Answers are grounded in this paper and cite their sources.
Researchers in the field of early adolescence interested in quantifying the environmental influences on a response variable of interest over time would use cluster sampling (i.e., obtaining repeated measures from students nested within classrooms and/or schools) to obtain the needed sample size. The resulting longitudinal data would be nested at three levels (e.g., repeated measures [Level 1], collected across participants [Level 2], and nested within different schools [Level 3]). A previous publication addressed statistical analysis issues specific to cross-sectional three-level data analytic designs. This article expands upon the previous cross-sectional three-level publication to address topics specific to longitudinal three-level data analyses efforts. Although all analysis examples are demonstrated using SAS, the equivalent SPSS and Mplus syntax scripts, as well as the generated example data and additional supplemental materials, are available online.
No takes yet. Share an insight, caveat, or question.
Peugh et al. (2016) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: