More complicated than I thought

 This title could apply to a lot of my posts, but the story behind this one is that in February Andrew Gelman sent an e-mail to several people who studied public opinion noting that quite a few Americans said they didn't want to take the covid vaccine and asking if this was true in other countries.  I didn't know of any data on other countries, but I looked up surveys on the polio vaccine in the 1950s, and found that in 1954, 31% said they didn't want to get it, and 17% didn't want their children to get it.  I added "That was more “no” answers than I expected. Maybe a significant number of people are always reluctant to try something new, or at least to be among the first to try it. I think there has been a change in the media—they used to be more deferential to the authorities and less inclined to report news that might promote doubts. Today they are more willing to report on anti-vaccine sentiments and on potential problems or limitations of the vaccines."  Recent news stories about continued reluctance to take the covid vaccine brought this back to mind, so I looked for more data about the polio vaccine.

In 1954, the vaccine was being tested.  The tests were successful and general distribution began in April 1955.   Initially the distribution was limited to children, who were more likely to get polio.  The results for questions on whether you would "like to" take it (1954) or were "planning to take it" (1955 and after).

                     Yes          No

May 1954      60%       31%

June 1955     37%        49%

Jan 1956       48%        38%

The results for questions on children:

                         Yes*        No          Mixed

May 1954          75%      17%       

April 1955         84%        5%

May 1955          69%      10%       10%

It seems that something happened between April and May 1955 to reduce confidence in the vaccine.  That something was the "Cutter Incident," which was mentioned in one of the comments on Andrew's blog post.  Several companies were manufacturing the vaccine.  One of them, Cutter Laboratories, used a process that did not kill all of the virus.  That is, their shots contained live virus, and some people got polio from the vaccine.  The Cutter vaccine was recalled on April 27, 1955, almost immediately once reports of problems appeared, but distribution of vaccines from other manufacturers continued.  So the drop in interest in taking the vaccine between April and May is understandable.  In June 1955, a report was released which concluded that the problems were limited to Cutter, and that the vaccines produced by other manufacturers were safe.  But although interest in taking the vaccine recovered to some extent, in January 1956 it was still lower than it had been in 1954.  In early 1957, when the vaccine had become widely available,  most children under 12, but only 39% of teenagers and 12% of adults under 40 had been vaccinated.**

So there was a good deal of "vaccine reluctance" in the 1950s.  I looked at the 1954 data on interest in getting your children vaccinated and found no clear differences among demographic groups and no relation to political views.  In my next post, I'll look at the data from May 1955 and after to see if any pattern emerged (especially any pattern involving politics).

[Data from the Roper Center for Public Opinion Research]

*Includes 11% "already vaccinated" in May 1955.  

**People over 40 were not considered to be at significant risk of getting polio. 

By the way

 I found a few interesting survey questions that I want to record here before I forget:

1.  A couple of years ago, I talked about a series of questions that distinguished between economic and social ideology that started in 1999.  Contrary to what is often suggested, people consistently rated themselves as more liberal on "social issues" than on "economic issues."  I recently found a couple of earlier questions on the same topic:

a.  A 1987 survey of people aged 18-44 asked them to rate themselves on "economic matters", "foreign policy issues," and "social issues."  The results

                        Liberal    Middle    Conservative

economic        18%        44%        31%

social                26%        41%       26%  

foreign             20%        35%        32%

b.  A survey of the adult population in 1978 asked "People who are liberal on economic matters usually are in favor of stepping up federal programs to solve the problems of poverty and unemployment, they tend to support labor rather than business, and they think there should be more government planning and regulation of the economy. People who are conservative on economic matters tend to side with business rather than labor, they believe that government should be as small as possible, and it should interfere less with free enterprise. When it comes to economic matters like these, would you say you area generally very liberal, somewhat liberal, middle-of-the-road, somewhat conservative, or very conservative?" and "People who are liberal in their lifestyles tend to support affirmative action for blacks and other minorities, they are more in favor of women's liberation, and they are not so likely to condemn marijuana smoking or sexual freedom. People who are conservative in their lifestyles tend to be strongly in favor of the death penalty and more often than not, they are opposed to laws that permit abortion or homosexuality. When it comes to social matters like these, would you say you are generally very liberal, somewhat liberal, middle-of-the-road, somewhat conservative or very conservative?"

