IRT Goodness-of-Fit Using Approaches from Logistic Regression

摘要: We present an IRT goodness-of-fit framework based on approaches from logistic regression. briefly elaborate the formal relation of models and regression modeling. Subsequently, we examine which model tests indices can be meaningfully used for IRT. The performance casewise deviance, a collapsed Hosmer- Lemeshow test is studied by means simulation that compares their power to well known tests. Next, various R2 measures are discussed in terms interpretability appropriateness within context. By treating as classifiers, several additional such hit rate, sensitivity, specificity, area under ROC curve defined. Data stemming social discomfort scale demonstrate application these statistics.