Slashdot Mirror
Teaching Natural Sciences To Social Science Students?
An anonymous reader writes "As a calculus professor for a small undergraduate institution, I normally lecture students who are majoring in the natural (or 'hard') sciences, such as mathematics, physics, and computer science. In fact, I have done so for almost thirteen years. However, for the first time this fall semester, we have a shortage of professors on our hands. As a result of this, I have been asked to teach a general education statistics class. Such classes are a major requirement for the large psychology student body we have here. I have never lectured social science students in any mathematics-related classes. My question to the Slashdot community is as follows: What are your experiences with teaching natural science classes to social science students? How is the experience the same or different in comparison to natural science students who may be more adept to the nuances of mathematics and other similar fields?"
Some will be apt and mentally up to speed with whatever you through at them.
Some will be unable to comprehend every third word.
Some will be uninterested. Others will be interested, but incapable.
Avoid using overly abstract concepts, and try to put things in terms they can understand. Since you are teaching statistics, try to use a lot of gambling references (lotto, roulette, etc.) since nearly all the students will have some familiarity with those.
I've found I can teach engineering concepts to elementary school teachers as long as I avoid formulae (and avoid using Latin references, so use the term "formulas" :-) ).
Betteridge's Law of Headlines. Did I do that right?
Isn't the use of statistics pretty much the only thing that distinguishes 'psychology' from 'talking about feelings'?
I realize that most psych majors don't actually go on to practice in psychology or psychiatry, and the ones that do generally have to do some flavor of graduate work; but I'm still rather alarmed by the implication of TFS that psych students might well be deeply uncomfortable with statistics...
As someone who's been on both of those academic sides (I started in hard, and moved into soft four years later), I never thought it was a lack of comprehension when fellow students have trouble with hard sciences. Instead, it's an appreciation for numerical conclusions.
Hard sciences basically tend to conclude three steps earlier than soft sciences -- because the math ends there. Hard sciences tend to describe a scenario, detail it numerically, hypothesize a numerical result, experiment numerically, solve for x, and x=n is the answer. The issue for soft science students is really that nobody ever cared about x. Hard sciences very quickly forget where x came from, because the entire scenario was translated into numbers. This affords hard sciences a certain level of abstraction, making problems faster to solve, easier to solve, and more widely relevant to re-apply.
Soft sciences tend to be industries where some aspect of the scenario can't be translated into numbers. It's usually a black-box scenario, and psychology is a good example. Such experiments don't attempt to describe certain behavioural anomalies numerically. Instead, 40% - 80% of a scenario is translated into numbers, leaving the remaining 20% - 60% as mysterious elements. Imagine a hard science equasion where six linear constants simply cannot be merged into a single constant -- for no seemingly good reason. As a direct result, after solving for x, the numerical abstraction must then be de-abstracted back into whatever the real-world scenario actually is. This procedure is not only an effort to grasp, but it's also a a major point of interpretation at the end of an experiment -- usually because x isn't the number of grams diluted; instead x is the likelihood that a person might turn left.
The nice part about de-abstracting at the end is that you wind up with a real-world answer, not a mystery number.
So my point is, that for a social science student used to walking in with a scenario, and walking out with a conclusion, you need to teach them how to appreciate the hard-science "datum result" without having a one-question-one-answer conclusion.
You can see this same effect in the business world. Big business corporate C.E.O.'s often make decisions from numbers in, to predict numbers out, without ever knowing where the numbers came from, nor how they'll be used on the way out. But if you've seen anyone go through "board of director" training, you know that the skills wind up applying to any business anywhere because they are all done at the hard-science executive level.
Constrast that to the entrepreneur of a small business, who needs to make all of the same decisions, but simply doesn't have the sample-size of data coming in to ever be able to make decisions numerically like the corporate guy -- which is one of the primary reasons that he has an advisory board instead of a board of directors. The decision-making process is very different, even though they are the same questions and the same answers. And each has a very difficult time in the other's business world.
Here's hoping someone else's response details a good way to actually teach that appreciation.
First thing to do is to get emacs and get the doctor watson mode working. Then have some sessions with Watson and understand how to talk to psychology students. To my best understanding, it involves rephrasing their questions and asking them why they ask that question or what their feeling is. All you need to do is to wing it for 50 minutes and charge them one hour of tuition fees. They will get the hang of it and learn to speak to their clients for 50 minutes and bill them for an hour.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
I'm sorry, am I misreading or are you saying statistics is a "soft science"? If you're that confused about things, then just go to the textbook, and teach one chapter a week.
