Squishy science versus Hard science

First, as a “hard” scientist I am somewhat worried about whether the “science” in the “science of how people learn” is “squishy” science vs “hard” science.  As a novice in this field I guess I am uncertain how much of what we are learning is just the latest passing fad backed up by some weak observational investigation versus how much is based on truly reproducible data in which confounding variables have been minimized and alternate interpretations have been ruled out.

I believe this skepticism is based on two factors:

1) My mom has been a secondary school teacher for 40 years and has seen teaching trends come, go, and return. This seems more like fashion than the kind of building progress that (forgive me) “hard” science relies on.

2) My dad is a college professor with little respect for “post-modern” jargon. Thus I am always suspicious of terms that seem purposefully puffed up. To me “Affective Domain” could be summarized much more succinctly as internal feelings or emotions (encompassing the range from passing mood to deeply held beliefs). “Affective field” could be summarized as environment: physical environment and emotional environment created by our interactions. To me words like “environment” and “emotions” convey actual meaning, even to the novice, while “Affective Domain” and “Affective Field” say nothing to the novice. You may now note the irony of a someone whose entire field is built on abbreviations complaining about jargon. For whatever reason, these jargony terms make me nervous that I am in “squishy” science world versus “hard” science world.

For me the take home question is: How much of what we are learning is an unshakable foundation upon which I can build with confidence and how much will shift with the sands of time?

Having the arrogance to write what I have just written, pretty much calling out social science as “squishy,” let me end with a slightly tongue-in-cheek characterization of the personalities of “hard” science by a Nobel prize winning physicist: “Biologists are stupid, physicists are children, and chemists are gangsters.” Please don’t take this post the wrong way.

9 responses to “Squishy science versus Hard science

  1. I am glad you were able to translate the language used. Coming from engineering, I tend to run into phrases that don’t seem to make sense, so I pass over them, only to find that I have passed over much of what qualifies the argument. I am glad that it appears to be simply a matter of silly vocabulary and unnecessary inflation, as opposed to genuine complexity that has been confusing me.

    Perhaps you’ll continue to translate some of this writing into “hard” language for us engineers.

    Thanks –

    Landon

  2. One of the questions I continue to raise is to what extent an understanding of the science of how people learn should be used to inform teaching. Some cognitive scientists aren’t very sure that much of it can be used in a way that directly translates into teaching practice. Dan Willingham, from UVa has an interesting pitch here http://www.youtube.com/watch?v=vdJ7JW0LgVs

    Your post here seems to conflate research on learning with teaching trends. Hoping you might clarify that with some examples.

    Gamson nad Chickering reviewed 50 years of educational research to identify 7 principles of good practice for undergraduate education (http://honolulu.hawaii.edu/intranet/committees/FacDevCom/guidebk/teachtip/7princip.htm).

    As a scientist, its obvious you have an interest in evidence-based decision making…where do you look for evidence to inform your practice of teaching?

  3. It’s all squishy!

    Even the hard sciences have a difficult time dealing with extraneous or intervening variables. Global warming, nutrition, the role of the mind in the recovery of illness (placebo effect), etc. Sure we can isolate the relationship between a couple of variables and produce a compelling correlation coefficient but in the real world there are always these confounding variables at work. The fact that some people value hard science over the “softer” sciences is evidence of the significance of human affect. Sure hard science can reduce bias through empiricism but you still have humans doing the science, interpreting the science, applying the science and politicizing the science. This is why I think the idea of being a “critically reflective practitioner is so important. Your skepticism is well-founded regarding squishy science but you should be equally skeptical about the hard sciences. For me, that is what is most important. Higher education needs to be more effective at producing “knowledge workers” and not just scientists of either ilk.

    I think the most compelling knowledge is that which result from triangulating methods (hard and soft) and through meta-analyses which looks at entire bodies of research over time to look for gaps and gems.

    The science behind human affect is still in its infancy in education but look at the advertising industry (as well as political science). Convincing the human race that smoking is cool seemed to work throughout the 20th century and the politics of fear seems to be working — whether it is global warming or the axis of evil.

    On the other hand, the science of how people learn is based on both hard and soft science and it is a meta-analysis so these tow criteria tend to persuade me more than usual.

  4. Actually, I was making a meta comment. The meta comment is that “science” is a label often slapped onto data to make it seem more substantive. To me an investigation can only be scientific if the hypothesis can be falsified. To me something can only be scientific if the field can form a consensus (after exhaustive research) about which hypothesis are correct, which were incorrect and then build on that foundation.

    Much of the data created in the world of human interaction probably fails to live up to these standards. That doesn’t mean that data is invalid or has no meaning, it just means it’s not “hard” science. No doubt some of the hypothesis on learning are falsifiable but much of it seems like descriptive conjecture to me (eg Affective Domain and Affective Field).

