Variable-Oriented and Person-Oriented Perspectives of Analysis: The Example of Alcohol Consumption in Adolescence

• NCJ Number: 216571
• Journal: Journal of Adolescence Volume: 29 Issue: 6 Dated: December 2006 Pages: 981-1004
• Author(s): Alexander von Eye; G. Anne Bogat; Jean E. Rhodes
• Date Published: December 2006
• Length: 24 pages
• Publication Format: Article
• Publication Type: Report (Study/Research)
• Language: English
• Country: United States of America

Annotation

This article explored two types of approaches to data analysis--person-oriented and variable-oriented approaches--in terms of the types of information they provided.

Abstract

Using data on the effects of the relationship between parent attitudes and gender on youth alcohol consumption, the analysis indicated that person-oriented and variable-oriented approaches to data analysis provided different, yet complimentary, information. The variable-oriented approach produced results indicating that boys drank more heavily than girls and that greater negative parental attitudes toward drinking were associated with less youth drinking. The person-oriented approach produced results indicating little change in drinking patterns based on parental attitudes toward drinking among both boys and girls. Although different, the results from the two analyses can be viewed as complimentary in that the person-oriented results provided information about the group of respondents who behaved counter to the general trend observed in the variable-oriented analysis. Thus, the variable-oriented approach provided information about overall trends while the person-oriented approach identified groups of participants whose behavior departed from the overall trends. Data on 3,558 control group youth were drawn from the National Cross-Site Evaluation of High Risk Youth Programs, which was a 5-year national study (1995 to 2000) that examined the effectiveness of alcohol prevention programs for 9- to 18-year-olds from a variety of ethnic and cultural backgrounds. Variables under analysis for the current study included the youths’ demographic characteristics, alcohol use, and perceptions of parental attitudes toward drinking. This information was collected at two time points (program entry and program exit) via bilingual self-report questionnaires. Data analysis for the variable-oriented approach relied on repeated measures ANOVA (analysis of variance) while the data analysis for the person-oriented approach relied on a first-order configural frequency analysis. Tables, figures, references