A study conducted by the Pew Research Center and published in February presents eight key findings on Americans’ confidence in science and their views on scientists’ role in society, drawn from a number of past surveys. Some of the findings show a clear divide in opinion between Democrats & Democrat-leaning independents and Republicans & Republican-leaning independents.

In particular, the report finds a large difference in opinion on whether scientists have a role in policy-making:

Democrats are more inclined than Republicans to think scientists should have an active role in science policy matters. Indeed, most Democrats and Democratic-leaning independents (73%) hold this position, compared with 43% of Republicans and GOP leaders.

Moreover, with relation to the quality of policies made by scientists:

54% of Democrats say scientists’ policy decisions are usually better than those of other people, while two-thirds of Republicans (66%) say that scientists’ decisions are either no different from or worse than other people’s.

The report also finds a substantial partisan divide in confidence in the scientific method to produce accurate conclusions:

Democrats are more likely to express confidence in the scientific method to produce accurate conclusions than do Republicans, on average. Most Democrats with high levels of science knowledge (86%, based on an 11-item index of factual knowledge questions) say the scientific method generally produces accurate conclusions. By comparison, 52% of Democrats with low science knowledge say this. But science knowledge has little bearing on Republicans’ beliefs about the scientific method.

What explains the large partisan divide on this subject?

Pew Research Center diagrams

  • 36
    Without saying anything partisan, it's important for people to realize that science and scientists are two different things. Scientific history shows this. – Panzercrisis Apr 24 '20 at 1:03
  • 1
  • 3
    @Panzercrisis In fact, that's the reason we have science, instead of just trusting everyone to use flawless Bayesian reasoning to make the best possible conclusions from the available evidence – scientists can't be trusted to do science right, so science ensures they can never worsen our understanding. It's a ratchet, not a railgun. – wizzwizz4 Apr 24 '20 at 18:57
  • 7
    The use of sound scientific study for non-sequitur policy decisions might have something to do with it. Stuff like "Science has given us X as a fact; we should therefore do Y" is reasonable on the face, until you specifically discuss X and Y and sometimes find they have nothing to do with each other. Unfortunately, throwing the baby out with the bath water is sometimes a response. For specific examples, consider liberal reliance on global warming and conservative reliance violent crime demographics. Each sometimes critiques the science therein, rather than the derived policies. – frеdsbend Apr 24 '20 at 19:08
  • 4
    @Panzercrisis: Agree with the emphasis on clearly laying out what people might think of as "science", "scientific experts", and "the scientific method" when responding to the survey. This is, unless the survey-givers sat everyone down and gave them a long lecture on exactly what they mean by these terms, then responses likely reflect not just differing opinions on such things, but also differing perceptions of what the questions on the survey were actually asking. – Nat Apr 25 '20 at 0:54

A good explanation might be confirmation bias. Confirmation bias in a nutshell is the psychological phenomenon that people generally tend to trust information which supports their views and distrusts information which contradicts their views. And this trust or distrust does of course extend to those who provide that information.

And it just so happens that the advice which can be taken from scientific consensus on multiple issues which dominate the political discourse in the United States happens to align more with the Democratic party position than with the Republican party position:

  • The biggest perhaps is Global Warming. The only conclusion which can be taken from the scientific consensus is that it can only be stopped by regulating both the industry and the personal lifestyles of people. This goes against Republican values like free market economy and personal liberty.
  • Immigration is another issue. Most economists come to the conclusion that the net effect on the US economy from immigration is positive. But this does conflict with the anti-immigration sentiments in the Republican party.
  • In education, there is the creationism debate and the sex education debate. The scientific perspective is of course that people should be taught what's scientifically correct: There is no evidence for intelligent design, the best way to prevent teenage pregnancies is information about contraception and LGBT people do exist. The more conservative view of many Republicans is that respect for religion and sexual morals should be taken into account when designing the school curriculum.
  • And now, during the COVID-19 pandemic, scientists come to the conclusion that the only way to reduce the expected number of casualties is to enforce draconian contact restrictions. This again conflicts with Republican values like the freedom to engage in economic activity, skepticism of government overreach and personal liberty.

I don't want to imply that just because science says what's best from an utilitarian perspective, that it is always the only right decision to make. Science does not consider political values like ethics and morals or how people want to live their lifes. Scientists can offer data which can be used as the basis for decision making, but making the decision is the responsibility of politicians. Or in other words: "Facts don't care about your feelings", but politicians do.

