American wokeness

 A few years ago, I compiled information from a number of surveys and concluded that perceived discrimination against blacks increased sharply around 2015.   I recently ran across a question that I hadn't noticed then, which I'll use to check and update those conclusions.  The question is "In general, do you think blacks have as good a chance as white people in your community to get any kind of job for which they are qualified, or don't you think they have as good a chance?"  It was first asked in 1963, then again in 1978, then in 1989 and pretty frequently since then.*  The figure shows the percent who said they did have as good a chance minus the percent who said they didn't.

It increased between 1963 and 1978, then stayed about the same for several decades.  Agreement was higher in two surveys from January and October 2009, which probably reflected general optimism associated with Barack Obama's election as president.  In February 2015, 72% thought black people had an equal chance and 28% that they didn't.  The next time the question was asked was in July 2016:  64% said they did and 36% that they didn't.  In July 2020, it was 58% and 42%.  In July 2021, it was 55% and 44%.  People often say that there was a period of "peak woke" for a year or two after the murder of George Floyd, but that it's receded as maybe even been replaced by a backlash.  However, when the question was asked in June 2025, the results were almost the same as in 2020 and 2021 (55%-42%).  

In my earlier post, I suggested that the change in opinion was the result of media coverage and viral videos about police misconduct and the mistreatment of black people in everyday life.  But why did they find such a large and receptive audience?  In a book published in 1981, Samuel Huntington proposed that American politics was marked by periods of "creedal passion"--when substantial groups of people became upset about the gap between widely held values (the "American creed") and actual social conditions.  Of course, the "American creed" is hard to define, but I would say that a belief in social mobility for both individual and groups is a major part of it.  Americans accept a lot of inequality, but don't like anything that resembles a caste system, with hereditary groups at the top and bottom.  After the civil rights laws of the 1960s, the general view was that in a generation or two, the class distributions of blacks would be similar to that of whites.  But although there was some decline in racial inequality, it was slow--that is, there was increasing tension between the ideal of a mobile society and the reality of enduring inequality.    And if racial inequality is the result of discrimination, then it's clear what can and should be done to reduce it.  As a result,  people were attracted to that explanation:  the alternative was that we might just have to live with it for a long time, maybe forever.

That leads to the question of why the change in views on race hasn't led to substantial reforms--if anything, public policy has shifted away from trying to reduce racial inequality.  I'd say it was a lack of political leadership.  In its absence, the desire to do something led to actions that were unhelpful or sometimes counterproductive (e. g., dropping standardized tests in university admissions).  

*It referred to "negroes" the first time it was asked.  I also include a few that asked:  "Do you think that blacks who live in your community do or do not...have as good a chance as whites to get a job for which they're qualified?"

Inside and outside

 A couple of weeks ago, I had a post on declining public confidence in universities.  Most people who've written on this topic say that the cause is internal--something universities have done or failed to do-- and that they need to reform in order to win back support.   I suggested that the major cause was external:  that leading Republicans had become more critical of universities, and the public (especially Republicans) followed.  In this post, I'll elaborate on that point, focusing on one frequently mentioned factor:  political bias.  Megan McArdle writes "in the wider world, asking whether academia really skews left makes you look like an idiot or, slightly more charitably, like someone so encased in a bubble that they don’t even know what they’re missing."   In 2021, an Axios/Ipsos survey asked if you agreed or disagreed with the statements "colleges and universities are biased in favor of liberal ideas and beliefs" and "colleges and universities are biased in favor of conservative ideas and beliefs."  I combined those into five groups:  those who thought they were biased in favor of only liberal beliefs, only conservative beliefs, both liberal and conservative beliefs, not biased, and don't know.*

                                    All            College grads
Liberal only                37%            49%
Conservative only        9%               7%
Both                             9%                6%
Not biased                   14%              17%
Don't know                  32%              21%

Less than half of people believe that there is a liberal bias.  Among the public, 32% think that there is a conservative bias or no bias; among college graduates, it's 30%.  And even among college graduates, a lot of people say they don't know.  So in the "wider world," there isn't a consensus on the topic.  

These results illustrate a more general point.  Although there's a good deal of media coverage of higher education, it's a niche topic--it rarely appears on TV news or in local newspapers.  Also, there are no standard measures that are regularly reported--even someone who follows the issue just gets a collection of individual events. As a result, actual conditions have little influence on public perceptions.  But if a prominent politician speaks out about universities, that is a major story, and people have a sense of how to react:  it depends on how they generally feel about that politician.  So what political leaders say does make a difference.  

*People who said they were biased in favor of one side and didn't know about the other were counted as biased in favor of only that side.  Disagree on one and don't know about the other were counted as not biased, so the "don't know" category means don't know on both questions.

