Hated by the right people

The campaign for the 2024 Republican presidential nomination is already underway, although the presence of Donald Trump means it's not as open as it would normally be.  Ron DeSantis seems to have moved to the top of the field--Predicit has him at 0.23, with Trump at 0.31 and everyone else below 0.10.  I wondered whether DeSantis's position was slipping given the rise of Covid in Florida, so I looked for figures on the popularity of state governors.  That led me to something that wasn't exactly what I was looking for, but was still interesting--the Covid States project, which as the name suggests collects data on the state level for various things related to Covid.  One of them is approval of how your state governor has dealt with the pandemic.  The data run only to October 2020 (they have another report from March 2021 but the more recent data aren't available for download), but some patterns are visible.  Here is the average level of approval, by respondent's and governor's party:

 

Unsurprisingly Republicans in states with Republican governors (RR) and Democrats in states with Democratic governors (DD) are pretty satisfied.  Republicans in states with Democratic governors (RD) are more satisfied than Democrats with Republican governors (DR).  Independents in states with Democratic governors are more satisfied than independents in states with Republican governors, suggesting that the DR/RD difference is because Republican governors tend to pursue less popular policies, not because Democrats are more partisan.  

The next figure shows the standard deviation of state level approval ratings within each group.

Democrats' ratings of Republican governors have the most variation--some Republican governors got high approval ratings from Democrats and some got very low approval.  Independents' ratings of Republican governors also have pretty high variation.  There's not much difference among the other groups, although it's worth noting that variation in Republicans' approval of Democratic governors declined over the period.

Then taking the approval ratings of Democratic governors among Democrats and Republicans in the last survey (October 2020).

 

and the parallel figure for Republican governors (they are arranged so that approval within their party is on the horizontal and approval from the opposite party is on the vertical axis).

 

As the standard deviations indicate, Republican approval of Republican governors is pretty similar among states.  nineteen out of twenty-six were over 65%, and the maximum was 77.7%.  But Democratic approval of Republican governors ranged from below 10% (Iowa and South Dakota) to over 90% (Vermont).  DeSantis's approval from Republicans was just slightly above average (71% vs. 68%), but his approval among Democrats was among the lowest (14%).  Kristi Noem, who is also getting attention as a possible contender (PredictIt has her at #7) also was about average in approval from Republicans, but second-lowest in approval from Democrats.  Mike Hogan, who had high approval from both, has also been mentioned, but PredictIt has him down at #16, with a price of .01.  Greg Abbott of Texas and Doug Ducey of Arizona have also been mentioned--like DeSantis and Noem, they are distinguished by low rates of approval among Democrats rather than high rates of approval among Republicans.  

So it seems like the key to getting ahead for Republicans is not popularity among Republicans but unpopularity among Democrats.  A recent story on J. D. Vance quotes him as saying "I think our people hate the right people."  The data here show another aspect of the same thing--for Republicans today, the strongest appeal is negative: who you dislike and who dislikes you.

 

 

Stop blaming white women

 The New York Times recently had a piece called "The March of the Karens," which was about "Karen" as "an epithet for a type of interfering, hectoring white woman."  To accompany it, "Carmen Winant created six collages exclusively for T, all of which are titled 'White Women Look Away' (2021). The works are a continuation of her 2016 series of the same name, made in response to that year’s presidential election and to the 'lack of support Prof. Anita Hill received from white women' during her testimony in 1991."  That made me wonder what the racial and gender differences were on that controversy.  I looked at a USA Today poll from Oct 13, 1991, as the Senate hearings were concluding.  It asked "Clarence Thomas has been charged with sexual harassment by Anita Hill, and he has denied the charges.  Who do you believe is telling the truth?"  The results broken down by race (black and white only).

                         Thomas       Hill      Both    Neither     DK

White                  47%         25%         6%     4%          18%

 Black                 47%          20%         8%     7%          18%

and gender:

Men                 49%             22%          5%      5%        19%

Women            45%             26%          6%       4%       19%

Little or no difference either way (the associations were not statistically significant).   But maybe there's an interaction--e.g. maybe support for Anita Hill was highest among black women and lowest among white women?

White M          51%             22%        5%          4%       18%

White F             44%            27%        6%          4%      18%

Black M            45%            21%        7%          8%      18%

Black F             49%            20%         8%          6%      19%

In fact, support for Hill was highest among white women, although once again the differences were not statistically significant.  

I also looked at a survey from 1994, which asked people what they thought about it then.  There was more support for Hill overall (about 45% believed her and 45% Thomas) and there was some gender gap, with women more likely to believe Hill.  But once again, there was no evidence of a difference by race or an interaction between race and gender.  

Another thing I noticed was that partisan differences were not all that strong.  In 1991, 18% of Republicans and 31% of Democrats thought that Hill was telling the truth.  In contrast, 8% of Republicans and 76% of Democrats said that they thought that Christine Blasey Ford was telling the truth in the Brett Kavanaugh hearings.  

[Data from the Roper Center for Public Opinion Research]

 

 

 

More complicated than I thought, part 3

 In January 1957, the Gallup Poll did a survey for the National Foundation for Infantile Paralysis (the March of Dimes), which was concerned about "the extent to which the public has failed to avail itself of the polio vaccine."  It had an open-ended question on what you thought about the vaccine, which they coded into groups.  I collapsed a few of the smaller ones, giving four categories--approval, qualified approval (their example statements were "not 100% effective" and "will be improved in the future"), disapproval (either because it was ineffective or because it was unsafe), and don't know.  Approval was strong for all educational levels.

                                      Approve        Qualified     Disapprove    DK

Not HS grad                      73%                 14%               2%          12%

HS                                     77%                 16%               2%            4%

Some college                    78%                  17%               2%            3%

College grad                     76%                   20%              1%            3%

It also asked whether you had received the vaccine, and if not, whether you definitely planned to get it.  

