Résumé :
|
[BDSP. Notice produite par INIST-CNRS 01d9MR0x. Diffusion soumise à autorisation]. The aim of this study was to investigate whether names (surnames and the combination of surnames and first names) are adequate to identify Turkish, Moroccan and Surinamese children in a Dutch population. Parents visiting the Emergency Department of the Sophia Children's Hospital with their child filled in a questionnaire about their ethnic background, socio-economic factors and cultural factors. Patients'names were classified by students with a Turkish, Moroccan and Surinamese (Hindustanic and Creole) background. A combined name-method (surnames in combination with first names) and surnames only were compared with a reference method (ethnicity by parents'country of birth). The validity of both name methods and the measurements of agreement (kappa) of both comparisons were analysed. Turkish names showed good validity measurements between the combined name-method and the reference method (sensitivity of 81% and a positive predictive value of 86%). The Arabic names were distinguishable with a sensitivity of 77%, specificity of 99% and positive predictive value of 87%. Surinamese names were less easy to distinguish from the other names (69,96 and 69% respectively). Overall, the measurement of agreement showed a K of 0.69 in the comparison between the classification of the combined names and the classification by parent's country of birth. The classification of the combined names was slightly better than the classification by surnames only. Conclusions : Names can be used to identify Turkish and Arabic (Moroccan) children in a routinely registered database of Dutch children. For Surinamese names this method is less suitable.
|