    I am looking to this class, the materials, my fellow students and instructors, to provide evidence-based teaching methods. At the same time, I am not opposed to using teaching methods that haven’t been thoroughly tested by the scientific method. Knowing that something works is enough for me, I don’t need to know without a shadow of a doubt exactly why it works to try a method.

  5. @zach,
    Science is emphatically NOT all squishy science. The hard science process does not expect every paper and every conclusion proffered to be correct or without human error. However over time, numerous investigations coming at a problem from numerous directions produce irrefutable facts. The genetic code, evolution, that MHC presents peptide antigens, these are irrefutable established facts on which scientists can build without a shadow of a doubt.

    Because science journalism is always reporting on the latest and greatest provocative finding, which is later refuted by other studies, the general public has very little insight into the scientific process of honing a hypothesis over time by millions of data points and thousands of papers by thousands of scientists into an irrefutable fact. The general public has very little idea the volume of data required to prove or disprove a hypothesis and rarely hears when a hypothesis is proven or disproved, they generally hear about a hypothesis when it is first proposed and thus still “squishy.”

  6. I couldn’t agree more. This is certainly a squishy science. I have two insights I’d like to share:

    1) Last class we gathered into groups and each group discussed one of the five themes that “changed conceptions of learning.” While going around the room after a group would speak, I couldn’t help but think to myself that at times we were repeating the same ideas from a previous group’s discussion. The difference being that our vocabulary was different with unnecessary inflation. For example, #1 “Memory and Structure of Knowledge” and #3 “Early Foundations”, I found to be entirely too similar to be categorized as two different themes that changed conceptions of learning. How can one’s memory and structure of knowledge not directly be influenced by their early foundations? In our discussion of “early foundations” we spoke of the need to know what a student knew when coming into our class. In the literature for “Memory and Structure of Knowledge” it states “knowing how learners develop…”, which alludes to the same principle of understanding your pupil when they walk into your class as mentioned from “early foundations”. All 5 themes we discussed I believe are very similar/interrelated, but each unique in their jargon.

    2) “teaching trends come, go, and return”… I believe teaching trends follow society and society is very trendy (society follows history, and they say history repeats itself). Our society constantly deems old methods of teaching no longer suitable in our society for one reason or another. Teachers used to be able to beat children, not anymore (probably for good reason, but the method worked, no one can deny that). Kids used to have to participate in gym class, not anymore because they are self conscious about being overweight (yet we put junk food vending machines and terrible lunch menus in schools). I was speaking with a 24 yr veteran high school teacher of Henrico County, She got in trouble from the administration for making kids write identical sentences 50 times (she’d done it for years). Look at the impact Columbine/Va Tech has had on our society and how it has affected the learning environment in schools. As society changes, so do our methods of teaching, whether its backed by an unshakable foundation or not…

  7. Agreed…I think we are saying the same thing…just from two different perspectives. Science in and of itself, may not be as squishy, but as soon as you add the human element, either as the researcher (in the hard sciences) or as the subject (in the softer sciences) then you have introduced the “squish. ” So for me, that makes it that much more important for us to be practical skeptics and critically reflective–as scientists, consumers of science, and as educators.

  8. Everything shifts with the sands of time–even the physical sciences. Isn’t the whole purpose of research and science to discover new ideas and to challenge the existing ones? As such, science is not the “rock” that you make it out to be.

    Also, I prefer the terms “physical” and “social” sciences. The sciences are different, yes, but that does not make one more or less valid than the other. In the social sciences, we have a mantra: correlation does not equal causation. We can show that there is a relationship between the two, but we cannot show for certain how that relationship works. We do have hypotheses that we reject/fail to reject, but that is our “null hypothesis” in which we set up that there is no relationship between the two variables.

    The real world is more complicated than a lab setting, and you cannot possibly know all potential influences on an activity that are occurring outside of your research. Some have said that makes social science more difficult than the physical sciences. 😉

  9. Meredith, I agree social science is harder than physical sciences.

    I maintain that the physical sciences are distinct from the social sciences in that in the physical sciences a rock of core foundation truths is built up over time. This makes the physical sciences easier. I do not have to question that the three base pairs make a codon that codes for an amino acid. This is an established fact, it will never change or shift with time, thus I can rely on it and build on it. I am not saying every new data point or hypothesis is a core foundational fact. However, I question whether social science can establish these foundational core facts that can be relied upon in the same way. I think no but maybe you can give me an example.

    Physical sciences are also easier because we have more range to experiment and manipulate. Thus in the physical sciences we can prove causation not just correlation.

    This is in no way an attack on social science, I just wasn’t sure how much of the science of learning belonged in social science and how much belonged in physical science. Even for the part of the science of learning that belongs in the physical sciences, I don’t know if this field is too new to have established core facts yet.

Leave a comment