                                      Already           Plan to       Don't plan      DK

Not HS grad                       7%                38%               31%          25%

HS                                     12%                44%              23%           20%

Some college                    20%                 41%              26%           14%

College grad                     19%                  42%             24%            14%

 Despite the similarity in views of the vaccine, more educated people were far more likely to have actually had it.  

Another question was "if there are enough polio shots for everybody, do you think all young adults should have themselves vaccinated against polio--or do you think it doesn't matter very much?"  Almost all of the people who had already been vaccinated or said that they definitely intended to said yes, but so did 73% of the people who didn't plan to and 82% of the people who weren't sure.  

Many contemporary discussions of the Covid vaccine assume that people who haven't gotten vaccinated by this point are firmly opposed to getting vaccinated, and go on to argue about whether that reflects mindless partisanship or "a reasonable uncertainty and wariness after a year of shifting public-health rhetoric, blunders and misleading messaging" (to quote Ross Douthat).    Of course, there is more outright opposition today, because Covid has become a political issue (the 1957 survey didn't ask any questions about politics, not even basic party identification).  But I think that the patterns found in 1957 are still relevant today--there are also people who think they should, but haven't gotten around to it, or who don't have strong feelings but would be willing.  Less educated people are more likely to fall into these groups.  

In my first post, I said I thought that the media "used to be more deferential to the authorities and less inclined to report news that might promote doubts."  I looked at New York Times stories in April-June 1955, as the story about the Cutter vaccine was developing, and found that was true to an even greater extent than I expected--the stories on the potential issues with the Cutter vaccine seemed to go out of their way to downplay the concerns.  A piece in the ancestor of the Sunday Review section about the withdrawal of the Cutter vaccine said "Unfounded rumors of black market operations are rife, the reputation of a trustworthy drug house is damaged, and parents throughout the country are thrown into a state of alarm."  It went on to say that the "few cases of polio . . . among the 300,000 who were vaccinated with the Cutter preparation" may have represented children who were already infected.  A few days later there was a little story about the California Medical Association calling for a thorough investigation   It mentioned that their statement "amounted to almost 100 percent disapproval of the handling of the vaccine program" but didn't expand on that--the whole story was only five sentences long.  A few weeks later, after a federal commission had recommended changes in the program, another Review piece treated the issue as closed, and framed it as scientists stepping in to sort out problems cause by blundering politicians:  "whatever had happened, the scientists were confident that it need not happen again--if the government would put their proposals into effect.  They won.  ... These moves are expected to go a long way towards raising public confidence in the safety of the Salk vaccine."  One thing that struck me is that they treated "the scientists" as a unified body--they just quoted the official commission, with no comments from outside experts.  There are a lot of criticisms that can be made of the media today, but I don't think I'd want to go back to this kind of coverage. 

[Data from the Roper Center for Public Opinion Research]

More complicated than I thought, part 2

 My last post noted that while the polio vaccine was being tested, a significant minority of people said they they were not interested in getting it for themselves (31%) or their children (17%), but views on that issue could not be predicted by demographics or political views.*  In April 1955, as vaccination distribution was starting, 84% of parents said that they planned to have their children vaccinated, and only 5% said that they didn't.  At that point, there was a difference by education--among people who hadn't graduated from high school, 78% said they did and 7% that they didn't, while among people with a high school diploma, it was 89% and 3%.  

Then came the "Cutter incident," which led to a decline in plans to take the vaccine.  The educational difference persisted, although it was not that large.  For example, in January 1956, 43% of people without a high school diploma said they intended to get the vaccine and 43% said they didn't; among college graduates, 53% said they did and 37% said they didn't.  A pretty substantial difference by race emerged--in the same survey, 65% of blacks said they would get the vaccine, against 27% who said they wouldn't; among whites, it was only 45%-39%.  The educational difference isn't surprising--more educated people tend to be quicker to adopt new things and to have more confidence in scientific authorities.  The racial difference is interesting, and I may say more about that in a future post, but now I'll turn to politics.  

At that time, the Gallup Poll regularly asked people how they had voted in the previous election.  In 1954, there was no relationship between 1952 vote and interest in the vaccine.  That survey had several other questions on politics (e. g., opinions about Joe McCarthy) and they didn't seem related either.  In April 1955, there was no difference between people who had voted for Eisenhower or Stevenson--people who hadn't voted were less likely to say that they planned to get their children vaccinated (more "don't knows" rather than more intending not to).  But starting in May 1955, a political divide appeared.  The May survey had a question about political ideology (which was not common at the time).  The relation to plans to get your children vaccinated (asked only of parents):

                    Yes          DK/Mixed       No

Liberal          79%           15%             5%

Con.              60%           25%            14%

DK                67%           24%              9%

This relationship may have existed be because many people thought of "liberal" and "conservative" in terms of a general disposition to favor new things rather than a more strictly political sense--the connection to 1952 vote was weaker (and not statistically significant).  But by 1956, there was a clear connection to politics.  That survey asked how people would vote if there was an Eisenhower-Stevenson rematch that fall:

                                          Yes      No        DK

Eisenhower                        45%    40%       15%

Stevenson                           54%    34%      12%

Undecided                          42%     34%     24%

That's nowhere near as large as the political difference in views of the Covid vaccine, but it suggests that there might be a tendency for conservatives to be more reluctant to accept new vaccines.  That might be reflect general attitudes about novelty, or a suspicion of government action. 

*There was one demographic characteristic that made a difference--age.  Since polio was a disease of children and young adults, that difference is easy to understand.

[Data from the Roper Center for Public Opinion Research]