Abstract
Traditional methods of test construction have worked rather well for the development and validation of tests. But, in recent years, tests and test-construction procedures have been examined more carefully than at any other time. The author has noted that when one has available knowledge of item-difficulty levels and of correlations of items with the total test score and with the criterion, it is possible to refine tests even better and simpler than we have in the past. A classic textbook example set of data in Gulliksen (borrowed from Mollenkopf) is analyzed to demonstrate the usefulness of the simplified method. The simplified method can greatly shorten a test while maintaining the full range of item difficulties, maintaining test validity, and even possibly raising test reliability.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
Flanagan, J. C. (1936). A short method for selecting the best combination of test items for a particular purpose. Psychological Bulletin, 33, 603–604.
Guilford, J. P. (1954). Psychometric methods. New York: McGraw-Hill.
Gulliksen, H. (1950). Theory of mental tests. New York: Wiley.
Horst, A. P. (1934). Item analysis by the method of successive residuals. Journal of Experimental Education, 2, 254–263.
Wherry, R. J., & Gaylord, R. H. (1946). Test selection with integral gross score weights. Psychometrika, 11, 173–183.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Schuh, A.J. A simplified method of test construction from traditional methods of item analysis. Bull. Psychon. Soc. 22, 566–567 (1984). https://doi.org/10.3758/BF03333909
Received:
Published:
Issue Date:
DOI: https://doi.org/10.3758/BF03333909