                        Liberal    Middle    Conservative

economic            19%        32%        48%

social                  22%        27%        48%

In both cases, the average ratings on social issues were to the left of average ratings on economic issues, although the difference was small in 1978.

 2.  In 1955, the Gallup Poll asked "SUPPOSE YOU WERE TALKING TO A PERSON, IN A GENERAL WAY, ABOUT THE UNITED STATES AND OTHER COUNTRIES. WHICH ONE OF THESE STATEMENTS ON THIS CARD BEST EXPRESSES YOUR OWN POINT OF VIEW?

I. THE UNITED STATES IS THE GREATEST COUNTRY IN THE WORLD, BETTER THAN ALL OTHER COUNTRIES IN EVERY POSSIBLE WAY

II. THE UNITED STATES IS A GREAT COUNTRY, BUT SO ARE CERTAIN OTHER COUNTRIES

III. IN MANY RESPECTS, CERTAIN OTHER COUNTRIES ARE BETTER THAN THE UNITED STATES

They asked the same question in 1991, and Pew asked it in 1996.  

              Better than all            Great          Others better

1955             66%                       31%              1%

1991             37%                       54%              9%

1996             37%                       49%            12%

[Data from the Roper Center for Public Opinion Research]

You started it, part 2

 Here are a few more questions on which it's possible to see if there's been a growth in extreme opinions.  Gallup asked " In your opinion, is the death penalty imposed too often, about the right amount, or not often enough?" a number of times between 2001 and 2018.  The results, as summarized on the Gallup website:

 

That looks like just a slight move against the death penalty, not a move towards the extremes.  Gallup does not give a breakdown by party.  

Then there's a question the GSS has asked since 2004:  "Do you think the number of immigrants to America nowadays should be .... increased a lot, increased a little, remain the same as it is, reduced a little, reduced a lot.  The means for Democrats and Republicans:

Democrats have moved in favor of immigration--on the average, they are somewhat in favor of increased immigration (zero means "remain the same"), while Republicans haven't changed much.  

Now the standard deviation of opinions: 

 

No trends--the one year that stands out is 2016, when the number of Republicans saying that immigration should be "reduced a little" increased at the expense of those saying that it should be should be reduced a lot and those saying that it should be kept the same or increased.   That is, Democrats have shifted in favor of immigration, but haven't shown any special tendency to move to the "increased a lot" option. 

Finally, between 1999 and 2019 Gallup asked about climate change:  "From what you know about global climate change or global warming, which of the following statements comes closest to your opinion?...Global climate change has been established as a serious problem, and immediate action is necessary, there is enough evidence that climate change is taking place and some action should be taken, we don't know enough about global climate change, and more research is necessary before we take any actions, concern about global climate change is unwarranted."  I couldn't get breakdowns by party for this one, but here is the overall mean:

 

Basically a movement towards regarding climate change as a serious problem.  The standard deviation:

A pretty clear increase.  The first time that the survey was taken (December 1999), 23% said that it had been established as a serious problem and 11% said that concern was unwarranted.  The last time (December 2019), 44% said it had been established as a serious problem and but the share who said that concern was unwarranted remained at 11%.  Support for the two middle options declined between 1999 and 2019.  

So putting the results from this post and the last post together, polarization increased on two out of the five issues, didn't decline for any of them, and stayed about the same for three.  In one of the cases where polarization has increased (abortion), it's roughly symmetrical--both Democrats and Republicans have moved towards extreme positions.  It's hard to say for climate change, since we don't have a breakdown by party and there's a strong general trend which can plausibly be attributed to new information (average temperatures have continued to increase).  Overall, there's no evidence for the general claim that Democrats have moved towards extreme positions more than Republicans.

