Dakota McCoy, a PhD student in the Department of Organismic and Evolutionary Biology at Harvard, and I published an article in Scientific American arguing that the "gender-equality paradox" is no paradox at all. You can read the full article here.
As part of that work, we re-analyzed some data from Nosek et al. (2009) and found that implicit attitudes toward women in science do not correlate with gender equality, as measured by the Gender Gap Index: data, R code.
UPDATE: Our analysis was pretty quick and dirty on a small amount of data, and we're thankful to David Miller for sharing this paper and data with us. His data confirm the same trend we saw: implicit attitudes do not correlate with the Gender Gap Index. Furthermore, implicit and explicit attitudes predict the proportion of women in tertiary STEM education.