Despite seeming like a straight forward process, I receive different advises when I want to translate and use a tool for research. I'd like to know the correct or most acceptable process, preferably in a step by step explanation. Also, if the tool is already validated in the original language, do I still need to revalidate the translated version as now it will be used with a different population? Or, can I simply cross validate the two versions by asking bi-lingual respondents to fill it in on two different occasions (e.g. two weeks apart)?
Here are some useful resources
http://www.naopindia.org/sites/default/files/ITC%2520Guidelines%2520-%2520Translating%2520and%2520Adapting%2520Tests%2520-%2520v2%25203.pdf
My recommendation would be back-translation and revalidation with the translated version using the population you are developing it for. Cultural or language differences could have substantial implications for the factor structure, validity and reliability
http://www.psyktestbarn.no/cms/ptb_mm.nsf/lupgraphics/ITC%2520guidelines.pdf/%24file/ITC%2520guidelines.pdf
http://www.who.int/substance_abuse/research_tools/translation/en/
Forward-backward method is the most recommended by expert to cross-cultural adaptation of instruments. In addition I recommend you to follow the COSMIN (consensus for instruments validation) checklist. Maybe this can be useful:
-http://www.who.int/substance_abuse/research_tools/translation/en/
-Beaton DE, Bombardier C, Guillemin F, Ferraz MB. Recommendations for the cross-cultural adaptation of health status measures. Rosemont, IL: American Academy of Orthopedic Surgeons; 2002.
-http://www.cosmin.nl/
Agree, Leonhard,
Measurement invariance, not mere structural invariance, is the crucial evidence.
Trevor
You have been provided with sound advice by others. I can only reinforce their views saying that yes, revalidation is necessary.
Best wishes with your study,
John
Get a subject matter expert to check the translation as well as going through the forward-backward translation method. If possible, get a subject matter expert who can speak both languages to ensure that it means the same thing in both languages - sometimes the translations can work perfectly in terms of forward-backward translations but actually mean something slightly different in each language. The translation will need to be analysed for reliability, validity and possibly re-normed as validation studies rarely transfer across languages.
The short answer to your question is yes, it does need to be re-validated. This is because validation and reliability analyses are for a specific population and when you are translating a protocol or instrument, it is clearly for use with a different population. Also, the other people who answered this are entirely correct about translating the instrument and then having someone else translate it back to check the translation. A final caution is that any protocol or measurement tool used clinically is also based on the culture of the group being assessed. Thus, most tools must be adapted rather than just translated. This is one of the main reasons you need to re-validate with the new population.
Really grateful for everyone who took time to respond. Your answers are very helpful.
- Badges
- Science topic
- Similar topics
- Psychology
- Psychometrics