BAYESIAN ANALYSIS IN A SPEEDED ITEM RESPONSE MODEL

We develop a Bayesian analysis of the model proposed by Goegebeur, de Boeck, Wollack and Cohen (2008) to work with speededness by considering a gradual process change IRT model. Prior specification and MCMC algorithm are discussed. Simulations studies are conducted to evaluate his performance to the recover the parameters by considering independence or dependence in examinee´s parameter or personal latent variables by using independent priors. A real-data example, EGRA test, is analyzed and show how the model can be used to identified personal latent variables associated with response in Speededness situations which is consistent with the results in the test.