Coding effects in comparative research on definitions of health

Journal/Book: Eur J Public Health. 1997; 7: Great Clarendon St, Oxford, England OX2 6DP. Oxford Univ Press. 364-372.

Abstract: In the frame of a comparative study of 'expressive' and 'instrumental' dimensions of the definition of health, a comparison was made between populations in The Netherlands and Canada (Toronto, Ontario), The open-ended question 'What is according to you the best definition of health?' was answered by 776 respondents in The Netherlands and 785 in Canada, In both countries the answers were coded by 3 coders, In order to safeguard the comparability between the 2 sets of data, the 2 groups of coders received an 'identical' coder training, The definitions of health among the Canadian respondents appeared to be more instrumental than among the Dutch, This difference remained constant through ail the subgroups in terms of sex, age, education and income, However, a detailed qualitative study of the coder training and the coding process showed that this difference was caused by differences in the coding practices rather than by differences between the respondents' answers, An implication of this finding is that quantitative studies must entail a detailed qualitative assessment of their own procedures.

Note: Article Hak T, Univ Liverpool, Dept Primary Care, Whelan Bldg, Liverpool L69 3GB, Merseyside, ENGLAND

Keyword(s): coder training; coding; comparative research; definitions of health; qualitative method

Zurück | Weiter

Coding effects in comparative research on definitions of health - A qualitative validation study