Controls

General Strategies Inclusion/Exclusion Criteria		< back		7 of 25		next >

We can control many extraneous sources of variance by controlling who is in our study. We want our samples to represent the population of interest without introducing undue bias. We know that probability sampling strategies are most likely to give us a representative sample (see Sampling Methods workshop). Even with probability sampling, we might need to develop inclusion and exclusion criteria to insure that we are getting exactly the people we need in our sample.

	For example, you want to do an experiment on stereotype threat and decide to study women and math. The stereotype holds that women are not as accomplished as men in mathematics, especially at higher levels of complexity. The concept of stereotype threat is that members of a group who are stereotyped become anxious when they are placed in a circumstance in which they may be judged by the stereotype rather than their abilities. This anxiety can decrease performance. This threat is particularly strong when the group member is invested in good performance in the domain of the stereotype.