Education research denialism in university STEM faculty


Scientists regularly contend with irrational denialism of simple facts. In our classrooms, communities and the media, we hear patently absurd things like:

This is the logo of the Flat Earth Society.

This is the logo of the Flat Earth Society.

  • The world isn’t getting hotter, or even if it is, it’s not from carbon emissions.
  • Humans didn’t evolve from nonhuman ancestors.
  • Transgenic foods are unsafe.
  • Vaccines cause autism.

Any scientist who operates on the basis of evidence will regard those ideas as total bunk*.

But, as I’ve mentioned before, people aren’t rational beings. Especially when their emotions are involved (and they usually are), they’re not prone to think an expert is correct if their intuition tells them otherwise.

And scientists are people.

So I am not entirely surprised, but I am disheartened, when scientists are guilty of their own flavor of denialism: Education Research Denialism.

Here’s a selection of things that I’ve heard many scientists claim to be true:

  • If a lecturer is good, then student attention span does not wane over the course of an uninterrupted hour-long lecture.
  • If you ask a question to your class, students learn just as much from volunteering to answer by raising their hand, compared to having all students discuss their answer with the person sitting next to them.
  • Active learning methods only increase student learning if the professor isn’t good at lecturing.
  • Whether a professor is likable does not affect student learning

Based on my familiarity with the education research literature (which is admittedly selective and heavy on secondary sources like these two), it seems pretty clear that those statements above are patently false. Research has shown them to be untrue. Counterfactual. Not correct.

If you’re one of the (many) scientists who denies some of the above facts, then what’s the reason for this denial? Here are my guesses that might apply:

  1. You reject the validity of education research that created these facts.
  2. You reject the notion that published education research findings can apply to what happens in your own environment.
  3. You trust your own experiences, intuition, and non-peer-reviewed personal investigation over the findings published by education experts who do this research for their careers.
  4. You suspect the research is driven by an pre-existing educational agenda and that the studies were conducted in a way that cannot adequately refute findings that go against the agenda.
  5. Education research is so filled with jargon and tautological ideas that it’s not possible to generate findings that outsiders can find to be objectively true.
  6. Education researchers have to pretend that their findings are novel and important to get funding, earn money as consultants, and to advance their careers.
  7. Educational researchers are all talking to one another so closely that they are in an information bubble and are not willing to consider outside ideas.
  8. There is a social pressure among the education community to accept these ideas as true because it would be “politically incorrect” to say otherwise**.
  9. Many people have been operating under the notion that that this education research is not true for a long while now and everything seems to be going just fine.
  10. What constitutes educational fact might just be a matter of personal belief, and your sources of information are widely held and respected and are more valid than the relatively new educational ideas that might be short-lived.

Now, I ask you to browse through items 1-10 again, but this time, read them and whenever you see the word “education” you substitute it with “science,” and whenever you read “educational,” you substitute it with “scientific.”

Aren’t these the arguments that you have to suffer from from your science-denying uncle-in-law at family events and on Facebook?

The reasons that scientists deny education research findings are the same reasons that other folks deny scientific research findings.

If you think that scientific research is fundamentally different than education research, and that what we learn from the field and the lab is different from studying human beings within the classroom, how about we heed the words of Jeremy Yoder: “If you are, like me, someone who studies things that are not humans and you’re interested in studying humans, you absolutely need to talk to people who do that for a living.”

A non-scientist denying scientific results isn’t all that different from a non-education researcher denying educational results.

Yeah, there is a lot of annoying jargon pushed by corrupt consultants with big egos who misuse information to push their own agenda. Do you think that the previous sentence is about scientists or education experts? Hmmm.

Next time you find yourself talking about what works in the classroom and what doesn’t, you don’t want to find yourself saying, “I’m not an expert researcher in the field but…” because that’s precisely what republicans are saying on Fox News every day about evolution and global warming. Consider whether your argument is more valid. And if so, can you couch it in a way that has more credibility?

Do I believe everything that education researchers tell us is true? Of course I don’t. Because knowledge isn’t a matter of belief. I don’t believe in education research any more than I believe in global warming. There are just a bunch of facts that I accept on the basis of evidence — and to some  extent, yes, taking the word of experts.

I haven’t done the phylogenetic reconstruction to confirm for myself that the closest extant relatives of whales are hippos. But you know what? I’ll take their word for it. I haven’t run all of those experiments to show that after fifteen minutes, students start learning less in a lecture. But I’ll take their word for it.

It’s okay with me if you feel like denying education research, but if you do, the quality of your argument better be a helluva a lot better than the silliness you hear from science deniers.



*Just in case you didn’t get the memo, let me be really overt that those statements above are wholly, entirely, factually incorrect. Each one has a mountain of evidence supporting the exact opposite. People burning carbon are making the world hotter. Humans are evolved from other apes, and even from unicellular organisms. There’s no evidence showing that transgenic foods (like “GMO anything”) are a risk to human health, and many studies so no association between vaccines and autism, and the one study that had purported such a link was retracted because of falsified data.

