An emerging issue in cross-cultural research methodology concerns how to test the invariance of measuring instruments used. Item bias refers to a systematic difference in the interpretation of scale items in different groups. In this study, we present a method for the analysis of item bias developed by van de Vijver and Leung (1997). The method uses Analysis of Variance to identify items showing uniform and nonuniform bias. We illustrate this procedure using a study into family processes in Italy and Great Britain. We analysed the cross-cultural invariance of the items of a scale of family enmeshment (Bloom, 1985). The presence of bias in various items shows important implications for the definition of enmeshment in countries with different family cultures.
Manzi, C., Lanz, M., Vignoles, V. L., Come si verifica l'invarianza cross-culturale degli strumenti? Un esempio nell'ambito degli studi sulla famiglia, <<TPM. TESTING PSICOMETRIA METODOLOGIA>>, 2005; (12): 261-276 [http://hdl.handle.net/10807/12968]
Come si verifica l'invarianza cross-culturale degli strumenti? Un esempio nell'ambito degli studi sulla famiglia
Manzi, Claudia;Lanz, Margherita;Vignoles, Vivian Laurence
2005
Abstract
An emerging issue in cross-cultural research methodology concerns how to test the invariance of measuring instruments used. Item bias refers to a systematic difference in the interpretation of scale items in different groups. In this study, we present a method for the analysis of item bias developed by van de Vijver and Leung (1997). The method uses Analysis of Variance to identify items showing uniform and nonuniform bias. We illustrate this procedure using a study into family processes in Italy and Great Britain. We analysed the cross-cultural invariance of the items of a scale of family enmeshment (Bloom, 1985). The presence of bias in various items shows important implications for the definition of enmeshment in countries with different family cultures.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.