I agree with you.
We had a presentation from another organization saying that if we sent an all-English survey into a French community, the survey would be useless, and I would agree. I thought it was a lousy example, because whether you send two or two million English surveys into an all-French community, you're going to get a lousy return.
Do you not agree that building biases into the survey is something that...? Your organization tries to remove the bias piece because it automatically affects the results. Is that not correct?