Poster #65 - Using Regression Trees to Identify Distinct Subgroup Performance and Achievement Gain

• In Event: 25.087 - MTCC Poster Session Four
  In Poster Session: 25.087-9 - Upcoming Researchers Showcase Multiple Examples of Research, Evaluation, and Assessment in Schools: A Graduate Student Poster Session

Abstract

The study applied regression tree analyses to examine student assessment data using Early Childhood Longitudinal Study-Kindergarten (ECLS-K) as a case study. It investigated factors from students' demographic backgrounds to ascertain their relationships to students' academic performance and achievement gain in reading. The reading performances of the upper, middle, lower end of kindergarteners on the SES scale were variously related to different factors (i.e., race, region, and gender). For reading gain related, important factors included some of the same factors but also depend on location. This study has shown that regression tree analyses can be utilized with assessment data to display complex patterns between factors and outcomes that would be obscured by trying to model the entire population as a whole.

Authors

• Ya Mo

• Nell Sedransk, National Institute of Statistical Sciences

• Ting Shen, University of Missouri - Columbia