[Data from the Roper Center for Public Opinion Research] 

You started it!

 Kevin Drum has attracted a lot of attention with an argument that "It is not conservatives who have turned American politics into a culture war battle. It is liberals."   He starts out by making a valid point--over the past couple of decades "Democrats have moved significantly to the left on most hot button social issues while Republicans have moved only slightly right."  I think he's correct on the leftward movement, but that wasn't what made American politics more polarized.  First, there's been a leftward movement on "social issues" for a long time, at least as long as surveys have been around, but polarization didn't start growing until much later.  Second, contemporary polarization isn't just (or even mainly) about "culture war" issues.  As I've observed before, during the Obama years the most polarizing issue was probably the Affordable Care Act, and several "culture war" issues faded (notably same sex marriage).  In fact, you could argue that the biggest source of polarization isn't any kind of policy issue, but a sense that the other side isn't playing fair--Democrats bring up Republican treatment of Merrick Garland, Republicans reply by talking about Democratic treatment of Brett Kavanaugh, and pretty soon we're back to arguing about whether Robert Bork was treated unfairly.   When people say that the Republicans are responsible for the growth of polarization, they usually mean that Republicans have been more aggressive about playing  "constitutional hardball".  Since this question involves the behavior of political elites, data on opinions in the general public can't shed much light on it.  

But Drum does bring up an interesting question about the public--has there been a movement towards extreme positions?  His primary data source is a series of Pew surveys which used ten questions to make an index of liberalism/conservatism.  This shows a substantial growth in the numbers on the extreme left, a slight growth on the extreme right, and a decline in the number of moderates.  But the questions in the index all just give two options--e. g. "homosexuality should be accepted by society" or "homosexuality should be discouraged by society."  So an "extreme" score just means taking liberal or conservative positions on every issue.  So public opinion has polarized in that sense, but it doesn't mean that there's been a growth of extreme positions on any particular issue.  In order to see if that's happened, you need survey questions (or series of questions) giving three or more options, which are relatively uncommon.*  Drum presents data on some specific issues, but only one of them gives more than two options--on this, both Democrats and Republicans have shifted towards the extreme positions.  I looked and found a few others.  

The GSS has a series of questions about whether abortion should be legal in various circumstances.  Here are the figures for the percent of Republicans who say that it should not be legal when the pregnancy is due to rape or when the woman's health would be endangered without one.  

Those who say abortion should be legal "if the woman wants it for any reason."

The number of Democrats who take this position has increased.  I also show Republicans because I was surprised to see that even today about 35% of them say that a woman should be able to get a legal abortion for any reason.  Is that because there's still substantial support for legal abortion among rank-and-file Republicans, or maybe because some people misunderstood the question as essentially asking whether abortion should ever be allowed.  

Another issue on which it's possible to identify extreme positions is gun control.  Between 1993 and 2013, the following question was asked a number of times:  "Which of the following statements comes closest to your view?...There should be no restrictions on owning guns. There should be some restrictions on owning guns. All guns should be illegal for everyone except police and authorized personnel."  The overall results:

                    None        Some        Ban

1993            6%            75%        17%

1998            8%            66%        24%

1999            4%            72%        22%

2009           15%            71%        15%

2011            14%            71%        14%

2012 (Aug) 13%            76%        10%     

2012            13%            76%        15%

2013            17%            70%        12%

Support for the extreme conservative position has increased and support for the extreme liberal position has declined.  The percent holding the middle position has stayed about the same, so the change is a shift to the right rather than an increase in polarization. 

I have other things to do, so I'll break here and continue the discussion in my next post.  

*Questions which give a range going from "strongly agree" to "strongly disagree" are pretty common, but strong agreement or disagreement doesn't necessarily mean taking an extreme position.  For example, someone might strongly favor an increase in the minimum wage, but think that it should only be a small one. 
  

[Data from the Roper Center for Public Opinion Research]