Donna L. Coffman

Asymptotically Distribution Free (Adf) Interval Estimation of Coefficient Alpha

The point estimate of sample coefficient alpha may provide a misleading impression of the reliability of the test score. Because sample coefficient alpha is consistently biased downward, it is more likely to yield a misleading impression of poor reliability. The magnitude of the bias is greatest precisely when the variability of sample alpha is greatest (small population reliability and small sample size). Taking into account the variability of sample alpha with an interval estimator may lead to retaining reliable tests that would be otherwise rejected. Here, the authors performed simulation studies to investigate the behavior of asymptotically distribution-free (ADF) versus normal-theory interval estimators of coefficient alpha under varied conditions. Normal-theory intervals were found to be less accurate when item skewness >1 or excess kurtosis >1. For sample sizes over 100 observations, ADF intervals are preferable, regardless of item skewness and kurtosis. A formula for computing ADF confidence intervals for coefficient alpha for tests of any size is provided, along with its implementation as an SAS macro.

Social Support, Stress, and Self-Efficacy: Effects on Students' Satisfaction.

Many variables impact a students adjustment to college and ultimate academic and social success. This study investigated the relationships between social support, perceived stress, self-efficacy, and life satisfaction among 94 first-year college students. Students who reported higher levels of social support and self-efficacy and lower levels of perceived stress also reported higher levels of life satisfaction. Social support and stress together accounted for 41 percent of the variance in life satisfaction ratings, with social support providing the largest contribution. The findings have implications for prevention and interventions with students experiencing difficulty in adjusting to college.

Estimating Causal Effects in Mediation Analysis Using Propensity Scores

Mediation is usually assessed by a regression-based or structural equation modeling (SEM) approach that we refer to as the classical approach. This approach relies on the assumption that there are no confounders that influence both the mediator, M, and the outcome, Y. This assumption holds if individuals are randomly assigned to levels of M but generally random assignment is not possible. We propose the use of propensity scores to help remove the selection bias that can result when individuals are not randomly assigned to levels of M. The propensity score is the probability that an individual receives a particular level of M. Results from a simulation study are presented to demonstrate this approach, referred to as Classical + Propensity Model (C+PM), confirming that the population parameters are recovered and that selection bias is successfully dealt with. Comparisons are made to the classical approach that does not include propensity scores. Propensity scores were estimated by a logistic regression model. If all confounders are included in the propensity model, then the C+PM is unbiased. If some, but not all, of the confounders are included in the propensity model, then the C+PM estimates are biased although not as severely as the classical approach (i.e., no propensity model is included).

Why do high school seniors drink? Implications for a targeted approach to intervention.

The transition from high school to college provides a potentially critical window to intervene and reduce risky behavior among adolescents. Understanding the motivations (e.g., social, coping, enhancement) behind high school seniors’ alcohol use could provide one important avenue to reducing risky drinking behaviors. In the present study, latent class analysis was used to examine the relationship between different patterns of drinking motivations and behaviors in a sample of 12th graders (N = 1,877) from the 2004 Monitoring the Future survey. Unlike previous variable-centered analyses, this person-centered approach identifies types of motivations that cluster together within individuals and relates membership in these profiles to drinking behaviors. Results suggest four profiles of drinking motivations for both boys and girls, including Experimenters, Thrill-seekers, Multi-reasoners, and Relaxers. Early initiation of alcohol use, past year drunkenness, and drinking before 4 p.m. were associated with greater odds of membership in the Multi-reasoners class as compared to the Experimenters class. Although the strength of these relationships varied for boys and girls, findings were similar across gender suggesting that the riskiest drinking behavior was related to membership in the Multi-reasoners class. These findings can be used to inform prevention programming. Specifically, targeted interventions that tailor program content to the distinct drinking motivation profiles described above may prove to be effective in reducing risky drinking behavior among high school seniors.

Assessing mediation using marginal structural models in the presence of confounding and moderation.

This article presents marginal structural models with inverse propensity weighting (IPW) for assessing mediation. Generally, individuals are not randomly assigned to levels of the mediator. Therefore, confounders of the mediator and outcome may exist that limit causal inferences, a goal of mediation analysis. Either regression adjustment or IPW can be used to take confounding into account, but IPW has several advantages. Regression adjustment of even one confounder of the mediator and outcome that has been influenced by treatment results in biased estimates of the direct effect (i.e., the effect of treatment on the outcome that does not go through the mediator). One advantage of IPW is that it can properly adjust for this type of confounding, assuming there are no unmeasured confounders. Further, we illustrate that IPW estimation provides unbiased estimates of all effects when there is a baseline moderator variable that interacts with the treatment, when there is a baseline moderator variable that interacts with the mediator, and when the treatment interacts with the mediator. IPW estimation also provides unbiased estimates of all effects in the presence of nonrandomized treatments. In addition, for testing mediation we propose a test of the null hypothesis of no mediation. Finally, we illustrate this approach with an empirical data set in which the mediator is continuous, as is often the case in psychological research.