**Whatever “politically correct” means. I think it just means treating people with respect.

23 thoughts on “Education research denialism in university STEM faculty

  1. … I have colleagues who think these things?!

  2. … by which I mean ‘I have colleagues who are education research denialists?!’

  3. All really good points. I haven’t taught much, and so haven’t considered my own ‘beliefs’ about teaching much, but I definitely hear the points you make from colleagues on occasion. I’m sure a lot of denial is due to the things you mention. But I wonder if some of it may also be statistical averages versus the individual. For example, if the research shows that, on average, method 1 is better than method 2 for learning in the classroom, that doesn’t necessarily translate to method 1 is better than method 2 for learning when I teach in the classroom. Just a thought.

  4. Terry, does this post represent a change of heart from your other recent posts on this topic? You’ve previously written that you’re not a fan of attacking those who don’t draw on modern pedagogical research in their teaching, because calling people “unethical” or whatever isn’t going to convince them to change their teaching methods. But now you’re comparing profs who don’t adopt modern pedagogical methods to climate change denialists, which at least skirts close to being an attack (and surely runs a pretty high risk of being seen as an attack, even if that wasn’t the intent).

    Or maybe I’m misunderstanding something, perhaps missing that you’re aiming different posts (and their associated rhetoric) at different audiences?

    To pick a handy concrete example: would you consider this old post of Brian’s (especially his point #8) to be the equivalent of climate change denialism?

    I’m asking to better understand where you’re coming from and what sort of claims you’re targeting with this post. Not trying to goad you into a fight with Brian! I wouldn’t have thought that a nuanced post like Brian’s would be the sort of thing you’re targeting. But your analogy to climate change denialism has given me pause, since more sophisticated climate change denialism often makes some of the same sorts of points Brian makes (e.g., highlighting possibly-important confounding variable existing research doesn’t much consider).

  5. It’s no change of heart of mine – I still think there are many ways of being an effective instructor, and it’s possible to eschew many approaches and still be effective.

    I am not saying that people who don’t adopt modern pedagogical methods aren’t akin to climate change denialists. I’m saying that people who don’t accept facts that are well-established by research are akin to climate change denialists.

    It’s possible to be an effective teacher without using some contemporary approaches. However, the research does say that one can be more effective by making some small changes. I am not of the opinion that pushing for someone to make changes in a prescriptive fashion will be useful. On the other hand, working to get people to accept straightforward and well-established facts is a foundation of academia.

    There are millions of people who accept that climate change is real but are flagrant carbon polluters and vote for republicans. And I imagine there are some people who accept educational facts but still don’t change their teaching. I understand that it many folks think it takes a lot of time and effort to change how you teach (I don’t think switching to some active learning approaches doesn’t take more work, but experienced folks have disagreed with me on this), so I can swallow the notion that someone can accept the facts from the researchers but not make the personal changes.

    So if someone says they won’t make changes because they’re curmudgeonly and it’s too much work and it’s not worthwhile? That might not be wise but at least it’s accepting of the facts.

    So I don’t intend to be judgy of anybody who isn’t changing their teaching. I’m just being judgy of those who are rejecting facts that are well-established in the literature.

    How can you tell what are the facts in education research are?

    Also, the books I linked to in this post are very solid.

  6. Thanks Terry, I’m with you now.

    I’ve never encountered anyone who just dismisses modern pedagogical research as a load of hooey. But that presumably just shows we move in different circles.

    Anecdotally, my impression is that blanket denial of the validity of modern pedagogical research is not one of the most important obstacles to adoption of non-lecturing approaches. Certainly, it’s not for me.

    It might be interesting to do a poll or something to get some anecdata on this. Ask people: how do you mostly teach, and why do you teach that way. I suspect you’d get a wide range of answers.

  7. I wonder how many of us work with deniers? Here’s a one-hour twitter poll. Curious to see what rolls in. I have no a priori expectations:

  8. I do recognise this, to an extent in my own practice. Not because I am an education research denialist ( I tend to have Ed research projects running alongside my STEM research), but mainly because implementing changes takes a lot of time and sometimes leaving thing things as they were is just easier (even if I know they are not ideal). I am only human after all. Change is a slow beast.

  9. Guilty of 3. I think the reason is that I always hated it when any of those new methods were used in teaching – and of course it’s hard to emotionally understand that I am not the average learner. Yes, yes, you can tell me that for most learners X or Y gives better results and I can intellectually understand that many (most?) learners are different from me but try telling that to my gut feelings. My gut feelings tell me that everybody learns just like me (or is just a stupid person, unable to learn anything, ever) and so any method that I would have hated as a learner must be bunk.

