Naslov
Validation of different procedures for the assessment of invariance measurement in cross- cultural studies

Autori
Takšić, Vladimir ; Mohorić, Tamara ; Ćosić Pilepić, Ana ; Molander, Bo ; Holmström, Stefan

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

Izvornik
23. Dani Ramira i Zorana Bujasa - Knjiga sažetaka / Arambašić, Lidija ; Erceg, Inja ; Kamenov, Željka - Zagreb : Odsjek za psihologiju Filozofskog fakulteta Sveučilišta u Zagrebu, 2017, 57-57

Skup
23. Dani Ramira i Zorana Bujasa

Mjesto i datum
Zagreb, Hrvatska, 06.04.2017. - 08.04.2017

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
cross-validation, ESCQ, invariance, differential item functioning (DIF), GMHDIF program

Sažetak
In the last two decades an increased interest in procedures for assessment of invariance emerged, resulting in the development and modernization of procedures. The main goal of cross-cultural studies is comparison of constructs and scores between samples in various countries. Early studies had some common methodological problems: a) cross-cultural similarities and differences were often not statistically tested, b) lack of rationale for selecting countries, c) no check on quality of translation/ adaptation. Application of assessment methods to different languages and cultural backgrounds, without taking in account these problems, can result in questionable cultural comparability. Researchers should give specific attention to structural equivalence and metric invariance of measures. Measurement invariance is related to the degree to which items and constructs have an equivalent meaning for individuals of different cultural backgrounds. Statistical techniques for evaluating DIF items allow us to compare any groups defined by the researcher in order to secure that the members of the compared groups have the same probability of answering the items correctly if they have the same level of latent variable. The Emotional Skills and Competence Questionnaire (ESCQ) has been translated and validated in several languages, and had shown proper psychometric properties. Several previous analyses of DIF items of ESCQ were published (Molander, Holmstrom & Takšić, 2009 ; 2011), using different DIF statistics – e.g. based on ordinal logistic regression or on Mantel-Haenszel methods (through newly developed GMHDIF program ; Fidalgo, 2011). Result from these studies will be presented with an additional examination of methodological issues specific for validation of different statistical procedures for accurate assessment of invariance in cross-cultural.

Izvorni jezik
Engleski

Znanstvena područja
Psihologija

POVEZANOST RADA