I am working on a project where I need to use a scale to measure certain psychological and political positions. I have a questionnaire that is used in the US, and whose validity was tested there quite well, but I need to use the scale in India.

I want to test the validity of the instrument in the Indian context, before using it for my study. I wanted to know the steps for testing different kinds of validity and reliability, the sample size required etc. Links to any good material to study these things (books/ video lectures/ notes/ other materials) would be of help, along with any comments or recommendations on what to do from your side.

  • $\begingroup$ If you want to provide more detail, like the scale you're interested in using, that might help. $\endgroup$
    – pep
    Jul 12, 2021 at 17:33

1 Answer 1


The first step is translation. Some projects have used a three-step process including 1) forward translation, 2) back translation, and 3) cultural adjustment. Two translators would complete step one, two more would complete step two, and then the materials are sent to external people (preferably non-academics) for review. Here's (1) a preprint that describes that process.

Beyond that, the process is very much dependent on the measure being validated. There are procedures to assess factor structure, measurement invariance, reliability, convergent validity, discriminant validity, predictive validity, etc. This is a huge field of study and there's no way to provide an answer that would apply to any scale. I would not assume that the measure was rigorously validated in the source in which you found it, unless you have the expertise to evaluate their methods and procedures.

I might try searching (or posting) on cross-validated to get information specific to the statistical approach.

I would think about whether you really want to conduct validation studies, or simply translate the measure as best you can. Validation studies will take a long time and they're quite complex. Here's (2) one example of a scale that very much needed to be validated in another language, because of the substantive differences in the translation.

Here is a video with someone talking about the EFA/CFA procedures. I have not watched it and because it's such a complicated topic, I doubt they cover everything you need to know in the 25 minutes.

Here are some powerpoint slides that have more information from Classical Test Theory. This is only one possible way to approach psychological measurement. Another approach is Item Response Theory.

  1. Forscher, P. S., Paris, B., Primbs, M., & Coles, N. A. (2020). PSACR: The Psychological Science Accelerator's COVID-19 Rapid-Response Project.

  2. Seema, R., Quaglia, J. T., Brown, K. W., Sircova, A., Konstabel, K., & Baltin, A. (2015). The Estonian Mindful Attention Awareness Scale: Assessing mindfulness without a distinct linguistic present tense. Mindfulness, 6(4), 759-766.

  • $\begingroup$ Nice answer but you are assuming zero knowledge in psychometrics and scale validation which is a bit patronising. With your assumption I think it would be better to recommend OP to talk to someone with experience in scale validation or follow a course in psychometrics. $\endgroup$
    – Aolon
    Jul 12, 2021 at 7:33
  • $\begingroup$ I do think it would be best to talk to someone with expertise! Thanks for clarifying! $\endgroup$
    – pep
    Jul 12, 2021 at 16:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.