  • 33
    If you pick other examples, you get the opposite results. For instance the perceived harm of GMOs seem to be by the democratic side quite unaligned with science. I wouldn't agree with you if your answer is reducable to "one side listens to science, the other doesn't". I do however, believe in your first statement about conformation bias. I think your post could be improved by adding a few examples where liberal / democratic values are unaligned with science. – Stian Yttervik Apr 24 '20 at 11:48
  • @StianYttervik Thank you for that comment. I actually considered to add some examples where democrats are skeptic of science which supports the Republican position to provide some balance, but couldn't come up with any good examples of my own. GMOs might be one. When I have the time I will do some more research on this and then perhaps add this point. – Philipp Apr 24 '20 at 11:55
  • 3
    @StianYttervik Is that actually true? This Pew study says that there's no difference between Conservatives and Liberals, though Moderates are more likely to think they're dangerous. The key factor seems to be education level and science knowledge – divibisan Apr 24 '20 at 14:26
  • 27
    @Philipp: Nuclear power would be another example of where a part of the Democratic base is against science, or perhaps more accurately, deliberately chooses to remain ignorant of it. – jamesqf Apr 24 '20 at 16:09

To understand the modern partisan divide over science, we need to go back to the origins of the Christian Fundamentalist movement in the late 19th century. There's an article on it in Brittanica that's worth reading in its entirety, but it sums up things nicely in the opening paragraph:

Christian Fundamentalism [is a] movement in American Protestantism that arose in the late 19th century in reaction to theological modernism, which aimed to revise traditional Christian beliefs to accommodate new developments in the natural and social sciences, especially the theory of biological evolution. In keeping with traditional Christian doctrines concerning biblical interpretation, the mission of Jesus Christ, and the role of the church in society, fundamentalists affirmed a core of Christian beliefs that included the historical accuracy of the Bible, the imminent and physical Second Coming of Jesus Christ, and Christ’s Virgin Birth, Resurrection, and Atonement.

Consider the condition of the US in the late 19th century. On the Eastern seaboard there was rapid industrialization, with all the consequent social problems that entails, and a growing intellectual/academic sector that promoted a secular worldview. The deep south was still reeling in the aftermath of the Civil War, which upended the traditional social institutions. The Western territories were beginning to open up, and was still a somewhat lawless and amoral region, known (anecdotally, at least) for violence, gambling, prostitution, gold hunting, and other less-than-upright activities. Huge numbers of immigrants were entering the country to work in factories, to settle in the West, to seek out their fortunes. And of course, this was the beginning of Marxism, Darwininsm, and eventually Freudianism, the three great changes in European thought that began filtering over into the the US. It was a moment of great change and unrest, and that moment put a strain on the largely Christian population that was already established.

Christian Modernism was a movement that called for Christians to embrace these social changes (within reason). It wanted to relax and revise some of the points of Christian Liturgy that conflicted with modern social changes: embracing blacks and foreigners as brothers in Christ, accepting scientific principles that conflicted with age-old teachings, tolerating secular amorality in order to better focus on redressing immorality. But many Christians found this 'modernist' movement troubling. They thought indulging these modernist revisions would gut the church and render it powerless to fight against moral depravity, resulting in the dissolution and destruction of Christian ideals and communities. And there was plenty of anecdotal evidence to back that up, particularly in stories of young people leaving small Christian communities to seek out their fortunes in cities, or to find that mythical gold strike out in the Wild West. These Christians organized themselves as Fundamentalists, who asserted the uncompromising and absolute truth of Christian teachings, and rejected outright both the social upheavals they saw and the secular-scientific worldviews those upheavals ostensibly came from. All of this came to a head in the 1920s, when Tennessee passed the Butler act which made teaching evolution a misdemeanor crime — the setting for the famous Scopes trial — and a number of other states followed suit with similar laws.

The point is that in the 1920s Fundamentalism went from being a purely religious movement to establishing itself as a political movement. And while Fundamentalism has been highly successful as a religious cause, it has (in the years since the Civil Rights movement) suffered any number of political setbacks. Laws against teaching evolution or requiring the teaching of creationism or intelligent design have been overturned; the right to abortion has become enshrined in law; religious displays and monuments have been forced out of public spaces: step by step, the US has moved away from being a Christian nation to being a secular or faith-independent state. These political losses have chaffed at the Christian Fundamentalist community.

