An Experimental Investigation of the Relationship Between Difficulty and Reliability in Tests
Abstract
The need of valid and reliable instruments in mental measurement becomes ever more apparent as the value of such instruments becomes more widely appreciated. The danger resulting from the magnitude of errors of measurement due to low test reliability is a matter of concern when tests are used for individual diagnosis, particularly when interpretation is left in uncritical hands. This investigation, although theoretical in nature, concerns itself with an extremely practical aspect of test construction, the means of increasing the reliability of objective type tests. The study may be best described as an attempt to arrive at satisfactory answers to the questions–what relationship, if any, exists between the difficulty of a test and its reliability? Further, if a relationship is indicated, to what factors may it be ascribed and how may it be utilized to the end of higher reliability?
Subject Area
Educational tests & measurements|Educational psychology|Education
Recommended Citation
Voss, Harold Anthony, "An Experimental Investigation of the Relationship Between Difficulty and Reliability in Tests" (1937). ETD Collection for Fordham University. AAI29282565.
https://research.library.fordham.edu/dissertations/AAI29282565
