Yes. Here is what you can do:
- Start the analysis with the data in the language that you are the most comfortable with. This data would be in a separate file. Do the analysis (create the model) by merging, grouping, or ignoring. and accepting synonym suggestions. Pin the important topics and groups.
- Go through all the pinned groups and topics and translate their (main) terms to their corresponding terms in each of the other languages from which you also have data. Translate these as carefully as possible. After you are done you should have lists of topics and groups with their English names but with terms translated into each target language.
- Get a native speaker for one of your other languages and have them do the analysis in that language. You can do this on a separate file with data in that language. As a gold standard give them the translation of groups and topics for that language. Their task will now be to do analysis (with all of the operations mentioned above) while trying to recreate the topics and groups on their gold standard sheet using the translated terms and any new synonym suggestions that they can find. They should make sure that they label topics and groups in English exactly as they were labeled on the gold standard sheet.
- Repeat for all the other languages.
- Now you have a collection of projects, one for each language, with a common model with labels in English but with the terms in each native language. You can now export each project to Excel and in each Excel file you will have a summary sheet which summarizes the strengths of each topic for each respective language. Copy these values to a new Excel file where you can compare them.