[Rain Man]

As someone who asks this question a lot on surveys, I think it's asked too often, just by force of habit. If a client wants it and it's not obvious why, I'll ask them a couple of questions:

1. Do you have any reason to believe that there's a noticeable difference?

2. If there is a difference, is it going to make a difference in your ad campaign/ program/ policy/ whatever else is being tested?

The answer to Question 1 isn't necessarily a question killer, but if they answer no to Question 2, we don't ask it.

I will say, though, that there are measurable differences in opinions and behaviors by race/ethnicity, and sometimes it's useful to know that. Why? Because a lot of minority populations have specific media preferences that differ from that of the majority population, and it might pay for you to buy some media to target them if they're well below the average.

I think the key is to view it as one more demographic variable to help understand the audience, just like recording age or gender or income. The key, though, is thinking through the two questions above before including it out of habit.