Suppressor Variables
I've recently become fascinated by suppressor variables and a couple of interesting ones have appeared in some of my recent studies.
The most common one I've seen in the literature is that of right-wing authoritarianism and religious fundamentalism in predictive models of prejudice (most often two different types: anti-gay and anti-black). In this subset of the literature, we see a shift from the normally positively correlation between religious fundamentalism and prejudice. However, when controlling for right-wing authoritarianism (which gobbles up much of the unique variance), religious fundamentalism oftens becomes negatively related (or sometimes not significantly related) with anti-black prejudice but not anti-gay prejudice.
What other interesting suppressors have you come across in your research?
The most common one I've seen in the literature is that of right-wing authoritarianism and religious fundamentalism in predictive models of prejudice (most often two different types: anti-gay and anti-black). In this subset of the literature, we see a shift from the normally positively correlation between religious fundamentalism and prejudice. However, when controlling for right-wing authoritarianism (which gobbles up much of the unique variance), religious fundamentalism oftens becomes negatively related (or sometimes not significantly related) with anti-black prejudice but not anti-gay prejudice.
What other interesting suppressors have you come across in your research?