Related Experiment Videos
Structural equation modeling with longitudinal data: strategies for examining group differences and reciprocal
1Department of Psychology, Virginia Commonwealth University, Richmond 23284-2018.
Journal of Consulting and Clinical Psychology
|June 1, 1994
Summary
Structural equation modeling (SEM) effectively analyzes group differences and causal relationships in longitudinal studies. This method offers advantages over traditional techniques for understanding complex data, such as adolescent alcohol use patterns.
Area of Science:
- Psychology
- Statistics
- Sociology
Background:
- Longitudinal designs are crucial for understanding developmental trends and causal pathways.
- Traditional statistical methods have limitations in analyzing complex, multi-wave data.
- Structural equation modeling (SEM) offers a robust framework for analyzing such data.
Purpose of the Study:
- To describe the application of SEM with latent variables for examining group differences in passive longitudinal designs.
- To compare SEM with other statistical methods like cross-lagged panel correlations, regression, and path analysis.
- To illustrate SEM's utility using a 3-wave longitudinal study on adolescent alcohol use.
Main Methods:
- Structural equation modeling (SEM) with latent variables was employed.
- Analysis involved a 3-wave longitudinal study design.
- Generalizability of measurement and structural models were assessed across gender and time.
Main Results:
- SEM demonstrated its capability to test competing models of cause-effect relationships.
- The study assessed the generalizability of the measurement and structural models.
- Competing hypotheses regarding adolescent alcohol use were evaluated.
Conclusions:
- SEM is a powerful tool for analyzing complex longitudinal data and group differences.
- The study highlights the strengths of SEM in modeling causal relationships in developmental research.
- Limitations of SEM for longitudinal data were also discussed.