Hi AnjaM! Welcome to the forums, and to UX. =)
I've not done any testing that deliberately worked with another culture. Other people might have more experience in this area?
I have done some work on projects that are pretty different contexts to my own. For example, I was working on a project that had a significant Japanese audience. I've never been to Japan or had the opportunity to properly understand their culture, but the sessions were in English, so I was taken aback when the first round of testing didn't do very well; I was confused by the findings and didn't get a good sense of mutual communication, despite good rapport and good participation in what I thought were clear expectations of speaking thoughts aloud, etc. I later realised that I was assuming the participants were explicit communicators (divulging information beyond the topic at hand based on the idea that the listener isn't aware of background or correlating issues). Implicit communicators, such as the Japanese, tend to assume the opposite. Once I'd worked out my problem I was able to continue testing by doing some activities with participants beforehand set the scene differently and better encouraged them to share.
For cross-cultural testing, I assume some of the same principles would apply?
* It's much harder to understand context when more of it is alien to you
* Avoid stereotyping, which is easier to fall into when you're floundering in new contexts.
* Rapport and establishing a relationship will be more important than usual
* Addressing fears and concerns upfront will also be pretty important
* Watch your assumptions. Suspend any judgements that aren't specifically related to the research
* Make sure that you can clearly communicate the requirements to the participant
* Be mindful of cultural taboos, inappropriate topics and the way race, gender and class might affect things
* Talking about previous activities and behaviour might be couched in modesty more than Western participants might exhibit
* Verbal and non-verbal cues are likely to be different, so be mindful.
* Use straightforward language. Metaphors, proverbs and colloquialism's are likely to confuse
* Be aware of the need for extra time - most people might translate the English into their native language to understand better before making a response.
It's a mutual and collaborative process and so also goes the other way. Participants may be reluctant or make their own assumptions about intrusion based on their reading of you or your role as a researcher. Your attempts to avoid bias when commenting may also be more difficult.
Given the increased potential for misreading signals or behaviour, the need to have a second person scribing notes is even more important than usual. A second perspective might help pick up things you miss.
If you're doing a lot of this, it sounds like it would be worth getting some language and cultural training so you can be competent and sensitive to a participants background?
Does the product you're working on provide a translated version of the original content?
Does anyone else have experience in cross-cultural interviewing? Or anything that could be useful to AnjaM?