Construction of Replicate Weights for Project TALENT

• Activity Number: 636
• Type: Topic Contributed
• Date/Time: Thursday, August 8, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Survey Research Methods Section
• Author(s): Zhulin He*+ and Alan F. Karr and Michael Cohen and Danielle Battle and Deanna Lyter Achorn and Alexander D. McKay
• Companies: National Institute of Statistical Sciences and National Institute of Statistical Sciences and American Institutes for Research and American Institutes for Research and American Institutes for Research and American Institutes for Research
• Keywords: Project TALENT ; Weights adjustment ; CHAID ; Jackknife replicate weights ; Survey analysis

Abstract

Project TALENT is a large, nationally representative longitudinal study developed by American Institutes for Research and conducted from 1960 to 1974. The goal was to assess the interests, abilities, and demographics of 9th-12th graders and their trajectories into adulthood. More than 1,200 junior and senior high schools participated. Replicate weights were not constructed at the time, preventing the estimation of standard errors. In this paper, the retrospective construction of 118 sets of student-level replicate weights is described. The process entailed adjustment of the original base year (1960) student weights and school weights to better estimate the educational and life experiences that are most important to individuals' life trajectories. CHAID analysis was performed to generate variance strata and variance primary sampling units. Finally, the student-level replicate weights were constructed using a jackknife procedure. The use of replicate weights is illustrated by estimating standard errors for quantiles of composite cognitive scores constructed from student questionnaires.