Mixture modeling is a widely applied data analysis technique used to identify unobserved heterogeneity in a population. Despite mixture models' usefulness in practice, one unresolved issue in the application of mixture models is that there is not one commonly accepted statistical indicator for deciding on the number of classes in a study population. This article presents the results of a simulation study that examines the performance of likelihood-based tests and the traditionally used Information Criterion (ICs) used for determining the number of classes in mixture modeling. We look at the performance of these tests and indexes for 3 types of mixture models: latent class analysis (LCA), a factor mixture model (FMA), and a growth mixture models (GMM). We evaluate the ability of the tests and indexes to correctly identify the number of classes at three different sample sizes (n = 200, 500, 1,000). Whereas the Bayesian Information Criterion performed the best of the ICs, the bootstrap likelihood ratio test ...

Deciding on the Number of Classes in Latent Class Analysis and Growth Mixture Modeling: A Monte Carlo Simulation Study

Statistical Analysis with Missing Data

OBJECTIVE: This report estimates the numbers of 18-24 year old United States college students who annually experience alcohol-related deaths, injuries and other health problems. METHOD: We examined traffic and unintentional injury deaths in 1998 reported by the National Highway Traffic Safety Administration and the Centers for Disease Control (CDC). We also examined results of national coroner studies, Department of Education college enrollment data, the National Household Survey on Drug Abuse (NHSDA), the CDC National College Health Risk Behavior Survey and the Harvard School of Public Health College Alcohol Survey (CAS). All survey participants were ages 18-24: 6,930 college and 12,394 noncollege respondents in the NHSDA survey; 3,077 college students in the CDC survey; and 12,217 full-time 4-year college students in the CAS. Based on the number and proportion of 18-24 year olds enrolled in college, data on alcohol involvement in injury deaths among 18-24 year olds and survey responses, we calculated the numbers of 18-24 year old alcohol-related injury deaths and other health problems. RESULTS: We estimate that over 1,400 students aged 18-24 and enrolled in 2- and 4-year colleges died in 1998 from alcohol-related unintentional injuries, including motor vehicle crashes. According to surveys conducted in 1999, in the preceding year, over 2 million of the 8 million college students in the United States drove under the influence of alcohol and over 3 million rode with a drinking driver. Over 500,000 full-time 4-year college students were unintentionally injured under the influence of alcohol and over 600,000 were hit or assaulted by another student who had been drinking. CONCLUSIONS: There is an urgent need for expanding prevention and treatment programs, to reduce alcohol-related harm among U.S. college students and other young adults. Language: en

/pdf/magnitude-of-alcohol-related-mortality-and-morbidity-among-u-2bgxvwjhzw.pdf

Magnitude of alcohol-related mortality and morbidity among U.S. college students ages 18-24.

Integrating data from the National Highway Traffic Safety Administration, the Centers for Disease Control and Prevention, national coroner studies, census and college enrollment data for 18-24-year-olds, the National Household Survey on Drug Abuse, and the Harvard College Alcohol Survey, we calculated the alcohol-related unintentional injury deaths and other health problems among college students ages 18-24 in 1998 and 2001. Among college students ages 18-24 from 1998 to 2001, alcohol-related unintentional injury deaths increased from nearly 1600 to more than 1700, an increase of 6% per college population. The proportion of 18-24-year-old college students who reported driving under the influence of alcohol increased from 26.5% to 31.4%, an increase from 2.3 million students to 2.8 million. During both years more than 500,000 students were unintentionally injured because of drinking and more than 600,000 were hit/assaulted by another drinking student. Greater enforcement of the legal drinking age of 21 and zero tolerance laws, increases in alcohol taxes, and wider implementation of screening and counseling programs and comprehensive community interventions can reduce college drinking and associated harm to students and others.

