Using a bifactor model to assess the factor structure of the Phonological Awareness Literacy Screening for Grades One through Three

Jan 1, 2014·
Francis L. Huang
Francis L. Huang
· 0 min read
Abstract
The factor structure of the Phonological Awareness Literacy Screening for Grades 1 through 3 (PALS 1-3), a widely used early literacy screener in the Commonwealth of Virginia, was investigated using a large sample of public-school second-grade students (n = 14,993). Three alternative factor models (i.e., a one-factor, two-correlated factors, and a bifactor model) were tested and explored using an exploratory sample consisting of a randomly selected half of the overall sample. Model fit indices using confirmatory factor analyses indicated that a bifactor model fit the best and supported the scoring methods used with PALS 1-3, which is largely measured by a general factor of orthographic knowledge. The model was found to replicate in the randomly selected hold-out sample as well and exhibited adequate measurement precision (ωh = .88).
Type
Publication
Journal of Psychoeducational Assessment, 32, 638-650