Non ti piace? Non importa! Puoi restituire gli articoli fino a 30 giorni
Non puoi sbagliarti con un buono regalo. Con il buono regalo, il destinatario può scegliere qualsiasi prodotto della nostra offerta.
Fino a 30 giorni per il reso
Many educational tests use testlets as a way of §providing context. When testlet effect and item §idiosyncratic features of individual items are both §considered to be the reasons of DIF in many §educational tests using testlets, it is interesting §to investigate the phenomena of DIF amplification §and cancellation resulting from the interactive §effects of these two factors. This research §presented a study based on a multiple-group §testlet item response theory model to examine in §detail different situations of DIF amplification §and cancellation at the item and testlet level using §testlet characteristic curve procedures with signed/ §unsigned area indices and logistic regression §procedure. The testlet DIF model was estimated using §a hierarchical Bayesian framework with the Markov §Chain Monte Carlo method. The simulation study §investigated all of the possible conditions of DIF §amplification and cancellation. Real data analysis §indicated the existence of testlet effect and its §magnitudes of difference on the means and/or §variance of testlet distribution between manifest §groups such as gender or ethnicity.
Ciao! Sono Libroamiko, il tuo consulente di libri.
Come posso aiutarti?