Sufficient Statistics and Latent Trait Models
Open Access
- 1 March 1977
- journal article
- Published by Cambridge University Press (CUP) in Psychometrika
- Vol. 42 (1) , 69-81
- https://doi.org/10.1007/bf02293746
Abstract
For questionnaires with two answer categories, it has been proven in complete generality that if a minimal sufficient statistic exists for the individual parameter and if it is the same statistic for all values of the item parameters, then the raw score (or the number of correct answers) is the minimal sufficient statistic. It follows that the model must by of the Rasch type with logistic item characteristic curves and equal item-discriminating powers.This paper extends these results to multiple choice questionnaires. It is shown that the minimal sufficient statistic for the individual parameter is a function of the so-called score vector. It is also shown that the so-called equidistant scoring is the only scoring of a questionnaire that allows for a real valued sufficient statistic that is independent of the item parameters, if a certain ordering property for the sufficient statistic holds.Keywords
This publication has 5 references indexed in Scilit:
- CONDITIONAL INFERENCE FOR MULTIPLE‐CHOICE QUESTIONNAIRESBritish Journal of Mathematical and Statistical Psychology, 1973
- Sufficient Statistics and Discrete Exponential FamiliesThe Annals of Mathematical Statistics, 1972
- Estimating Item Parameters and Latent Ability when Responses are Scored in Two or More Nominal CategoriesPsychometrika, 1972
- Sufficiency and Exponential Families for Discrete Sample SpacesJournal of the American Statistical Association, 1970
- Sufficiency and Statistical Decision FunctionsThe Annals of Mathematical Statistics, 1954