Dietary patterns and diet quality among diverse older adults: The university of Alabama at Birmingham study of aging

ObjectivesTo characterize dietary patterns among a diverse sample of older adults (≥ 65 years).DesignCross-sectional.SettingFive counties in west central Alabama.ParticipantsCommunity-dwelling Medicare beneficiaries (N=416; 76.8 ±5.2 years, 56% female, 39% African American) in the University of Alabama at Birmingham (UAB) Study of Aging.MeasurementsDietary data collected via three, unannounced 24-hour dietary recalls was used to identify dietary patterns. Foods were aggregated into 13 groups. Finite mixture modeling (FMM) was used to classify individuals into three dietary patterns. Differences across dietary patterns for nutrient intakes, sociodemographic, and anthropometric measurements were examined using chi-square and general linear models.ResultsThree dietary patterns were derived. A “More healthful” dietary pattern, with relatively higher intakes of fruit, vegetables, whole grains, eggs, nuts, legumes and dairy, was associated with lower energy density, higher quality diets as determined by Healthy Eating Index (HEI)-2005 scores and higher intakes of fiber, folate, vitamins C and B6, calcium, iron, magnesium, and zinc. The “Westernlike” pattern was defined by an intake of starchy vegetables, refined grains, meats, fried poultry and fish, oils and fats and was associated with lower HEI-2005 scores. The “Low produce, high sweets” pattern was characterized by high saturated fat, and low dietary fiber and vitamin C intakes. The strongest predictors of better diet quality were female gender and non-Hispanic white race.ConclusionThe dietary patterns identified may provide a useful basis on which to base dietary interventions targeted at older adults. Examination of nutrient intakes regardless of the dietary pattern suggests that older adults are not meeting nutrient recommendations and should continue to be encouraged to choose high quality diets.

Evaluating Latent Growth Curve Models Using Individual Fit Statistics

The usefulness of assessing individual fit in latent growth curve models was examined. The study used simulated data based on an unconditional and a conditional latent growth curve model with a linear component and a small quadratic component and a linear model was fit to the data. Then the overall fit of linear and quadratic models to these data was examined. The quadratic models fit the data well but the linear models did not, even though the quadratic component was very small. Then fit was examined at the individual level. Based on individual log likelihood values, a linear model fit most of the individual trajectories well. The article then demonstrates the use of individual fit indexes in empirical data and offers suggestions for assessing individual fit in applied research.

Can we spin straw into gold? An evaluation of immigrant legal status imputation approaches.

Researchers have developed logical, demographic, and statistical strategies for imputing immigrants’ legal status, but these methods have never been empirically assessed. We used Monte Carlo simulations to test whether, and under what conditions, legal status imputation approaches yield unbiased estimates of the association of unauthorized status with health insurance coverage. We tested five methods under a range of missing data scenarios. Logical and demographic imputation methods yielded biased estimates across all missing data scenarios. Statistical imputation approaches yielded unbiased estimates only when unauthorized status was jointly observed with insurance coverage; when this condition was not met, these methods overestimated insurance coverage for unauthorized relative to legal immigrants. We next showed how bias can be reduced by incorporating prior information about unauthorized immigrants. Finally, we demonstrated the utility of the best-performing statistical method for increasing power. We used it to produce state/regional estimates of insurance coverage among unauthorized immigrants in the Current Population Survey, a data source that contains no direct measures of immigrants’ legal status. We conclude that commonly employed legal status imputation approaches are likely to produce biased estimates, but data and statistical methods exist that could substantially reduce these biases.

Predicting substance use behavior among South African adolescents: The role of leisure experiences across time.

Using seven waves of data, collected twice a year from the 8th through the 11th grades in a low-resource community in Cape Town, South Africa, we aimed to describe the developmental trends in three specific leisure experiences (leisure boredom, new leisure interests, and healthy leisure) and substance use (cigarettes, alcohol, and marijuana) behaviors and to investigate the ways in which changes in leisure experiences predict changes in substance use behaviors over time. Results indicated that adolescents’ substance use increased significantly across adolescence, but that leisure experiences remained fairly stable over time. We also found that adolescent leisure experiences predicted baseline substance use and that changes in leisure experiences predicted changes in substance use behaviors over time, with leisure boredom emerging as the most consistent and strongest predictor of alcohol, cigarette, and marijuana use. Implications for interventions that target time use and leisure experiences are discussed.

Dietary patterns and relationship to obesity-related health outcomes and mortality in adults 75 years of age or greater

BackgroundThe prevalence of obesity-related adverse health outcomes is increasing among older adults. Because it is thought that nutrition plays an important role in successful aging, there has been considerable interest in the association between dietary patterns of older adults and obesity-related health outcomes.ObjectiveThis study examined the association between dietary patterns and mortality and prevalence of obesity-related health outcomes, namely cardiovascular disease (CVD), type 2 diabetes mellitus, hypertension, and metabolic syndrome (MetSyn), over a 5-year follow-up period in adults aged 75 years or greater.DesignA longitudinal observational study with cross-sectional dietary assessment.SettingRural Central Pennsylvania.ParticipantsCommunity-dwelling older adults (N = 449; 76.5 years old; 57% female).MeasurementsMultiple, unannounced, 24-hour dietary recalls were used to collect dietary intake. Cluster analysis was used to derive dietary patterns. Prevalence of CVD, diabetes mellitus, hypertension, and MetSyn was extracted from outpatient electronic medical records. Logistic regression was used to examine the associations between dietary patterns and health outcomes and mortality.Results‘Sweets and Dairy’, ‘Health-Conscious’ and ‘Western’ dietary patterns were identified. Compared to the ‘Health-Conscious’ pattern, those in the ‘Sweets and Dairy’ pattern had increased odds of hypertension over the follow-up period; adjusted odds ratio (95% CI) was 2.18 (1.11–4.30). No significant associations were found for CVD, diabetes mellitus, MetSyn or mortality with dietary patterns.ConclusionsThese findings support the potential value of healthy dietary patterns in the management of hypertension in older adults. We did not observe any other strong associations between dietary patterns and health outcomes or mortality in persons ≥ 75 years of age; thus failing to support the use of overly restrictive diet prescriptions for older persons, especially where food intake may be inadequate.