Abstract
This article highlights current thoughts on the design and analysis of longitudinal evaluation. It introduces cross-sectional, trend, cohort, panel, and time-series designs. Two general research questions are formulated that speak to (1) whether and (2) when change has occurred. The multilevel growth model can address whether change has occurred and apply to all designs with special model specifications. The survival model can address when change has occurred and apply to true longitudinal designs (cohort, panel, and time series). The cross-lagged model is introduced as an effective longitudinal way to address cause–effect relationships in nonexperimental conditions and applies to true longitudinal designs.
| Original language | English |
|---|---|
| Title of host publication | International Encyclopedia of Education, Third Edition |
| Pages | 757-764 |
| Number of pages | 8 |
| ISBN (Electronic) | 9780080448947 |
| DOIs | |
| State | Published - Jan 1 2009 |
Bibliographical note
Publisher Copyright:© 2010 Elsevier Ltd. All rights reserved.
Keywords
- Cohort design
- Cross-lagged model
- Cross-sectional design
- Longitudinal analysis
- Longitudinal design
- Longitudinal evaluation
- Longitudinal research
- Multilevel growth model
- Panel design
- Survival model
- Time-series design
- Trend design
ASJC Scopus subject areas
- General Social Sciences