/pdf/magnitude-of-alcohol-related-mortality-and-morbidity-among-u-13kwtgwns9.pdf

MAGNITUDE OF ALCOHOL-RELATED MORTALITY AND MORBIDITY AMONG U.S. COLLEGE STUDENTS AGES 18–24: Changes from 1998 to 2001

This transdisciplinary review of the literature addresses the questions, Do stress and negative affect (NA) promote smoking? and Does smoking genuinely relieve stress and NA? Drawing on both human and animal literatures, the authors examine these questions across three developmental stages of smoking--initiation, maintenance, and relapse. Methodological and conceptual distinctions relating to within- and between-subjects levels of analyses are emphasized throughout the review. Potential mechanisms underlying links between stress and NA and smoking are also reviewed. Relative to direct-effect explanations, the authors argue that contextual mediator-moderator approaches hold greater potential for elucidating complex associations between NA and stress and smoking. The authors conclude with recommendations for research initiatives that draw on more sophisticated theories and methodologies.

https://www.unizd.hr/Portals/12/Nastavnici/Zvjezdan%20Penezic/Nastava/PEM/PDF/Negativno%20potkrepljenje%20i%20ovisnost/Smoking,%20Stress,%20and%20Negative%20Affect.pdf

Smoking, stress, and negative affect: correlation, causation, and context across stages of smoking

Aims To review three topics pertaining to the validity of alcohol self-reports: factors that influence response accuracy; the relative merits of different self-report approaches; and the utility of using alternative measures to confirm verbal reports.

Findings  Response behavior is influenced by the interaction of social context factors, respondent characteristics, and task attributes. Although research has advanced our knowledge about self-report methods, many questions remain unanswered. In particular, there is a need to investigate how task demands interact with different patterns of drinking behavior to affect response accuracy. There is also a continuing need to use multiple data sources to examine the extent of self-report response bias, and to determine whether it varies as a function of respondent characteristics or assessment timing.

Conclusion  Self-report methods offer a reliable and valid approach to measuring alcohol consumption. The accuracy of such methods, however, can be improved by research directed at understanding the processes involved in response behavior.

The validity of self-reports of alcohol consumption: state of the science and challenges for research.

Substantial correlational evidence supports a causal (mediational) interpretation of alcohol expectancy operation, but definitive support requires a true experimental test. Thus, moderately to heavily drinking male college students were randomly assigned to 1 of 3 conditions in a pre-post design: Expectancy challenge (designed to manipulate expectancy levels), "traditional" information, and assessment-only control. Expectancy challenge produced significant drinking decreases, compared with the other 2 groups. Decreases in measured expectancies paralleled drinking decreases in the challenge condition. Significant increases in alcohol knowledge in the traditional program were not associated with decreased drinking. These experimental findings support a causal (mediational) interpretation of expectancy operation. The implications for a cognitive (memory) model of expectancies and for prevention and intervention programs for problem drinking and alcoholism are discussed.

Expectancy challenge and drinking reduction : experimental evidence for a mediational process

Surveys have documented excessive drinking among college students and tracked annual changes in consumption over time This study extended previous work by examining drinking changes during the freshman year, using latent growth curve (LGC) analysis to model individual change, and relating risk factors for heavy drinking to growth factors in the model Retrospective monthly assessments of daily drinking were used to generate weekly estimates Drinking varied considerably by week, apparently as a function of academic requirements and holidays A 4-factor LGC model adequately fit the data In univariate analyses, gender, race/ethnicity, alcohol expectancies, sensation seeking, residence, and data completeness predicted growth factors (ps <05); gender, expectancies, residence, and data completeness remained significant when covariates were tested simultaneously Substantive, methodological, and policy implications are discussed

Jack Darkes

Papers

The validity of self-reports of alcohol consumption: state of the science and challenges for research.

Expectancy challenge and drinking reduction : experimental evidence for a mediational process

Up close and personal: temporal variability in the drinking of individual college students during their first year.

Alcohol expectancy theory: The application ofcognitive neuroscience

Alcohol expectancy theory: The application of cognitive neuroscience