Statistical Reasoning in Psychometric Models and Educational Measurement
- 1 March 1989
- journal article
- Published by Wiley in Journal of Educational Measurement
- Vol. 26 (1) , 81-90
- https://doi.org/10.1111/j.1745-3984.1989.tb00320.x
Abstract
This paper explores three points raised by Divgi (1986) in order to explicate the important distinction between deterministic and statistical reasoning in the application of statistical models to educational measurement. The three issues concern the relationship between data and esimation equations, the distinction between parameters and parameter estimates, and the power of tests of fit of responses across the ability continuum.Keywords
This publication has 8 references indexed in Scilit:
- DOES THE RASCH MODEL REALLY WORK FOR MULTIPLE CHOICE ITEMS? NOT IF YOU LOOK CLOSELYJournal of Educational Measurement, 1986
- A Strategy for the Graphical Representation of Rasch Model ResidualsEducational and Psychological Measurement, 1985
- Consequences of Using the Rasch Model for Educational Assessment*British Educational Research Journal, 1979
- Fitting the Rasch model—A heady taleBritish Journal of Mathematical and Statistical Psychology, 1978
- Maximum Likelihood Estimates in Exponential Response ModelsThe Annals of Statistics, 1977
- Conditional Versus Unconditional Procedures for Sample-Free Item AnalysisEducational and Psychological Measurement, 1977
- Sufficient Statistics and Latent Trait ModelsPsychometrika, 1977
- Person and population as psychometric concepts.Psychological Review, 1965