In early September, Quinnipiac University asked Americans how they felt about the nomination of Judge Brett M. Kavanaugh to serve on the Supreme Court. Responses were mixed, as was generally the case in polls conducted at that point. Since then, a lot has happened, including the allegation that Kavanaugh assaulted a woman at a small party in 1982 when he was in high school.

On Monday, Quinnipiac released the results of a poll that began asking Americans their views the same evening as last week’s hearing addressing the allegation. Americans are now more likely to say that they oppose his confirmation, by a six-point margin.

Given how polarizing Kavanaugh’s nomination has become, it’s probably not a surprise that most Democrats oppose the nomination and most Republicans support it. But the shift in opinions by party is interesting. Democrats are 12 points more likely to oppose Kavanaugh than a month ago. Independents, more supportive than not early last month, have now flipped to be more likely to oppose his nomination. While support for Kavanaugh’s nomination increased among Republicans, opposition to his nomination increased more. (The percentage of undecided respondents decreased.)

There are obvious gender divides in views of Kavanaugh’s confirmation, in part because of partisan identity, in part out of concern about his position on abortion and in part, it’s safe to assume, because of the nature of the allegations he faces.

A plurality of women opposed Kavanaugh last month, while a plurality of men supported him. Now, more than half of women oppose his confirmation, an increase of eight points.

Those gender differences, though, aren’t uniform across party identity. Independent men, for example, still support Kavanaugh’s confirmation by a narrow (statistically insignificant) margin. The margin among Republican men narrowed while Democratic men became much more likely to oppose his nomination.

But not as much as Democratic women, who went from opposing Kavanaugh by a 68-point margin to opposing him by 84 points — a net 16-point shift. Among independent women, who were about split on Kavanaugh last month, there’s now a wide split. Part of this is because fewer independent women were polled, making the margin of error larger. But part of it, clearly, is a shift in attitudes.

Republican women, interestingly, didn’t see much change in net support for Kavanaugh’s confirmation.

In fact, Republican women were the only group in which net support for Kavanaugh’s confirmation increased. Otherwise, net support fell. That was particularly true among Democrats and independents, driven by the drop in support among Democratic and independent women.

In the abstract, this isn’t great news for Kavanaugh’s confirmation. An increase in overall opposition is not a trend his supporters would want to see.

But in the context of an election where women are poised to support Democrats over Republicans by historic margins, it should be particularly troubling for Republicans. Men are about where they were before the last month of revelations, but women have shifted against Trump’s nominee. The Supreme Court is cited as the most important issue to midterm voters.

Kavanaugh, subject to a vote not by the American public but by a Republican-led Senate, may well still be confirmed. Republicans have argued that confirming him is necessary to keep Republicans motivated to come out to vote. Quinnipiac’s poll results don’t appear to show the sort of surge in support that might reinforce that claim, but even if the argument is true, it will be interesting to see whether it was worth potentially turning away independent women in particular.