We’re teaching the scientific method incorrectly


As a scientist, I am often doing science. It’s my job. I know science. By any measure, I’m as much a scientist as any other scientist.


The scientific method as incorrectly taught to our children. Image from Washington Prep High School, Los Angeles Unified School District. photo: T. McGlynn

But if you look at what I do on a day to day basis, it looks absolutely nothing like what people think science should look like. Fixing misconceptions about science requires much more than correcting stereotypes of what scientists look like, though that’s a great start. (By the way, here’s my entry to This is what a scientist looks like.)

Science is taught in school as a linear process. In practice, it never is a linear process. It’s not even a linear process in the labs in which we teach the scientific method.

I was in a high school classroom last week, and on the whiteboard of this classroom was that odiously wrong conception that we see everywhere. I see this all the time, and if I was doing my job better I would openly contest it every time I see it.

In its stead, let me share with you what it looks like when I am doing science:

The Method of Science, as it happens in my lab

The Method of Science, as it happens in my lab

What science really looks like is a little more complex than how it is marketed by publishing companies to our children when they are in school. They’re not training kids to be scientists, they’re selling textbooks to teachers who are not scientists. Many of these teachers are reluctant to teach science because they are not adequately prepared, and because their bosses are making them overdose on math and English to maintain test scores.

Teaching science isn’t easy for those who aren’t used to doing science, so this cute linear process that kids see in school is what publishers have done to make teaching science as simple and boring as possible.

How do our kids really learn what science is? By actually doing science. By having teachers that understand science and do real science with them. What is often missing is the red arrow in the figure above. Even those who buy into the linear model of science need to realize that it is cyclical, that answers lead to questions. It’s not a plodding march of progress. It’s a messy tumble and jumble forward in which new information leads to even more confusion, but with broader horizons. Science expands the circumference of our ignorance.

How can teachers get our kids to do science if they don’t even know what real science looks like? We need to teach real science to the teachers. We can do this in college, but if you look at the science coursework that is required by future elementary- and middle-school teachers, you’d be either dismayed or outraged. This is the starting point in fixing the science education crisis in the US. We need elementary and middle-school teachers who understand, enjoy and prioritize science.

As scientists in science departments, we have the latitude to seize this curriculum and teach these classes the right way, and by the right people. And we can make sure that people don’t leave our classes without understanding and being excited about science. We can make sure that they’ve been involved in a genuine science experience. We can use genuine inquiry in our teaching.

We also can skip the middle man and do science with current teachers.

This summer I’m taking one of many small steps. I’m having an experienced master teacher at the middle school level joining my group in Costa Rica for a month. He should go home with a better idea what science looks like, I expect. If you want a teacher in your lab, and you’re one of those (declining few) with federal funding, just call up your program director and you probably could get hooked up mighty quickly once you find your teacher. To find a teacher, just ask around, and many will jump at the chance as long as they’re getting paid. Even if it’s just a lot of pipetting. Having a teacher in your lab can change science education for hundreds of kids in a short period of time.

To be clear, I’m not the only one who has this idea in mind. The more of us working to explode the notion that science is linear, the more opportunity kids have to get to do real science.

11 thoughts on “We’re teaching the scientific method incorrectly

  1. Hi Terry – great post. I share your dislike for the simplistic method and have spent a lot of time educating undergrad and grad students away from it. But I agree we need to go further and educate K-12 and the public away from it.

    I think we scientists have a love-hate relationship with that view. Love because it makes it seem like we have a magic formula that differentiates us from the muddling through approach everybody else uses. But we should hate it because if we leave it as what people expect us to do then when we don’t do it (or more precisely when they catch us not doing it because we never do it), then the public feels betrayed and distrusts science. In short this simplistic view in peoples minds I think has been reinforced by scientists as much as anybody, but I would argue it is ultimately very bad for the public view of science and scientists

  2. Just to be contrarian: it can be argued that one weakness of the “real” method of doing science, as compared to the idealized way, is that the real way confounds hypothesis generation and hypothesis testing and so totally screws up your ability to reliably differentiate random noise from real signals. 😉


    I’m semi-kidding here, of course. But only semi (and I include myself among those who use the “real” method)…

  3. I wonder if both models are really that different from eachother, except for the possible (and possibly very important emphasis on) the order of things relating to hypothesis testing. If I understood correctly, it may be important in hypothesis testing that you have a hypothesis upfront and that it might be important, from a statistical standpoint, to engage in testing this properly (e.g. see above post + link). So, this part of the “real scientific method” would be point 2 (“form hypothesis”) & 5 (“analyze data”).

