Naslov
Different approaches to cross-cultural validation

Autori
Pilepić Ćosić, Ana ; Mohorić, Tamara ; Takšić, Vladimir

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
Knjiga sažetaka / Jelić, Margareta ; Tomas, Jasmina - Zagreb : Odsjek za psihologiju Filozofskog fakulteta Sveučilišta u Zagrebu, 2019, 29-29

Skup
24. međunarodni psihologijski znanstveni skup: Dani Ramira i Zorana Bujasa (DRZB 2019)

Mjesto i datum
Zagreb, Hrvatska, 11.04.2019. - 13.04.2019

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
cross-cultural validation, measurement equivalence, DIF

Sažetak
Cross-cultural validation refers to whether measures that were originally generated in a single culture are applicable, meaningful, and thus equivalent in another culture (Matsumoto, 2003). Equivalence is by far the most important concept that researchers need to be aware of when conducting cross-cultural research. It is defined as a state or condition of similarity in conceptual meaning and empirical method between cultures that allows comparisons to be meaningful. Perhaps the most important area in which equivalence needs to be ascertained is measurement. Measurement invariance or equivalence is present if there is no item bias, which is a situation when a test item unfairly favors one group of examinees over another. A biased item will exhibit differential item functioning (DIF). DIF occurs when examinees from different groups with equal knowledge exhibit different probabilities of success on an item. Methods used to determine differentially functioning items are numerous and can be roughly divided into two conceptual frameworks: classical test theory and item response theory. The most basic DIF indices are based on classical test theory and analysis of variance and are no longer recommended for use. Item response theory (IRT) framework is rich with different DIF methods. They all share the use of a matching variable which is an estimate of latent ability rather than the observed score. The general framework involves estimating item parameters separately for the reference and focal groups and then comparing them. Third group of methods are the so-called contingency table methods which range from the relatively simple chi-square based Mantel- Haenszel approach to logistic regression. The advantages and disadvantages of each of the three groups of methods will be discussed and illustrated using the data on emotional intelligence measures collected in samples of different cultural origin.

Izvorni jezik
Engleski

Znanstvena područja
Psihologija

POVEZANOST RADA