6533b82afe1ef96bd128b678

RESEARCH PRODUCT

Double validation of translated questionnaires in international studies

subject

description

Background Study questionnaires developed in one country can be translated and used for similar studies in other countries. The main problem of this process is that people speaking different languages can understand the meaning of some questions differently due to disparate mentalities and backgrounds. To avoid a misunderstanding, to perform correct conclusions in accordance to the mentality of each country, and to permit the comparison of results in different countries, the intellectual validation of questionnaires is widely used. The main problem of such a validation is its qualitative nature that can be biased by the subjective approach of the investigator. Therefore, there is a necessity of some objective quantitative methods for better assessment of results of validation. Purpose The objective of the current study was to perform an intellectual validation with subsequent quantitative assessment of questionnaires translated from German to Latvian and Russian in a sample of medical personal in Latvia. Material and methods We obtained the German version of questionnaires on beliefs about older workers, quality of life, and workability from the main investigator - the Centre of Health and Society, University of Dusseldorf, Germany. Questionnaires were translated into the two languages that are mainly used in Latvia: Latvian and Russian. The analysis of translated versions was performed in two stages: qualitative and quantitative. Four focus groups, two groups for each language, were involved in the qualitative stage. Each focus group contained 4 participants (16 participants altogether, 8 for each language), all with medical education. The mean time of discussion in each focus group was 70 minutes. The interpretation of each question was discussed within focus groups, and for those questions that were interpreted differently (both within and between focus groups) four different variants of understanding were formulated. At the second stage, we performed a quantitative analysis of collected equivocal questions. Forty-five participants, all medical workers, were enrolled in this stage. Each participant marked one of the four variants of questions that were formulated after the qualitative stage - the mostly close for their understanding of the initial question. We assessed the most frequently marked variant and performed Chi 2 test for differences between participants with different mother tongues. Results During the qualitative stage, we identified 19 questions that were interpreted differently. In the quantitative stage, we obtained 25 Latvian and 20 Russian-speakers answers on the variant of the question closest to their understanding. In both, Latvian and Russian-speakers’ answers, the same variant was chosen in 10 of 19 questions. Significant differences between Latvian and Russian-speakers was observed in 4 out of 19 questions (Chi 2 test, P Conclusions Double intellectual validation of questionnaires translated from foreign languages is recommended for better interpretation of the study results.