Higher order factor structure of a self-control test: Evidence from confirmatory factor analysis with polychoric correlations

David B. Flora*, Eli J. Finkel, Vangie A. Foshee

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

39 Scopus citations

Abstract

The self-control test developed by Grasmick, Tittle, Bursik, and Arneklev was designed to measure each of six components of self-control, namely, impulsivity, a preference for simple rather than complex tasks, risk seeking, a preference for physical rather than cerebral activities, a self-centered orientation, and a volatile temper. This conceptualization clearly suggests that self-control may be defined as a higher order construct that leads to each of these components, which in turn may be represented as first-order factors or constructs. However, due to various limitations, previous analyses of the test failed to establish this factor structure. By employing proper methods for the factor analysis of Likert-type items and explicitly testing a higher order structure, the authors show that the self-control test may provide more valid measurement of the psychological constructs it was intended to measure than previous research suggests.

Original languageEnglish (US)
Pages (from-to)112-127
Number of pages16
JournalEducational and Psychological Measurement
Volume63
Issue number1
DOIs
StatePublished - Feb 2003

Keywords

  • Factor analysis
  • Higher order models
  • Self-control
  • Weighted least squares

ASJC Scopus subject areas

  • Education
  • Developmental and Educational Psychology
  • Applied Psychology
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'Higher order factor structure of a self-control test: Evidence from confirmatory factor analysis with polychoric correlations'. Together they form a unique fingerprint.

Cite this