Una aplicación de valores plausibles a la calificación de pruebas estandarizadas vía simulación

The use of Plausible Values in large scale standarized tests, develop the role of imputation when the reference framework is too large and each person can not adress the totality of the items in the production. In that case appleal to block design definition that ensures an appropiate share of indivi...

Full description

Autores:
Córdoba Perozo, Michel Felipe
Tipo de recurso:
Fecha de publicación:
2016
Institución:
Universidad Santo Tomás
Repositorio:
Universidad Santo Tomás
Idioma:
spa
eng
OAI Identifier:
oai:repository.usta.edu.co:11634/6435
Acceso en línea:
https://revistas.usantotomas.edu.co/index.php/estadistica/article/view/2643
Palabra clave:
valores plausibles, imputación de datos, rasgo latente, pruebas estandarizadas .
Rights
License
Copyright (c) 2016 Comunicaciones en Estadística
Description
Summary:The use of Plausible Values in large scale standarized tests, develop the role of imputation when the reference framework is too large and each person can not adress the totality of the items in the production. In that case appleal to block design definition that ensures an appropiate share of individuals per item aiming that the framework is addressed smugly across the study population is necessary. In general the imputation methos consist in find the posterior distribution of the feature latent that is associated to the individual hability, by weighting distribution that induces model item response theory and regression associated with some latent variables measured in the individual. This paper shows an example of simulation where you can easily see the advantages offered by the method in the aggregate results of this particle scheme application