tl;dr the FT/HN reported data may have flaws; based on a slightly different data set TFA finds: - Since the 1990s, both men and women have become more likely t

tl;dr the FT/HN reported data may have flaws; based on a slightly different data set TFA finds: - Since the 1990s, both men and women have become more likely to identify as liberal. - Women are more likely to identify as liberal by 5-10 percentage points. - There is no evidence that the ideology gap is growing. (in particular for the last point, TFA points out the wide error bands resulting from noisy data)