  10. There’s a significant difference between research in a pure science and between educational studies. The latter deals with far more variables that cannot be controlled. That’s not to say that there aren’t valid pedagogical finds, but what I’ve learned form the best is that teaching is a craft that has to make use of many tools. From that perspective, the lecture versus active learning or tech versus books or student vs subject-centered polarities are artificial in most cases. When practitioners take one approach to an extreme, mediocrity is the norm regardless of the method adopted. Teaching is always compromised by our own imperfections, by sheer numbers of large institutions that pull K-12 and even college education in all sorts of directions.

  11. Ok, poll in the queue for next week at Dynamic Ecology, asking readers if they mostly lecture, and if so, why. It won’t be a random sample of any well-defined population, of course, but I will be very interested to see the responses anyway.

    To test my precognitive abilities, some predictions:

    -I predict the majority of respondents will mostly lecture, but I don’t think it will be a massive majority (say, >90%). I predict this in part because I suspect the DE readership skews towards people with an interest in modern pedagogical approaches.

    -I predict the most common reasons for not lecturing will be (in no particular order): it’d be too much work to teach any other way; lecturing seems to work adequately for my students; I don’t know how to do anything besides lecturing.

    -I predict that “I deny the validity of pedagogical research” will be one of the rarest reasons for lecturing (not that you claimed it was a common reason, Terry. But it seems to me quite possible that many people might know a pedagogical research denialist without pedagogical research denialism being a common reason for people to lecture. For instance, many people know someone over 6′ 2″ tall, but few people are over 6′ 2″ tall.)

  12. Jeremy, I also predict that “I deny the validity of pedagogical research” won’t get many clicks, given the readership of DE. Nor would you have many people admit to such a blunt statement if you asked around at a meeting. (The question as you give it corresponds to Terry’s #1.) But I’ve certainly heard and seen many statements to the effect of Terry’s #2 or #3 — the pedagogical literature might be valid but that doesn’t mean it applies to Course X for whatever reason. (Material is not suited for anything but a lecture, I am such a good lecturer that it doesn’t matter, I tried something else once and it sucked, those aren’t students at my university)…

    That would be more analogous to “well maybe most vaccines aren’t unsafe but I’d rather not vaccinate my kids anyway.” The statement protects the speaker from complete science denialism (you didn’t ask them “is all science bunk?”) but then invokes personal choice to have an intuition about something in particular.

  13. This “There are just a bunch of facts that I accept on the basis of evidence — and to some extent, yes, taking the word of experts” is a problematic way to form beliefs. Experts within homeopathy, chiropracty, nutrition, social psychology, etc. etc. practice a sort of evidence-based science and if we trusted these experts, we’d believe many wrong things. What is happening within these fields is a combination of many of your reasons to be a “denier”. So where does education research stand? Are the studies focussed on p-values (probably) or effect sizes? How generalizeable are the results? Are alternative explanations tested with well-designed experiments (i.e. an infinite number of models can explain the rejection of a null, but how are these alternative models tested?)?. I’m not a denier, but I lean toward skepticism, given the few studies that I’ve read in detail.

  14. Jeff, I think your “sort of” invalidates your point. I’m using evidence from well-designed experiments with appropriately robust analyses that consider alternative hypotheses. Which does characterize a bunch of the education literature. I’m a scientist, dude. Also, as I’ve already pointed out, I’m not forming beliefs. I’m accepting facts. The distinction between belief and fact is pretty much something that people discuss a lot when it comes to science denial.

  15. @Abigail:

    That’s why my little polls offers lots of different choices to answer the question “why do you mostly lecture?”, and lets respondents pick up to the three most important options.

    As for people who think the main conclusions of the pedagogical literature don’t apply to them and/or their students, you seem to think that’s just rationalization and excuse-making on their part (apologies if I’m misunderstanding you). And in some cases I’m sure it is. But in other cases it might be people making defensible professional judgment calls. For instance, if somebody, say, tries out non-lecturing approach X, and it doesn’t work, and then they tweak it and it still doesn’t work, surely at some point it’s at least defensible–I’d say advisable–for them to go back to lecturing. This is not an unrealistic hypothetical by the way–it’s currently happening to me. I just taught intro biostats at Calgary as a flipped course for the first time. I thought it was working well–then I marked the midterm. Turns out it’s working worse than when I lectured 100% (with occasional clicker questions tossed in). I’m not going to give up on the flipped structure after one go. But if I keep trying it for another year or two and it’s still not an improvement over lecturing, yeah, I’ll probably just go back to lecturing. In the event that admittedly-hypothetical but very real possibility comes to pass, tell me: am I just rationalizing and making excuses for my own partial or complete “denialism” of the pedagogical literature? Or am I doing what every good teacher does: making professional judgments as to how best to teach my students, based on inevitably imperfect information, which includes but is not limited to the studies that comprise the pedagogical literature?

