Measurement

  • William Miller
Chapter

Abstract

Classical item analysis is used to estimate the reliability of test scores obtained from measures of subjects on some attribute such as achievement, aptitude or intelligence. In classical test theory, the obtained score for an individual on items is theorized to consist of a “true score” component and an “error score” component. Errors are typically assumed to be normally distributed with a mean of zero over all the subjects measured.

Keywords

Covariance Posit 

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  • William Miller
    • 1
  1. 1.Iowa State UniversityAmesUSA

Personalised recommendations