Now, somewhere around the Reagan years (the 1980s), the GOP began constructing a new coalition among various groups that felt disenfranchised by the Liberal swing of the Federal arena. They brought together deep South segregationists, small town ('family values') conservatives, Right-libertarians and hard-line capitalists, neo-McCarthyists, and yes, Fundamentalist Christians, many of whom had little in common with each other except:

  1. They opposed the inclusive communitarianism of the left (for widely divergent reasons), and
  2. They opposed the entry of science into social and political affairs

The second point is derived from the first. Since the 1960s, the Left has bolstered its moral worldview with scientific research, using such research (variously) to show the damaging effects of pollution; describe social stratification, systematic racism, or the oppression of women; undercut religious teachings and attack religious viewpoints; etc. Conservative groups don't dislike 'science' per se. They dislike science when it is used as part of a social/moral argument. Part of that, of course, is the fault of those who use science poorly and aggressively — certain anti-theists, certain progressive 'warriors,' etc; those who try to leverage science as a hammer to destroy social institutions they dislike — but part of it is the fact that scientific arguments, done well, are extremely difficult to contest, and that creates a moral dilemma. For instance, it would be perfectly feasible for someone to look at climate science and make a moral argument like: "I accept climate science, and accept the idea that humanity is altering the world's climate, but I reject the conclusion that this is a moral wrong, and see no reason to change anything." But making such a moral argument often sounds heartless and cold; it is socially and psychologically easier to discredit or obfuscate the science itself.

So in short, the reason why there is such a strong partisan divide on the credibility of science is that the GOP has explicitly constructed a coalition that opposes the Federal trend towards liberalization, and rejects the mass of scientific research that trend relies on.

  • 1
    Comments are not for extended discussion; this conversation has been moved to chat. – yannis Apr 29 '20 at 13:02

The question as titled comprises two related but distinct issues.

What is the quality of scientific data?

Unfortunately, "science" is an unhelpfully broad category here, and "the scientific method", while a useful illustration for school children, is a very rough approximation for how the sausage gets made. Even in the "hardest" of sciences (e.g., high-energy physics), the selection of what topics to pursue and how is influenced by funding politics (government policy, industry influence, fashionability of certain theories, even intra-departmental rivalries), the decision of what gets published depends on a number of people's idea of what qualifies as "interesting" (what to write up, what the editor approves, what passes peer review), and the "consensus position" is subject to all sorts of human biases and errors.

Strikingly, the Particle Data Group publishes a set of graphs showing the consensus values of various physical constants over time. There are two important lessons to be learned from these graphs: The consensus (with error bars!) has often been substantially off from the current (hopefully better) consensus values; and even more important, there are several cases where the scientific community has accepted and held onto a conclusion for a substantial period of time before finally accepting overwhelming evidence that a revision is necessary.

Particle Data Group historical constants

Keep in mind that these graphs represent consensus for specific numeric values of parameters needed in physical equations and measured by repeated lab experiment!

In contrast, some other disciplines have been undergoing a "replication crisis", in which entire swaths of undergraduate curricula are being undermined wholesale. The infamous Stanford prison experiment, a mainstay of psychology, is credibly accused of being outright fraudulent.

Given the reality that these phenomena of conflicting incentives exist at every step of the process of producing scientific data, there is a kind of meta-question: To what extent do these incentives influence the approach we take, from the most minor unconscious omissions to intentional fraud? In general, people on the right tend to expect people to respond to incentives, while people on the left tend to expect people to respond to ideals. When these conflict, expectation of outcome will differ.

To what extent is technocracy a good mechanism for making complex decisions?

Konrad Lorenz (among others) is credited with the remark that

Scientists are people who know more and more about less and less, until they know everything about nothing.

A simple policy question with competing criteria is what sort of approach to take when handling the COVID-19 pandemic: what kinds of restrictions should be in place where and for how long. Epidemiological experts have provided opinions regarding the likely disease effects of different measures.

Even assuming perfect models, however, such decisions aren't made in a vacuum. As raised by this previous question, restrictions of various sorts have very serious costs. For example, it is well established that unemployment raises suicide rates.

If a certain containment policy is expected to prevent X deaths from disease, but the likely economic damage will also cause 0.75X more suicides, is the policy "worth it"? What if the policy is not expected to prevent the deaths in the long term but merely to slow their rate? How do you mix in all the other tangible and intangible costs, like non-suicidal depression, lost opportunity at life events, delayed education, and damage to communities? And all this is presuming that the experts' advice on outcomes is reliable!

