Your dropouts might not be necessarily due to technical reasons, but maybe there are reasons within the sample or the questionnaire itself. Can you see where exactly those dropouts happen?
Reasons I can imagine top of my head that might cause problems with multilingual surveys:
- Approach / language of the questionnaire
- Bad sample quality
- People receive the wrong language as default
In general, we usually try to avoid to run surveys multilingual, but rather have one survey per country/language. Because often you also need to change the structure / questions of the survey a bit, when adapting to different countries, etc.
But of course, multilingual surveys should work. No idea if the size of the session files increases, maybe someone else can help there.
But maybe you can check what the size of these session files currently is, that might be a good indicator already.
Your server specs look pretty good, how is it connected to the internet? What bandwidth does it have?
I answer at the LimeSurvey forum in my spare time, I'm not a LimeSurvey GmbH employee.
No support via private message.