[Data from the Roper Center for Public Opinion Research]

Aftermath

 A few thoughts in the wake of Tuesday's election:

1.  The basic story: the voters who put Trump over the top in 2024 were expecting something like his first term, which was a pretty standard Republican administration in terms of policy, although not in terms of style.  His second term has been much more radical, and people reacted against that.  

2.  Paul Krugman writes "the 2024 election was mainly about economics", particularly inflation, and that 2025 was too.  This overstates the importance of inflation and understates the importance of immigration in 2024.  I'm not sure about the overall impact of immigration in 2025:  most voters like the reduction in illegal immigration, but regard Trump's methods as too harsh.  But in 2024 the sense that the Biden administration had let things get out of control definitely helped Trump. 

3.  Jamelle Bouie writes "As a presidential candidate, Donald Trump is a phenomenally effective vote-winner, capable of turning out millions of otherwise infrequent voters to deliver the White House and Congress to the Republican Party."  I think this is wrong--his record in the 2016 and 2020 general elections was very poor.  In 2016, he got 45.9% of the vote against Hillary Clinton's 48%, even though Clinton was one of the weakest major-party candidates in a long time.  In 2020 he got 46.8% against Joe Biden's 51.2%--it was only the peculiarities of the Electoral College that made it a close election.  This time, Trump had the advantage of incumbency, which should have been even larger than usual because people tend to rally around leaders in a crisis.  All Trump had to do was act like he was taking Covid seriously, but he couldn't bring himself to do that.  Bouie is right that Trump turned out millions of infrequent voters, but many (or most) turned out in opposition to him.  

This leads to the question of why the party renominated him.  People sometimes say that it was because he was so popular among ordinary Republican voters that he couldn't be stopped.  I don't think that this is it.  The figure shows very favorable ratings (4 or 5 on a -5 to +5 scale of major party presidential candidates since 1952:

31% gave Trump a very favorable rating in 2024, which was better than 2016, but still pretty ordinary--Mitt Romney had 30% in 2012 and George W. Bush had 34% in 2004.  Trump was only a slightly ahead of Harris, who had 29%.*

I think the major reason for Trump's renomination is that leading Republicans didn't put up much resistance.  Why not?  My (tentative) answer it's because being seen as divided is a negative for a party, and it was clear that if Trump lost a battle for the nomination, the party was going to be divided--he wouldn't accept defeat and urge everyone to unite around the nominee.  So their strategy was to hope that once he was out of office, he wouldn't be the center of news news coverage, and that his support would just fade away.  This wasn't unreasonable, since there was no previous experience with a situation like this.  

[Data from the Roper Center for Public Opinion Research and Gallup]

*There were two surveys in 2024, one in May and one in October.  Both Democratic and Republican very favorable ratings were higher in October:  29% for Harris vs. 20% for Biden, 31% for Trump in October vs. 25% in May.  My initial version of this post mislabeled May 2024 as 2020.  There was a survey in October 2020, but the report doesn't have a breakdown of the degree of favorability.  

We're going wrong

 Andrew Gelman had a post on "the last time that it seemed that the United States was coming apart, in the late 1960s."  He says that there was there was "the same feeling of things spinning out of control and the idea that conventional politics was failing."  I was going to leave a comment, but it expanded into this post: 

The most important difference is that in the late 1960s there was a lot more confidence in political and institutional leaders.  The Gallup poll sometimes asks people to rate public figures on a scale of +5 to -5.  In October 1968, 79% gave Richard Nixon a positive rating, with 23% giving him a +5, and only 6% giving a -5.   Hubert Humphrey got 72% positive, with 16% giving a +5 and 8 a -5.   In October 2024, 51% gave Donald Trump a positive rating, with 21% at +5 and 35% at -5.  Kamala Harris got 48% positive, with 16% at +5 and 29% at -5.   That is, in 1968 many people thought that both candidates were pretty good, and few people thought that the one they were voting against was terrible; now it's the reverse.  We also have data for confidence in "the people running" various institutions, from the Harris Poll in 1967 and the General Social Survey in 2024.  In 2024, confidence was lower for 12 of the 13 institutions, and about the same for the remaining one (organized labor.

But I think that the sense that things were spinning out of control continued through the 1970s and early 1980s.  Although there were ups and downs, there was a constant background of economic problems ("stagflation"), rising crime and social disorder, and international setbacks.  Confidence in all thirteen institutions fell between 1967 and 1980, although the 1980 levels are higher than today's, with the exception of the military and organized labor.  As far as political leaders:

Both Carter and Reagan got low ratings compared to most previous candidates, but both were favorable on balance:  Carter had 68% positive and Reagan 64%, and both had more +5 than -5 ratings.  Independent candidate John Anderson also had over 60% positive.  