The opinion on the right is generally that while experts are helpful at predicting likely outcomes of various policies in narrow terms, expertise is nevertheless narrow, and experts are likely to overlook unintended consequences. Furthermore, expertise is not especially useful in making value judgments, such as "what is an appropriate level of risk to take on in order to gain a benefit?" For these, those on the right would rather make decisions at as low a level as is practical (e.g., epidemic policies appropriate to Wyoming are not appropriate to New York, and policies appropriate to New York City are not appropriate to upstate), and have value judgments made by people the most responsive to those affected by them—individuals where possible, and elected officials before appointees.

  • Comments are not for extended discussion; this conversation has been moved to chat. – Philipp May 11 '20 at 7:37

This Q. is an abstraction of previous Q.s about why conservatives in 2020 aren't conservationists like:

Different US states and regions favor different styles of religion and culture. The Texas oil boom provided that state with a kind of first mover advantage. Thus the culture of the national and even multinational oil business as a whole tends to disproportionately reflect the relatively bolder values (both secular and religious) of wealthy Texans. And the oil business is a primary funder of the style of political lobbying and propaganda that supports such partisan scientific divides.

If this theory is correct, we might speculate that if from 1901 on Texas had had no oil, but Massachusetts did, then the first-mover oil barons would be old-money New Englanders whose more liberal political propaganda would support a more technocratic style of divide, and the Texans might have become conservationist conservatives.


It's important to note that the study in question doesn't address the issue of whether a person has confidence in the scientific method, or whether they support any actual policy advocated by any actual scientist. It only addresses the issue of whether a person says, in the abstract, that they do. So the question really is, how did democrats come to think of themselves as the party of science?

I don't have numbers on this, but I can tell you that when I was a teenager in the 90s, Christian fundamentalists featured largely in the public conversation. Their unscientific views on evolution and climate change, and their opposition to any scientific study of human sexuality were much discussed. Democrats had (and continue to have) wildly unscientific views on gender differences, the origins of racial inequality in the US, guns, and the effects of minimum wage laws, but those things weren't (and still aren't) usually thought of as subjects of scientific inquiry. A lot of people seem to think those are matters of opinion.

So why do some unscientific ideas get you called out for being unscientific more than others? Some of it has to do with how hard the science is. Measuring the age of the Earth, for example, depends on some pretty basic physics. There is no serious question about it in the scientific community because it's so simple. Soft sciences are more difficult because there are more variables to control for. There is serious question in the scientific community about some aspects of gender. It's more obvious when someone has ideas that contradict a hard science.

There's also the issue that hard sciences tend to be less personal. No measurement of the age of the Earth is going to tell you that you're a bad parent. But a study of gender differences might. So it tends to be a lot more incendiary to make definite claims about soft sciences, and people tend to avoid doing so out of politeness.

So, democrats tend to check the "Science is good" boxes on surveys because 1) their unscientific beliefs are less obvious than the unscientific beliefs of republicans and 2) those subjects on which their beliefs are unscientific tend to be those that people avoid talking about in scientific terms.

(I make no claim that this is a complete answer to the question.)

  • 3
    What is your definition of "unscientific"? You seem to be conflating natural sciences with the humanities – you mention gender theory, history, and economics as the fields in which Democrats hold "wildly unscientific views". But how can you hold an unscientific view of history? And how can you claim that our understanding of guns, gender, and the minimum wage are so settled that disagreements with your preferred policy are "unscientific" in the way that disagreements about, for example, evolution or the (broad) effects of increasing CO2 concentrations in the atmosphere are? – divibisan Apr 24 '20 at 18:24

The partisan difference goes broadly with the political ideologies of the two sides, large State vs small State, rights vs liberties, intervention vs self-help. Higher tax vs lower tax etc etc.

Science-based policy fits better with the Democrat ideology than Republican ideology.

Republicans have more reason to be induced in to circumspection of science and consensus when it seems to threaten their liberties (albeit merely in the education of their children).

But it's not that the Republicans are necessarily being bloody-minded in their circumspection. Statistics is infinitely malleable, prone to vested interests,and hard to get right, while too much consensus science of the past with excellent pedigree, has proven false.


The public have had their faith in science shaken by various reversals, starting with the great scandal of stomach ulcers/cancers and Helicobacter Pylori, to saturated animal fat and vascular/heart disease. The latter was a State-sponsored dogma. It looks increasingly like fruit is next (if you don't believe me, consider that fruit is mostly sugar and fibre).

What's surprising is that Democrats are not themselves more circumspect and knowledgeable about the frailties of science. But this may be one of a piece with believing in a bigger State, that they are also more trusting of a given authority. (Amusingly it may also be that they are pleased to accept science-with-coercive-policy-implications because it riles Republicans.)

