|Institution:||University of Pittsburgh|
|Full text PDF:||http://d-scholarship.pitt.edu/31405/1/Leventhal_Dissertation.pdf;http://d-scholarship.pitt.edu/31405/|
More robust and rigorous psychometric models, such as multidimensional Item Response Theorymodels, have been advocated for survey applications. However, item responses may be influencedby construct-irrelevant variance factors such as preferences for extreme response options. Throughempirical and simulation methods, this study evaluates the use of the IRTree Model, themultidimensional nominal response model, and the modified generalized partial credit modeldesigned to account for extreme response tendencies. The modified generalized partial creditmodel was found to have the best overall fit in terms of test-level, item-level, and person-levelposterior predictive model checks performed. Estimation of this model also resulted in the lowestmean squared error between observed total score and expected total score. The multidimensionalnominal response model had the lowest deviance information criterion among the three models.The empirical study, data validation from the simulation study, and the simulation results providedevidence that the IRTree Model was measuring a unique construct-irrelevant variance factorcompared to the two other methods. For all simulation conditions of sample size (500, 1000),survey length (10, 20), and number of response options (4, 6), the modified generalized partialcredit model had the most adequate model fit with respect to mean item mean squared error. Themultidimensional nominal response model was found equally suitable for surveys measuring onesubstantive trait when responses to 10 4-option forced-choice Likert-type items were explored.