For the record, I love Kentucky, Louisiana, and Utah.
Similarly, a YouGov poll last week found that there is a gap between who married men think their wives will vote for, and how married women plan to vote:
Shy Hillary voters? Discuss.
How is that a shy Clinton effect? A "Shy X" effect is when people are reluctant to tell pollsters their true voting intentions. Here they are telling the pollsters their voting intentions. They're just not telling their husbands.
Presumably some people are answering polls in front of their family members, so there will be some women who don't tell pollsters their true intentions because they don't want their husbands to hear.
However, the opposite effect might still exist, and in some cases it could simply be couples not talking about politics with each other.
So you're saying that the fact that there's a gap between what married women say their voting intentions are and what their husbands think suggests that the gap is even bigger, because there might be additional women who are lying to pollsters, who actually support Clinton?
I don't think there's any way to tell if that's the sense in which there's lying going on. For all we know, the husbands are the ones telling pollsters the truth about their wives' voting intentions, and the discrepancy occurs because some Trump-supporting women don't want to tell pollsters that they support Trump. So this could just as easily indicate a "Shy Trump" effect as a "Shy Clinton" effect. How can we possibly know which way these effects are running from the data at hand?
Like I said, it also depends on whether it's a case of women being afraid of openly disagreeing with their domineering sexist husbands, or simply men assuming that their wives agree with them even though they don't.
Either way, we clearly don't have enough information to know.
