Hostname: page-component-745bb68f8f-kw2vx Total loading time: 0 Render date: 2025-01-07T18:02:04.140Z Has data issue: false hasContentIssue false
  • English
  • Français

An Index of Item Validity Providing a Correction for Chance Success

Published online by Cambridge University Press:  01 January 2025

A. P. Johnson
A. P. Johnson*
Affiliation:
Purdue University
Get access Rights & Permissions [Opens in a new window]

Abstract

The KG Index described below is proposed for evaluation as one approach to the problem of providing an index giving comparable values for items (1) of equal discriminative power at all levels of difficulty and (2) of different numbers of alternative responses.

Type
Original Paper
Information
Psychometrika , Volume 12 , Issue 1 , March 1947 , pp. 51 - 58
Copyright
Copyright © 1947 The Psychometric Society

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

* Votaw, D. F. Notes on validation of test items by comparison of widely spaced groups. J. educ. Psychol., 1934, 25, 185-191.

† The writer uses hereafter the symbol K for the number of the total test sample who are estimated to know the correct answer.

* Lee, J. M. and Symonds, P. M. New type or objective tests: A summary of recent investigations (October 1931-1933). J. educ. Psychol., 1934, 25, 161-184.

† Guilford, J. P. The determination of item difficulty when chance success is a factor. Psychometrika, 1936, 1, 259-264.

* As Votaw indicates, the value N/n or in this instance K/n may well vary, as estimated by. the formula S. E. of . According to the probability tables for the normal curve of error, in 9973 cases out of 10000, K/n should not vary beyond ± 3 . It is thus possible that values of less than 1/n may occur.

† Idem.

‡ Idem.

* The use of these suggested groups is believed to provide sufficient accuracy while avoiding the necessity of computing different base groups for each item.

* The suggested base group values of this table have been arrived at on the basis of both theoretical and practical considerations: except for a very few special test construction situations it is expected that they will prove quite satisfactory.

The values of R have been derived from the basic formula K = (nR – N/ (n – 1) solved for R namely, R = (n – 1) K/n + N/n.

* Guilford, J. P. The phi coefficient and chi square as indices of item validity. Psychometrika, 1941, 6, 11-19.