Feasibility of Teacher-Centric Analyses Using NAEP Data

Session Type: Symposium

Abstract

NAEP’s contribution to research on teachers and teacher effects is possible due to data gathered through the NAEP teacher questionnaire. Yet in spite of the immense value of NAEP teacher-level data, their analysis has been limited until now because teachers have not been used as the unit of analysis. This symposium examines the feasibility of constructing representative sampling weights for NAEP teachers and using them in teacher-centric analyses of NAEP data. We will show that it is possible to conduct teacher-weighted analyses of NAEP teacher data, and then demonstrate the results of single-level and multi-level analyses using teacher weights. Our study is intended to inform how NCES collects, disseminates and analyzes NAEP teacher data.

Sub Unit

• SIG-NAEP Studies

Chair

• Markus Broer, American Institutes for Research

Papers

• Introduction and Importance of the Study - Taslima Rahman, University of Maryland - College Park

• Probability of Selecting a Representative Teacher Sample - Austin Lasseter, American Institutes for Research

• Adjusting NAEP School Weights for Teacher Nonresponse - Jiao Yu, American Institutes for Research

• Validating NAEP Teacher Weights - Fei Liu, Macrosys LLC

• Single-Level Weighted Analyses of NAEP Teachers - Jonathan Phelan, American Institutes for Research

• Multilevel Weighted Analyses of NAEP Teachers and Students - Austin Lasseter, American Institutes for Research

Discussants

• Keith Rust, Westat

• Avinash Singh, American Institutes for Research