BackHome
Refresh: OffView Personal Schedule
Session Summary

10.010 - PDC06: Fitting Multilevel Models to Early Childhood Longitudinal Study Data Using R

Fri, April 5, 8:00am to 5:00pm, Metro Toronto Convention Centre, Floor: 700 Level, Room 713B

Session Type: Professional Development Course

Abstract

This course is designed to provide an applied understanding and hands-on practice in multilevel mixed models using the R package EdSurvey, which was developed for analyzing NCES cross-sectional and longitudinal assessment data. Using the Early Childhood Longitudinal Study, Kindergarten Class of 2010-11 (ECLS-K:2011) public-use datafile provided in the workshop, participants will be introduced to the EdSurvey package and learn how to use the package to obtain results from various types of mixed models including the null model and random intercept and slope models. Multilevel models for binary outcomes and longitudinal data will also be covered. This workshop will introduce the unique design features of ECLS-K:2011 data to researchers and provide guidance on specific analytic strategies related to these features, including centering and the selection and use of appropriate sampling weights and variance estimation procedures. The knowledge and analytic approach covered in this course can be applied to working with other ECLS datasets. There will be time designated for participants to use EdSurvey to practice the techniques discussed with variables of their own interest.

Participants should have basic knowledge of statistical techniques including statistical inference and multiple regression analysis. It is helpful, though not required, to have some experience in binary logistic regression. Having working knowledge of R and multilevel modeling is preferred. Participants need to bring a laptop preloaded with the latest version of the R and RStudio software to participate in the hands-on portion.

Sub Unit

Director

Instructors