How Do You Teach Critical Thinking?

Give them labs to do and lie about the expected results.

In a paleo class I teach, we'll cover a topic and then watch some highlights from a Discovery Channel show on the same issue. A very, very good way to help students begin to recognize fact and fiction in many of the"documentaries" out there.

Read, read, read. Write, write, write.

Have them read material on both sides of a controversial issue and write a compare/contrast essay.

In labs, have them write up their expected results with a supporting rationale before conducting the lab work.

Where possible, explore connections between different areas of study - biology and chemistry, chemistry and physics, history and politics, for example.

I have to say, Johan, I never want to be your student.

Besides lab experiences, I would say discussing research articles. One of my advanced science courses that is all it was. It developed my skills in reading and evaluating an article as well as to think about what experiments I would do. Really put me at an advantage when I went to graduate school. Where I am at now has such a course for their undergrads and a more intense version for their first year grad students.

Critical thinking evolves when the thinker's conclusions come back at him good and hard. President Obama today released $75 billion to reward those too stupid to responsibly run their lives, checkbook vs. income. "Orthodoxy means not thinking - not needing to think. Orthodoxy is unconsciousness." George Orwell, 1984.

Critical thinking is heresy, blasphemy, and treason. The singular biblical sin enraging otherwise disinterested Yahweh was eating of the Tree of Knowledge. Sodom earned a historical slot for sins of the flesh. Gommorrah was expunged for sins of the mind.

I've never like English class and that's partly why I became a scientist, but I did have one really great English teacher. Her main point of the whole course was "question your own assumptions". That one idea has been useful in many areas of life, from lab work to political opinions.

Give them labs to do and lie about the expected results.

In my engineering labs, it was common to get answers greatly different from literature values. It was OK as long as we were within an order of magnitude. My teacher called this "engineering accuracy".

I think the best way to do it personally (meaning: what worked on me), was interesting problems with non-intuitive solutions. I'm not talking advanced mathematics here, just something as simple as a blacksmith's puzzle. You can play with it forever and hope you get lucky, or you can look at the shapes and think about how they interleave and how to unlock them. That's the simplest and easiest form, but even fairly simple logical puzzles on paper can achieve the same thing.

Put me on the "question your assumptions" bandwagon. As an adjunct picking up primarily political/social classes, the best tool for me is encouraging the students to realize that their beliefs have a history and are to some extent influenced by the student's environment and interactions. We do interview projects so they understand how their views (especially political ones) are shaped by those around them and then ask the students to interrogate their own perspectives. Incidentally, this is also a good way of getting them to actually form an opinion about something like the utility of American power in the contemporary world when they haven't really thought of the topic systematically before. Once you can think truly critically about yourself, the transition to doing it about something else is easier. I appreciate, though, that this kind of an approach is not really applicable for something like chemistry.

Carl Sagan's "Baloney Detection Kit"

I think reading and critiquing the work of others can be really helpful, and can also be done at any level. (By "others" I don't mean classmates, but things like published papers, advertising, magazine articles, or really just about anything, depending on the area you want to develop.)

Chad Orzel:

What's the best way to teach students to think critically?

Depends what you mean by "think critically". It seems to me that clearly defining that phrase ought to be the first step.

If you're willing to risk introducing some Formal Systems, teach them to identify the premises used to reach the conclusion (including implicit premises such as validity of rules of inference, as well as explicit ones), distinguish whether these premises are being taken as Primary versus as Inference from other premises and evidence, and look for whether there is any step of inference where the conclusion does not follow from the given Premises. An overview of the assorted types of fallacies is probably worth including at some point in the education process; the taxonomy of formal and informal types at FallacyFiles.org might be handy for this.

Addressing correctness of premises themselves is a bit more complicated. However, it's worth pointing out that it's philosophically valid to consider the Refutation of any premise taken as primary Assertion. The catch of doing so is that you must be willing to re-examine ALL secondary premises which might have been taken as inference... which may include "How do I tell when a group of perceptions is probably connected to some 'thing' and not a collection of unrelated coincidences?" It's also worth noting that the only way to challenge a conclusion is to reject a premise, or claim that some step of inference is an invalid application of the particular rule of inference claimed.

Johan Larson:

Give them labs to do and lie about the expected results.

