Wednesday, August 02, 2006

Learning styles: Validity (or lack thereof) of the Gregorc Style Delineartor

I've always been a skeptic of some of the popular learning style instruments that seem to "catch on" in education. Recently I ran across an empirical study of the validity of the Gregorc Style Delineator.  This study, the abstract (and link to article) which is provided below, failed to support the structural validity of this "pop"-ular instrument.  The abstract and article speak for themselves.

However, having been subjected to these type of instruments while working in the public schools (the early part of my career), I did see some positive benefits from the "process" of taking and discussing the results among staff.  The process often made educators more aware of individual differences, a topic that could then be expanded upon as a consultant.

Reio, T. & Wiswell, A. (2006). An Examination of the Factor Structure and Construct Validity of the Gregorc Style Delineator.  Educational and Psychological Measuremen, 66(3), 489-501 [click here to view]

Abstract
  • Cognitive learning style instruments continue to enjoy widespread use in educational and occupational contexts by researchers and practitioners alike, despite widespread questions about the valid measurement of cognitive learning styles. This investigation examines the psychometric properties of one well-known measure of cognitive learning style—the Gregorc Style Delineator (GSD). The GSD was administered to 467 undergraduate and graduate students (49% female) from two major colleges in the mid-Atlantic region of the United States. Using confirmatory and exploratory factor-analytic techniques, Gregorc’s channel theory was investigated as deployed and represented by the instrument. The Cronbach’s alpha coefficients, ranging from .54 to .68, were appreciably lower than those reported by the instrument’s developer. The data disconfirmed both the two- and four-factor confirmatory models. In the post hoc exploratory factor analyses, many of the factor pattern/structure coefficients were ambiguously associated with two or more of the four theoretical channels as well. Overall, there was little support for the GSD’s theoretical basis or design and the concomitant accurate portrayal of one’s cognitive learning style.


Technorati Tags: , , , , , , , , ,

powered by performancing firefox