I don't see why not as long as the teachers understood English. The CFA results would tell you (to a certain extent) whether the untranslated scale worked OK for your sample. For example, if the expected factor model fits the data well and if the indicator reliabilities (R-squared values) are decent (showing small measurement error variances) this would indicate that the scale worked fine.

Christian Geiser , thank you for your insight. Is there any reference available to use CFA for scales that i will adopt in my cultural context without translation?

Yes, you can still use Confirmatory Factor Analysis (CFA) if you are using an international scale in its original English language with Pakistani teachers, without translating it into Urdu. However, there are a few points to consider. First, ensure that the participants (Pakistani teachers) are proficient in English, as the scale will be in English. If the teachers do not have sufficient understanding of English, it could affect the validity of their responses. Second, even if the scale is in English, be mindful of potential cultural differences in how certain items or concepts are interpreted by the teachers. If any item is culturally biased or unclear, it may affect the factor structure or the overall validity of the CFA. Third, while you do not need to translate the scale into Urdu, it’s important to confirm that the scale is measuring the same construct for Pakistani teachers as it does for the original population. You could perform a measurement invariance analysis to test if the factor structure holds across the two different populations (i.e., the original population and Pakistani teachers). Lastly, it’s important to check the validity of the scale in the new context (i.e., Pakistani teachers) through pre-testing or pilot testing, ensuring that the scale is suitable and the results obtained from CFA are meaningful. If these considerations are met, you can proceed with CFA without the need for translation.