Multilevel Analysis, Mixed-Effect Modeling, and Longitudinal Data Analysis with R

About

This 5-day course presents the highly popular methodology of multilevel and mixed effect modeling (MLMEM) across the social, marketing, business, behavioral, educational, life, health, biomedical, and organizational sciences. In addition, it discusses the specific applications of MLMEM in the analysis of longitudinal data that are currently very frequently collected in these and cognate disciplines. MLMEM provides a widely applicable approach to modeling and accounting for clustering effects that impact the majority of contemporary empirical studies in the above and a number of related sciences. A key feature of these effects is the associated relationship among and similarity of the observed scores collected from members of studied groups or clusters of units of analysis (usually persons, respondents, patients, employees, students, clients, etc., but could also be higher-order aggregates of them). If this similarity (within-group ‘correlation’) is not properly handled, as would be the case if using standard statistical analysis and modeling methods, incorrect statistical results ensue (e.g., deflated and incorrect standard errors, confidence intervals, and hypothesis test results). These results typically yield substantive conclusions that can be seriously misleading. A main achievement of MLMEM is the proper handling of the clustering effects, leading to valid and dependable statistical findings and substantive results. A particular field of application of MLMEM is that of longitudinal data analysis and modeling. Designs and studies producing such data are very frequently utilized in the social, marketing, business, behavioral, medical, life, health, and organizational as well as related sciences. Through appropriate use of MLMEM for longitudinal data analysis, whose detailed coverage is also part of this course, data resulting from studies involving repeated measures can be analyzed and modeled applying relevant statistical procedures that lead to valid and dependable results entailing correct statistical findings as well as substantive interpretations and conclusions.

Course leader

Tenko Raykov

Target group

Master | PhD | Postdoc | Professional

Fee info