Turning to the general mood, I have to rely on impressions rather than data.  I was born in 1959 and wasn't paying attention to politics and public affairs in the late 1960s, but my sense that most people thought that the country had serious problems, but that its leaders could solve them.  By the late 1970s, there was a feeling that maybe our problems were too big for anyone to solve.  Now there's a feeling that the leaders aren't even trying to solve problems, but are causing them.  

[Data from the Roper Center for Public Opinion Research]

One way or the other

 Megan McArdle writes "Public trust in universities has plummeted over the past decade, among independents as well as conservatives, which is why Republicans now feel so free to attack them."  McArdle says that the major reason that trust has declined is that the prevailing political climate in universities has moved to the left:  they developed "a left-wing culture that appeared increasingly hostile to the society paying its bills."  It is true that the political climate at universities is well to the left of the nation as a whole, but that isn't a recent development:  the culture at leading universities has been moving to the left for a long time.  So if that's the cause, we should see a gradual decline of confidence over a long period of time.  Between the 1970s and the early 2010s, the Harris Poll regularly asked how much confidence you had in the "people in charge of running major educational institutions, such as colleges and universities."  There wasn't much trend--it declined from the 1970s to the 1990s and then bounced back (see this post).  Gallup started asking a similar question in 2015, and confidence declined substantially between 2015 and 2024.   Since the decline in confidence is limited to the last decade or so, the cause must also be a relatively recent development.  McArdle offers another possibility:  "the value of a college degree is stagnant and might fall as AI takes over much knowledge work."  The timing on this is a better fit:  payoff to a college degree rose until about 2010 and has stayed about the same, or maybe declined a little, since then.  However, the claim that stability in the college wage premium leads to a decline in confidence doesn't make sense:  it amounts to saying that people want to see college graduates continue to pull away from everyone else.

What if she has it backwards:  Republicans have attacked universities over the past decade, so public trust (especially among Republicans) has plummeted?  I don't have any measure of Republican rhetoric about universities, but my impression is that it has become increasingly negative in the past ten years or so.  There also seems to have been an increase in media coverage of campus "activism":  two prominent examples, at Yale and the University of Missouri, happened in 2015, and there have been many more since then.   Since 2023, there have been many protests about the conflict in Gaza.  The rate of very little or no confidence answers rose from 23% in 2023 to 32% in 2024, and then fell back to 23% in 2025.  That suggests that changes in public opinion are partly a reaction to actual events; although there were a lot of protests in 2024-5, they were less disruptive and administrations handled them more effectively (the surveys are taken in June).  There's also probably also a self-reinforcing process of more coverage creating more interest and more interest creating more coverage.  

Left, right, and elite, part 2

In my last post, I suggested that opinion differences between elites and the public didn't fall into the standard left/right pattern; this post will take a more systematic look.   The Chicago Council of Foreign Relations/Chicago Council on Global Affairs has a long-running series of surveys of foreign policy elites and the general public which include questions on the importance of different foreign policy goals.  The figure shows the importance that self-described liberals and conservatives in the general public put on these goals in 2016 (3 represents "very important," 2 "somewhat important" and 1 "not important at all").   

:

The correlation is -.37:  that is, the ones that are important to liberals tend to be less important to conservatives, and vice versa.  The figure suggests that they fall into two groups:  one which is strongly related to ideology (the ones that fall in a downward sloping line) and another on which there is a consensus that they are very important.  These are "preventing the spread of nuclear weapons," "combatting international terrorism," "protecting the jobs of American workers," and "attaining US energy independence."

There is a moderate positive correlation between elite and liberal ratings of importance (0.43), and essentially no correlation between elite and conservative ratings.  The figure shows elite and liberal ratings:

There are substantial differences for some of the items:  protecting jobs and attaining energy independence are very important to liberals in the public, but not that important to elites.  On the other side, "defending our allies' security" and "maintaining superior military power worldwide" are more important to elites than to liberals.  Going back to the first figure, you can see that military power is more of a conservative priority in the public, while defending allies is in the middle.  

The list of possible goals offered changes from time to time, so I will show the liberal/elite comparisons for a couple of other years.  

In 2014, they asked about "strengthening the United Nations"; liberals in the public regarded it as a lot more important than elites did.  As in 2016, elites rated protecting American jobs as less important and maintaining military power and defending allies security as more important then liberals did.  As in 2016, both rated "controlling and reducing illegal immigration" low, but it was substantially lower for elites than for liberals.  

