A Bayesian multidimensional IRT approach for the analysis of residents' perceptions toward tourism

Abstract

In this study, a Bayesian multidimensonal item response theory (IRT)model assuming the presence of correlated general and specic latent traitsis proposed for the investigation of residents' perceptions toward the tourismindustry. Data collected in 2012 in the Italian Romagna area were used tostudy the perceived benets and costs related to tourism. By using posteriorpredictive model checks and Bayesian deviance, the additive IRT model wasfound to t the data well. More importantly, the results could be interpretedmeaningfully, showing the most and least important perceived advantagesand disadvantages of tourism for the local community. Finally, thanks to thecompensatory structure of the model, the dierent inuence of the overallattitude and the specic perceptions of the respondents could be investigatedfor each aspect included in the questionnaire.