  16. My point was that we cannot trust experts in all fields just because they call themselves experts. They may be fooling themselves. Is this point invalid? Here is my issue with education research (see also the link to Brian McGill’s post from Jeremy Fox above): Any causal network in a system involving humans will be very complex, meaning that many (most?) causal relationships will be highly contingent and poorly generalizable. This is what makes human medicine, economics, sociology, and psychology so difficult and why within these fields a kind of cargo-cult-like science culture can easily arise. At least in some of these fields there has been outside policing (Andrew Gelman on social pyschology or Science Based Medicine on alternative medicine) and even self-policing (John Ioannides for example in medicine or nutrition more specifically). I wouldn’t think education research is any different from other human-science fields. I would think many education results would be conditional on sex/age/personality/knowledge/charisma of teacher, time of class, day of week of class, size of class, composition of the students and their myriad histories and experiences, level of the class, etc., etc. etc. Certainly, some effects are large and generalizable, but which? I’m not skeptical of education research because the findings go against my politics or I’m too lazy to incorporate some of these practices (I do!). I’m skeptical because I’m not aware of a self-policing, best practices culture within the field. Who is the John Ioannides of education research? Or maybe the field doesn’t need one because the bulk of the research is based on “well-designed experiments with appropriately robust analyses that consider alternative hypotheses”.

  17. @ Jeremy – your experience with the flipped class highlights an example of why I’m skeptical. Tests during the semester are a proxy for what was learned. What if the tests are consistently worse using a flipped classroom but the students do better in upper level classes because of better quantitative skills that they acquired from a flipped but not conventional class? Or what if the students are better prepared for working in biotech or Ph.D. research because of the flipped but not conventional class? Or what if effect is heterogenous (it has to be, the question is how much), for example, 2/3 of the students do worse on tests but 1/3 do better on tests? What if it’s the lower end that does better? What if it’s the upper end that does better? Who should you target? Should the target vary among schools with different demographics? I recognize that isn’t novel thinking, but it means there are few bullet point answers to best practices

  18. Jeremy, yes, I think there’s a fundamental difference between “I tried this and assessed it properly and it is not working” and “I know based on assumptions (gut feeling) that this will not work.” The more I think about it, the more it seems like the difference between troubleshooting one of my lab methods properly versus refusing to, say, optimize the food level I give my animals because that Couldn’t Possibly be the problem.

  19. @Abigail,

    Fair enough. Except that trying some forms of non-lecturing even once is a lot of work. Even if the data say “approach X works better than lecturing on average”, there’s variance around that average and so it’s reasonable for an instructor considering approach X for the first time to be somewhat risk-averse. Reluctant to invest a lot of prep for an uncertain payoff. Especially if lecturing is working adequately for you.

    Put another way, the switch from lecturing to, say, a full-on flipped classroom isn’t really analogous to trouble-shooting your lab’s existing methods. It’s analogous to switching to a whole new system and the associated brand new (to you) methods, which you’ll then have to trouble shoot. Making that kind of big, risky switch isn’t an easy decision, even if the expected (average) payoff of the switch is positive.

    Which is maybe an argument for trying out those forms of non-lecturing that can be tried out relatively easily. For instance, breaking up your lectures with some clicker questions or short pair-and-share exercises isn’t much work. Of course, my own experience was that those sorts of tweaks didn’t make any detectable difference to student mastery in intro biostats. Small, low-risk investments have small expected payoffs, as it were. That’s part of what motivated me to take the plunge and go to a full-on flipped class ( A risk that so far hasn’t paid off. I’m still glad to have taken the risk even in retrospect, in large part because I was very fortunate to be able to draw heavily on someone else’s prep. I was playing with “house money” in a sense. But I wouldn’t question the judgment of anyone who decided not to take the same risk, especially if they had to do all the prep themselves.

  20. To ericouva, I challenge the idea that there are fewer variables in science research. I think we are just accustomed to making assumptions about a large number of variables and then not explicitly saying so, while in education research we are far more explicit about them.

  21. Thank you for your blog. An additional note — practices that enhance learning are the same ones that make all students successful in the classroom. They maintain rigor and close achievement gaps. As a professional developer of college faculty, I have begun to (gently) ask my faculty how they balance the time it takes to alter teaching practices with the idea of social justice and equity. It makes many of them pause. Here is more information:

  22. Poll results in. Also, some folks were kind enough to share links to larger and more rigorous surveys on the same topic. Headline result: education research denialists exist but are rare (as are people who lecture almost all the time). People who mostly lecture, mostly do so for reasons having nothing to do with education research. The most common reasons for mostly lecturing are things like class size, lack of TA support, desire to use existing prep, lack of formal training in other teaching methods, and risk aversion.

Comments are closed.