The 2004 survey also included a question on protecting jobs, and again that was less important to elites than to liberals in the public.  Strengthening the UN was also included, and was rated higher by liberals, although the gap was smaller than in 2014.  Illegal immigration was also included:  liberals rated it as of moderate importance, while elites rated it low.  On the other side, elites rated "helping to improve the standard of living of less developed nations" as a lot more important than the public did.  

Finally, turning to changes in priorities, here are the ratings for liberals and conservatives on the six items that were included in both 2004 and 2016 (blue is 2004, red is 2016).  Movements on the horizontal axis represent change among conservatives, while movements on the vertical axis represent change among liberals.  Military power and illegal immigration became less important to liberals and more important to conservatives, while "combatting world hunger" become less important to conservatives and stayed about the same among liberals.  The other three stayed about the same for both.  

I won't show the figure for elites, just say that hunger become less important, maintaining military superiority more important, and the others stayed about the same.

So elite opinions are different from liberal opinions--and not in the sense of being farther to the left on all issues.  Elites have consistently rated protecting jobs and controlling and reducing illegal immigration as relatively unimportant.  Elite priorities also can't be characterized as straightforward internationalism:  they rate defending allies as important, but strengthening the UN as not very important.  

What's the source of the differences?  One possibility is demographic differences:  elites are more educated, and more likely to be white, male, and older.  But these factors don't account for much of the differences in opinion.   Another possibility is factual knowledge:  e. g., elites will know more about how close other nations are to getting nuclear weapons.  Then there are ideas about how things work--e. g., elites will be more aware of economists' view that protecting domestic jobs will have the cost of reducing our standard of living.  You could also add values, but these tend to go along with ideas about how things work:  e. g., that a higher standard of living will make less developed nations more stable and more peaceful, so helping them to achieve a higher standard of living is a matter of enlightened self-interest.  It's hard to measure "ideas about how things work" but it's safe to say that some combination of knowledge, experience, and interactions within elite groups may combine to produce a distinct viewpoint.  I offer more thoughts about the difference between elite and popular priorities on illegal immigration in this paper.  

[Data from the Roper Center for Public Opinion]

Left, right, and elite

 Ross Douthat writes "the political right has plenty of popular support but considerably less influence inside the managerial systems through which elected officials actually exercise their power," while progressivism has  "an extraordinary advantage in the meritocratic institutions, private as well as public, that actually staff and shape the power structure."  As a result, "activists and elites effect dramatic change outside the democratic process and then try to survive or sidestep backlash from the voters."  He offers several examples, including "a new regime of euthanasia in Canada."  

Back in 1947, the Gallup Poll asked, "when a person has a disease that cannot be cured, do you think doctors should be allowed by law to end the patient's life by some painless means if the patient and his family request it?"  This question was asked again in 1950 and has been included in the General Social Survey since 1977.  The percent who say that doctors should be allowed to end the patient's life:

Support passed 50% in the 1970s, and has been over 60% since the 1980s, but assisted death is allowed in only eleven states plus Washington, DC.  Oregon was the first to legalize it, by a referendum--it wasn't until 2013 that it was enacted through a state legislature.

I don't have as much data for Canada, but in 2000 Gallup Canada asked "When a person has an incurable disease that is immediately life-threatening and causes that person to experience great suffering, do you, or do you not think that competent doctors should be allowed by law to end the patient's life through mercy killing, if the patient has made a formal request in writing?"  and 72% said yes.  It also asked about a disease that "is not immediately life-threatening but causes that person to experience great suffering" and 54% said yes.  This was 15 years before laws allowing that were passed.

So assisted dying is not something that was imposed by elites on an unwilling public.  In fact, the question is why political elites have been reluctant to do something that has strong support among the public.  Political forces are probably part of the answer--opponents are more likely to participate in organized religion, so they are better organized and more committed than supporters.  On the other hand, more educated people are more likely to support assisted dying, and political elites have more education than the general public.  So it seems that there is something that keeps them from following the usual inclinations of their class.  One thing that separates political elites from educated people in general is that they think about legislation for a living, and I suspect that they were concerned about the possibility of a "slippery slope" (which has happened to some extent in Canada):  it's hard to write rules that draw a clear line saying exactly when it will be allowed, so they are reluctant to change the status quo.  

The general point is that Douthat, like many other people, implicitly uses a broad definition of elite--"institutions that staff and shape the power structures."  But the people who make and interpret the laws are a much smaller group, and they may have a distinct outlook that doesn't fit into the standard left/right spectrum.  I've had a few posts on this topic, and I'll return to it soon (probably in my next post).

[Data from the Roper Center for Public Opinion Research]