[article]
| Titre : |
Evidence of cross-cultural differences across multiple translations of an autism screening tool |
| Type de document : |
texte imprimé |
| Auteurs : |
Michaela DUBAY, Auteur ; John SIDERIS, Auteur ; Erica ROUCH, Auteur ; Yun-Ju CHEN, Auteur ; Grace T. BARANEK, Auteur ; Linda R. WATSON, Auteur ; Elizabeth CRAIS, Auteur |
| Article en page(s) : |
202738 |
| Langues : |
Anglais (eng) |
| Mots-clés : |
Culture Measurement Invariance Screening Translation |
| Index. décimale : |
PER Périodiques |
| Résumé : |
Background Examining the psychometric alignment of translations of autism assessments with original language versions allows for cross-cultural comparisons and data pooling. This study tested measurement invariance between three versions of a parent-report autism screening tool, the First Years Inventory v3.1(FYI): 1) the original English, 2) a traditional forward-backward (FB) Spanish translation, and 3) a culturally adapted Spanish translation (TCA). Methods Participants were caregivers of children 6-16 months of age in the United States. 5974 native English-speakers completed the English FYI. 506 native Spanish-speakers were randomized to complete one of two FYI Spanish translations, either the FB (N = 257) or the TCA (N = 249). Results We performed measurement invariance testing using multigroup confirmatory factor analysis to evaluate the psychometric equivalence of the Spanish versions to the English version across the three groups. Neither Spanish version was invariant with the English version, however partial models were identified. Analysis of mean differences in factor and item means did not indicate that one Spanish translation was more aligned with the English version than the other. Conclusions Datasets including multiple language versions of an instrument should undergo measurement invariance testing to ensure equivalence between language versions and avoid the risk of making unfounded conclusions. Further research should identify specific translation methods that are sufficient in generating instruments that are valid for both clinical and empirical purposes as neither translation methodology used here maintained psychometric equivalence to the original English version. More robust cultural adaptation procedures may be necessary. |
| En ligne : |
https://doi.org/10.1016/j.reia.2025.202738 |
| Permalink : |
https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=573 |
in Research in Autism > 129 (January 2025) . - 202738
[article] Evidence of cross-cultural differences across multiple translations of an autism screening tool [texte imprimé] / Michaela DUBAY, Auteur ; John SIDERIS, Auteur ; Erica ROUCH, Auteur ; Yun-Ju CHEN, Auteur ; Grace T. BARANEK, Auteur ; Linda R. WATSON, Auteur ; Elizabeth CRAIS, Auteur . - 202738. Langues : Anglais ( eng) in Research in Autism > 129 (January 2025) . - 202738
| Mots-clés : |
Culture Measurement Invariance Screening Translation |
| Index. décimale : |
PER Périodiques |
| Résumé : |
Background Examining the psychometric alignment of translations of autism assessments with original language versions allows for cross-cultural comparisons and data pooling. This study tested measurement invariance between three versions of a parent-report autism screening tool, the First Years Inventory v3.1(FYI): 1) the original English, 2) a traditional forward-backward (FB) Spanish translation, and 3) a culturally adapted Spanish translation (TCA). Methods Participants were caregivers of children 6-16 months of age in the United States. 5974 native English-speakers completed the English FYI. 506 native Spanish-speakers were randomized to complete one of two FYI Spanish translations, either the FB (N = 257) or the TCA (N = 249). Results We performed measurement invariance testing using multigroup confirmatory factor analysis to evaluate the psychometric equivalence of the Spanish versions to the English version across the three groups. Neither Spanish version was invariant with the English version, however partial models were identified. Analysis of mean differences in factor and item means did not indicate that one Spanish translation was more aligned with the English version than the other. Conclusions Datasets including multiple language versions of an instrument should undergo measurement invariance testing to ensure equivalence between language versions and avoid the risk of making unfounded conclusions. Further research should identify specific translation methods that are sufficient in generating instruments that are valid for both clinical and empirical purposes as neither translation methodology used here maintained psychometric equivalence to the original English version. More robust cultural adaptation procedures may be necessary. |
| En ligne : |
https://doi.org/10.1016/j.reia.2025.202738 |
| Permalink : |
https://www.cra-rhone-alpes.org/cid/opac_css/index.php?lvl=notice_display&id=573 |
|  |
Evidence of cross-cultural differences across multiple translations of an autism screening tool in Research in Autism, 129 (January 2025)