    What I don’t understand is the emphasis in “the method of science in my lab” (red arrow) in the model and in the text. Isn’t that part also part of the “real scientific method” (i.c. 1: “state the problem/question” and 6: “draw conclusions”) only the locigal results of these points (i.c. that the questions at hand leads to certain specific answers/conclusions which leaves one with further questions) isn’t contained in the model of “the real scientific method”. I think this is because these “linear” steps would then be followed all over again, i.c. starting from point 1 “state the problem/question”.

    The rest of all the terms and arrows in “the method of science in my lab” are all fine and good, but I wonder if they truly represent a different method from “the real scientific method”. I reason they might represent more of a different use of terms in combination with viewing matters on a slightly different level. To give an example: “the method of science in my lab” contains the terms “questions” & “learning what is already known” which have arrows between them. I wonder if they both nicely fit within point 1 of “the real scientific” method “state the problem/ question”. The only possible difference is that they are perhaps depicted on a slightly different level of specificity (i.c. you could if you want to, change the ‘/’ sign in 1. “state the probloem/ question” into two arrows).

    • The way that *kids* see the classic and flawed model is that scientists have an idea, that arrives magically. Then, they do an experiment to test that specific idea. Then, they magically have an answer. And the process is over.

      I highlighted the red arrow to couple with the notion that science is iterative, or circular. A conclusion, in its essence, leads to more questions. This is what every scientist knows and what almost no kids are taught in school.

      The reason that there are arrows between “question” and “learn about what is already known’ is to emphasize the fact that we refine our questions even before running an experiment. (Also, during an experiment as well. We don’t wait for it to be all over before we rethink our questions.) In the classic model of “state the problem” it’s not shown to the kids that this formulation of the problem (my “question”) is informed by what we already know and what earlier experiments and observational work that we have done has shown us.

      “State the problem” is NOT the starting point. But that’s what kids are shown. How do I know that? Look at the whiteboard. Talk to a kid who is learning science in middle school. That is absolutely how science is never done. The problem comes from many things that are part of the scientific method. Exclude those arrows, and kids feel that they’re not meant to be scientists because they don’t have the problems emerge from the ether. None of us do, but the kids don’t realize that.

  4. Would it help if we called it the “experimental method” or something like that? I haven’t given this notion that much thought yet, but it seems like it does fit nicely into a small portion of the overall scientific process you depict in your diagram. Yours seems to summarize a research program, whereas perhaps this method taught to so many is intended to reflect approaching a single problem. In that case, the problem would not be that this method is wrong, per se, but rather than it is an element of something far more messy, unpredictable, and exciting. To that end, I think the best method of teaching the scientific method is to jump directly off an observation (notice something that raises a question, figure out how to learn more about it.) Having judged middle school science presentations, most kids seem to pluck a question out of nowhere (about half of them were testing which random household products would kill bacteria).

    • Great point. The questions that kids typically have are often boring, and often prompted by others. Starting with observation would be a great start.

  5. From Lomolino, Riddle and Brown “Biogeography”, 3d ed., page 7:

    “Most people have a vague and perhaps misleading impression of what science is, how scientists work, and how major scientific advances come about. Philosophers and historians of science, viewing its progress with 20/20 hindsight, often suggest that it is possible to give a recipe for the most effective way to conduct and investigation. Unfortunately, as most practicing scientists know, scientific inquiry is much more like working on a puzzle or being lost in the woods, than like baking cookies of following a road map. There are numerous mistakes and frustrations”.

You can leave a comment anonymously, just don't give your name or email.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s