Premium
ITEM ANALYSIS BY PROBIT AND FRACTILE GRAPHICAL METHODS
Author(s) -
Das Rhea S.
Publication year - 1964
Publication title -
british journal of statistical psychology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.157
H-Index - 51
eISSN - 2044-8317
pISSN - 0950-561X
DOI - 10.1111/j.2044-8317.1964.tb00243.x
Subject(s) - probit model , probit , statistics , econometrics , ordered probit , regression analysis , test (biology) , mathematics , regression , selection (genetic algorithm) , computer science , artificial intelligence , paleontology , biology
This paper considers five basic questions of item analysis and describes an integrated approach for obtaining the answers which utilizes the methods of probit and fractile graphical analysis. Item difficulty is interpreted as the limen at which 50 per cent of the subjects pass; discriminating power is given by the probit regression coefficient; the item‐ability relation is specified by the probit regression equation; chi square provides an objective test of whether or not the item meets the normal ogive assumption; and plotting items in terms of the limen and regression coefficient permits the selection of items comparable in terms of ability, difficulty, and discrimination. Techniques for the practical application of this approach are also outlined and illustrated.