At college level, maybe. Not before that, however, and preferably late-in-course. Stunts on those lines can destroy the trust of the students for the teacher. Without trust, it is hard to have respect, and lack of respect for the instructor is counterproductive to learning. I base this opinion on my experience with one engineering professor I had, who did something very much on those lines (sabotage of lab equipment); the only things I learned for the rest of the course were a few points of pure math almost entirely incidental to the curriculum.

Give them labs to do and lie about the expected results.

Actively lying to them is a terrible idea, but the "inquiry-based learning" approach recommended by a lot of education experts isn't too far from this in practice. The key thing there is to get them to commit to incorrect predictions, and then do an experiment that proves them wrong. Essentially, you let them lie to themselves about the expected results, and then discover the right answer.

It's tough to do well, though. Especially in intro college physics-- the vast majority of students have taken physics in high school, and have some idea of what they're supposed to expect.

It was OK as long as we were within an order of magnitude. My teacher called this "engineering accuracy".

This attitude is captured in the old joke which translates phrases that commonly appear into scientific papers into ordinary everyday English. The phrase "correct within an order of magnitude" means "wrong".

Give them labs to do and lie about the expected results.

Let me add to the chorus of people saying this is a truly awful idea. It's one thing to not tell the students what to expect--that's probably a good idea. But if you tell the students to expect X, and they instead get Y, you have a problem. The students do not have the experience to know whether they got Y because they botched the experiment, the equipment was broken, or their teacher misled them. And as abb3w pointed out, when they discover that the last reason is the true one, it will destroy their trust in you.

A better approach, if you still think you need to predict a lab result that is not true, would be something along these lines: You predict X if Theory 1 is true, and Y if Theory 2 is true. X and Y differ by an amount large enough to measure with the crude equipment at hand. Which theory is correct? In this case, the conditionals mean you haven't actually lied ("P implies Q" is always true when P is false; it is only false if P is true and Q is false).

Show them that you are creating an environment where original thinking (including critical thinking) isnât penalized. For years, they have been rewarded for coming up with the ârightâ answers (meaning the expected answers). Give examples of the kind of âthinking outside the boxâ that will be well-received. Use analogies, examples from the course material, or both. You may still have some limits outside which they can not go; make that clear, too, otherwise their trust in their own critical faculties could be damaged when they step outside a set of unspoken limits which you expected them to intuit.

Try to break the model, is what usually do-- figure out what parameters can be varied with impunity, what variables really aren't, what variables may be coupled more tightly than the model is implying, and when do you just get obvious nonsense as a result of playing with the model.

I don't know how to teach someone to do that, other than to just keep doing it in front of them. Generally to the models that they're putting together.

Fermi problems.

If nothing else, you can get an order of magnitude answer for lab analyses or assigned problems. This gives the student a rough guide to the validity of his answer.

(The number of times I've seen a student blunder by 6 orders of magnitude and be blissfully unaware.... Grrr.)

Chad @ #13 finds the nugget of value in a dangerous suggestion.

I am opposed to ever lying to a student. You can decline to answer a question -- in my Biology or Anatomy class I had no trouble with questions about human sexuality, but refused to say anything about my own sex life, other than that I was married.

It is extremely important for students to trust you, if you are the teacher. Trust, once lost, is almost impossible to regain.

Getting people not to lie to themselves, that is a worthy goal. I believe that there is a different path for each person. No two students have exactly the same learning style.

You'll have students who have been lied to by parents, friends, other teachers, and you should distinguish yourself from such unreliable people.

I've had students doubt that people really walked on the moon, doubt that Darwin was right, doubt that Einstein was right, but the rare times I saw students doubt me were when I said something truthful that fell outside their perceptions of me, such as that I conversed at length with Johnny Depp, or that I'd played in a Rock & Roll band.

There is a sense that students feel lied to when they are told at a more advanced level that what they were told earlier was a simplification. "What do you mean, that electrons don't orbit around the nucleus like planets orbit around the Sun? Then I was I told about the Bohr model?"

There is the sense that all textbooks lie, by hiding the history of science, and making a series of revolutions seem like a straightforward step-by-step advanced of "Whig History."

So I suggest always telling the truth, as best you can.

Critical thinking? Our job is not to tell them WHAT to think, but to lead them to discover HOW to think.