Meanwhile for Republicans....

If a scientist or Expert says "We know that..." this is an attitude that will inevitably get a strong negative reaction from those not ideologically well disposed to public policy implications of statistical findings presented as fact, especially if they affect religious sensibilities. And that, in the main, is going to be a Republican issue.


As previously mentioned, confirmation bias.

There may also be another effect though: the preference / excitement about "new" things VS preference for "the old way" i.e. conservatism in a classical sense of the word. Some studies indicate a fairly consistent ratio between the people in a society that just like new ways and things, and the ones who like things the way they are / were / dad did it. It is probably good for the stability of society to have ways of changing but not so fast it stops being a society.

This kind of conservatism is more sceptical of change and prefer status quo and conformism, and has a significant overlap with political conservatism. Hence, with science having all these new things, and new thinking; pushing for change in how things used to be done, I think people who are conservative find appeal in political conservatism and mistrust science and change in some cases. Likewise for some the "certainty" and limited change of faith and religion may provide the answers and security they crave where science with its changing less certain world view probably makes them uneasy at some deep level.

This like or distrust of change is probably at the foundation. Then add politics and emotion, and it is easy to discount what does not fit your goals be it science or other "facts" for all sides of political spectrum, or cherry-pick the science "facts" that support your position. Even IN science confirmation bias is a real obstacle. Scientific data to be published also has cherry-picking.

Humans just seem to be political animals, so even science is political to some degree.


Start with Hofstadter’s Pullitzer winning book Anti-Intellectualism in American Life (1963) to understand how Protestant evangelicals rejected science and intellectual study. Given that they are the “base” of modern conservatism in American culture, I suspect you’ll find the core of your answer there.

The Pew Research Center has reported the partisan divide in the US towards scientific experts and expertise. For recent results:

While anti-intellectualism plays an important part in the issue, it doesn't completely answer your question, because many people wouldn't identify themselves as conservatives but would reject science when it doesn't fit their beliefs. Many conservatives may not be simply rejecting science, but handpicking only the (perhaps bad) science that fits their interests, which likewise happens outside conservatism too.


I think there is a strong philosophical rationale for why Republicans are less likely to be influenced by scientific evidence then Democrats.

Republicans are more likely to come from a conservative political tradition, and conservatism can be considered to be a reaction to the beliefs of the enlightenment, which were pro-scientific to a very great extent. Utilitarianism, Socialism and contract theory were three enlightenment systems very much based around the idea that a scientific approach to policy can solve all political problems, they arose due to the great progress during the enlightenment period (17th to 19th centuries). The enlightenment could be considered to have reached its peak with the French revolution, which led to excesses (there was the terror, and strange things happened like the decimalisation of time and renaming of the months). The revolution led to a reaction - the publication of Reflections on the Revolution in France by Edmund Burke, which is considered to be a defining conservative document.

Conservatism is more based around the idea that communities and traditions have evolved over time to fit the needs of populations, and should not be changed to fit what are considered to be abstract external notions. Judges and case law, and political evolution are how views change, rather than based on an idea from a philosopher.

This also provides some explanation why religious groups are more likely to be conservative (religions tend to be rooted in tradition, rather than any scientific basis, and most of the enlightenment traditions paid little attention to religion). As there is a traditional element to conservatism, conservatives are more likely to be patriotic and inward looking. Enlightenment thinkers are more likely to believe in root and branch reform.

Note most of my ideas here are based on Ian Shapiro's course on the Moral Foundations of Political Theory, which I would thoroughly recommend.

  • 6
    Hi. I downvoted because I disagree that this is a one sided issue. There are cases where it flips around. Republicans are, in agreement with science, more acceptant of GMO crops and their benefits - for instance. – Stian Yttervik Apr 24 '20 at 11:54
  • 9
    @StianYttervik I'm not sure that's actually true. There doesn't seem to be any difference in trust of GMOs by politics, only education level. Where the parties differ is on labeling, which hits more at ideas on government mandates, trust of corporations, and transparency. – divibisan Apr 24 '20 at 14:30
  • 1
    Perhaps my comment can be seen as defending the conservative viewpoint, I think it is valid to do so as the pro scientific argument is so easily made (we owe our prosperity to science, and I work in scientific research), I personally find the concept of anti enlightenment thinking very hard to follow. However I think there is a valid argument there, and so I try and make it. It can be valuable to try to understand other people's points of view, especially if you wish to build a political discussion community. – dilaudid Apr 25 '20 at 10:31

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .