-capturing-life-as-it-is-lived-2tmv49odub.pdf

Diary Methods: Capturing Life as it is Lived

The authors conducted a meta-analysis of 150 studies in which the risk-taking tendencies of male and female participants were compared. Studies were coded with respect to type of task (e.g., self-reported behaviors vs. observed behaviors), task content (e.g., smoking vs. sex), and 5 age levels. Results showed that the average effects for 14 out of 16 types of risk taking were significantly larger than 0 (indicating greater risk taking in male participants) and that nearly half of the effects were greater than .20. However, certain topics (e.g., intellectual risk taking and physical skills) produced larger gender differences than others (e.g., smoking). In addition, the authors found that (a) there were significant shifts in the size of the gender gap between successive age levels, and (b) the gender gap seems to be growing smaller over time. The discussion focuses on the meaning of the results for theories of risk taking and the need for additional studies to clarify age trends.

Gender differences in risk taking: A meta-analysis.

A 4-factor measure of drinking motives based on a conceptual model by Cox & Klinger (1988, 1990) is presented. Using data from a representative household sample of 1,243 Black and White adolescents, confirmatory factor analyses showed that the hypothesized model provided an excellent fit to the data and that the factor pattern was invariant across gender, race, and age. Each drinking motive was related to a distinct pattern of contextual antecedents and drinking-related outcomes, and these relationships did not generally vary across demographic subgroups. Results support both the conceptual validity of Cox and Klinger's model and the utility of this measure for clinical and research purposes across a diverse range of adolescent populations

Motivations for alcohol use among adolescents: development and validation of a four-factor model

Discusses substantive issues in predicting drinking patterns from expectancies, as well as issues of content and measurement of the scales developed to measure these expectancies. In recent years, much research has suggested that alcohol expectancies--or the beliefs that individuals hold about the effects of alcohol on their behavior, moods, and emotions--are an important factor in motivating drinking behavior. Although measures of these expectancies have consistently been shown to be correlated with measures of alcohol use, conceptual and methodological problems remain to be addressed. In order to progress in understanding this potentially important psychosocial factor in abusive and nonabusive drinking, alcohol expectancy research, which has been atheoretical in nature, should attend to potential contributions from other areas of research in psychology.

In search of the Seven Dwarves: issues of measurement and meaning in alcohol expectancy research

The goals of this research were to develop a scale to measure alcohol outcome expectancies that incorporated important features suggested by previous research; to examine the psychometric properties of the instrument, with particular attention to item discrimination; and to examine the relationship of positive and negative expectancy to self-reported alcohol use. In Study 1, a preliminary expectancy scale was constructed; factor analysis showed 2 general constructs representing positive and negative consequences of drinking. In Study 2, the scale was refined through tests of item discrimination and was used to predict alcohol use using structural equation modeling

Alcohol Outcome Expectancies: Scale Construction and Predictive Utility in Higher Order Confirmatory Models

Aims Because expectancies about the effects of alcohol change as drinking experience is accumulated, it is likely that the relationship of expectancy to drinking will differ with age. In this study, we examine the prediction of drinking behavior from positive and negative outcome expectancy at different ages.

Design  Data were collected as part of the National Alcohol Survey, using a multi-stage area probability sample of the household population of the 48 contiguous United States.

Participants  US residents aged 12 and older (n = 2875).

Measurements  Survey questions included drinking habits (frequency, quantity, frequency of drunkenness, maximum quantity) and beliefs about the effects of alcohol (alcohol expectancies).

Findings  Structural equation models tested the relationship of positive and negative expectancy to drinking behavior in six age groups. Outcome expectancy accounted for a larger portion of the variance in drinking among younger respondents than among older respondents. However, suppression effects were common. When suppression effects were considered, positive expectancy predicted drinking better than negative expectancy only among respondents under 35, while negative expectancy was a better predictor of drinking status in most respondents over 35 years. Among drinkers, positive expectancy predominated over negative expectancy when suppression effects were considered.

Conclusions  These results suggest that negative expectancy predicts abstention, while positive expectancy predicts level of drinking among drinkers. In expectancy research, differences between drinkers and abstainers, age of participants and the presence of suppression effects should be taken into account.

Alcohol expectancies and drinking in different age groups.

A questionnaire on beliefs about the effects of alcohol on the behavior of self and others was administered to a sample of college students and a sample from the general population. Respondents reported that alcohol effects are more likely to happen to others than to themselves, particularly for socially undesirable behaviors. Women reported a greater likelihood of behavioral impairment; men reported more aggressive behavior. Nondrinkers reported a greater likelihood of both positive and negative effects; heavier drinkers reported more pleasurable effects.

Beliefs about the effects of alcohol on self and others.

Traditional sex‐role stereotypes suggest that men and women engage in sex for different reasons. Previous studies have supported the notion that women are more motivated by emotional expression in having sex, and men are more concerned with physical gratification. In a survey of sexual behavior, heterosexual and homosexual respondents were asked to rate the importance of a variety of reasons for having sex and not having sex. The results showed that men and women differed in the importance attached to emotional and physical motives, with gender differences appearing in both heterosexuals and homosexuals. Certain practical motives (e.g., to reproduce, fear of AIDS, fear of pregnancy) differentiated between homosexuals and heterosexuals. Motivations predicting frequency of sexual behavior in the last month depended on the relationship status of the respondent. These findings, although suggesting that gender differences in motivations for sex persist in all kinds of relationships, point to many similarities ...

Barbara C. Leigh

Papers

In search of the Seven Dwarves: issues of measurement and meaning in alcohol expectancy research

Alcohol Outcome Expectancies: Scale Construction and Predictive Utility in Higher Order Confirmatory Models

Alcohol expectancies and drinking in different age groups.

Beliefs about the effects of alcohol on self and others.

Reasons for having and avoiding sex: Gender, sexual orientation, and relationship to sexual behavior