Psychometrika

, Volume 26, Issue 2, pp 239–246 | Cite as

Empirical comparison of item parameters based on the logistic and normal functions

  • Frank B. Baker
Article

Abstract

Maximum likelihood estimates of item parameters of a scholastic aptitude test were computed using the normal and logistic models. The goodness of fit of ogives specified by the pairs of item parameters to the observed data was determined for all items. While negligible differences in the limen values were found, differences in item discrimination indices indicated that interpretation of these indices requires separate frames of reference. The empirical results showed the logistic model to be a useful alternative to the normal model in item analysis.

Keywords

Public Policy Empirical Result Maximum Likelihood Estimate Likelihood Estimate Logistic Model 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    Baker, F. B. Univac Scientific computer program for test scoring and item analysis. Washington, D. C.: Amer. Documentation Inst., Library of Congress, Document 5931.Google Scholar
  2. [2]
    Berkson, J. Why I prefer logits to probits.Biometrics, 1951,7, 327–339.Google Scholar
  3. [3]
    Berkson, J. A statistically precise and relatively simple method of estimating the bio-assay with quantal response, based upon the logistic function.J. Amer. statist. Ass., 1953,48, 565–599.Google Scholar
  4. [4]
    Berkson, J. Tables for the maximum likelihood estimate of the logistic function.Biometrics, 1957,13, 28–34.Google Scholar
  5. [5]
    Birnbaum, A. Probability and statistics in item analysis and classification problems. Efficient design and use of tests of mental ability for various decision-making problems. Ser. Rep. No. 58-16, USAF Sch. Aviation Medicine, Randolph AFB, Texas, 1957.Google Scholar
  6. [6]
    Birnbaum, A. Probability and statistics in item analysis and classification problems. Further considerations of efficiency in tests of a mental ability. Ser. Rep. No. 58-17, USAF Sch. Aviation Medicine, Randolph AFB, Texas, 1957.Google Scholar
  7. [7]
    Birnbaum, A. Probability and statistics in item analysis and classification problems. On the estimation of mental ability. Ser. Rep. No. 15, USAF Sch. Aviation Medicine, Randolph AFB, Texas, 1957.Google Scholar
  8. [8]
    Finney, D. J. The application of probit analysis to the results of mental tests.Psychometrika, 1944,9, 31–39.Google Scholar
  9. [9]
    Finney, D. J. and Stevens, W. L. A table for calculating of working probits and weights in probit analysis.Biometrika, 1948,35, 191–201.Google Scholar
  10. [10]
    Garwood, F. The application of maximum likelihood to dosage-mortality curves.Biometrika, 1941,32, 46–58.Google Scholar
  11. [11]
    Hodges, J. L., Jr. Fitting the logistic by maximum likelihood.Biometrics, 1958,14, 453–461.Google Scholar
  12. [12]
    Lawley, D. N. On problems connected with item selection and test construction.Proc. Roy. Soc., Edinburgh, 1943,61A, 273–287.Google Scholar
  13. [13]
    Layton, W. L. Construction of a short form of the Ohio State University psychological examination. Univ. Minnesota, Student Counseling Bureau, 1957.Google Scholar
  14. [14]
    Lord, F. A theory of test scores.Psychometric Monogr. No. 7, Chicago: Univ. Chicago Press, 1952.Google Scholar
  15. [15]
    Maritz, J. S. Estimation of the correlation coefficient in the case of a bivariate normal distribution when one of the variables is dichotomized.Psychometrika, 1953,18, 97–110.Google Scholar
  16. [16]
    Maxwell, A. E. Maximum likelihood estimates of item parameters using the logistic function.Psychometrika, 1959,24, 221–227.Google Scholar
  17. [17]
    Tucker, L. R. Maximum validity of a test with equivalent items.Psychometrika, 1946,11, 1–14.Google Scholar

Copyright information

© Psychometric Society 1961

Authors and Affiliations

  • Frank B. Baker
    • 1
  1. 1.University of MarylandUSA

Personalised recommendations