I understood the summary to mean that the OP is teaching a statistics course to soft science students (those who are majoring in social science and phychology), and not that (s)he considers statistics to be a soft science.
Not to be rude, but reread the post.
The OP says he normally teaches hard sciences to students with a major in a hard science meaning that they are more likely prepared for the learning of hard sciences. Because of some staffing issues the OP now must teach his hard science classes to students with a major in soft sciences, thus previous classes may not have fully prepared them for a hard science class.
Because of this the OP is asking how to mold his teaching strategy to better target those soft science majors.
You make it sound like you are teaching physics to special ed classes.
They are as smart as everyone you've had so far. You may see some differences in their backgrounds, but that's easy enough if you make allowances to give more basics or point them to appropriate resources. I'd give an example, but I have no idea what "natural science" is to you. Geology and oceanography are natural sciences, same as physics, but they share little in common.
One thing you may notice is that arts students in hard classes may want more "why" than "how" answers. So be prepared for more philosophical discussions, or correct, if silly, comments (i.e., the "why" for valence electrons is that the stable ones are like a comforable couch, and the unstable ones are hard benches. You want the better seat, but you don't really want to get up, and the worse the chair, the sooner you'd move) or something like that. The "why" as an expression of potential energy in MeV won't get the point across as well as a discussion of musical couches, and they'll remember it better, isn't that the goal, over the goal of the hard science students where accuracy is above all.
Learn to love Alaska
A major problem with these sorts of courses is that they're often not taught in a way that emphasizes their utility to the student. If you're thinking about being a psychologist for example why is calculus important? I'm not saying it isn't. I can think of several different ways it could be very important especially as it regards understanding statistics.
But you might want to create some test questions that relate to their majors.
In business calculus they focus on it's relationship to various economic calculations. So you might want to look at drug trial statistics or anthropological/demographic statistics.
And for the love of God... please tell them that correlation is not causation. You'd be doing everyone a huge favor. These guys are going going write stupid papers or write blogs or something similar that will pop up in the media. And everyone here at slashdot will be facepalming over another dumb paper that didn't acknowledge that simple fact.
Just saying.
I've decided to stop wasting my time responding to AC trolls/sockpuppets... so if you want a response from me... login.
Always interesting to see the categories different parts of academia place each other in. The post's author is calling math, physics and comp-sci "natural sciences" and apparently considers statistics to be "social science". I'm a geology professor and, as far as I'm aware, my colleagues and I tend to consider Earth, environmental, and biological sciences to be the "natural sciences"; physics, chemistry, engineering, and any math to be "physical science"; and psychology, sociology, (cultural) anthropology, etc. to be "social sciences". Everything else is art and/or humanities.
I wonder how other groups categorize one another? Right off the bat I'd suspect that mathematicians don't always consider themselves scientists. Perhaps ditto for engineers. People tend to form and place each other in groups of varying degrees of subjectivity. How you place others probably says something about the standards and values of one's own group.
This sounds like it'd make a great piece of social-psych research! They love this kind of fluff, right? (j/k)
I think that professor was me. I am from Trinidad but my parents were from Nigeria where I am living now. I would love to get back in touch and have a beer with you. If you could please send me a small advance for travel along with your SSN and a passport photo I can begin to make arrangements.
Citing a specific example of hard science that is encompassed within a science does not prove the encompassing science itself is hard science. (notice I didn't say anything about statistics...just as the OP didn't say statistics was a soft science).
My thoughts exactly.
He might end up losing some of his elitist attitude before the course is over. It would be better if lost the attitude ahead of time, and approached the experience like he was at least teaching the same species.
Indeed. I teach statistics to mathematicians, biologists, psychologists and social scientists and I would say the social scientists 'get' the principles of statistics better than the 'hard' scientists do. The main reason is that soft scientists (which is a horrible term) can think about uncertainty and its consequences, whereas hard scientists (mathematicians included) are unhappy if they don't have a yes/no answer to a question. Obviously this is a generalisation but it may inform your approach to teaching.
Also, statistics is not 'just math'. I know this because I can do statistics but I can't do math(s) any more. :-)
If you're not familiar with it, I recommend you read Ken Bain's What the Best College Teachers Do (2004) which provides a wide range of insights and approaches that can help you out in any classroom. Speaking as a former science major who went on to a Ph.D. in history, the number one difference I notice between the streams is that many of the social science and humanities students believe they're bad at math and statistics. Problems in high school convinced them that they can't cut it - a high proportion will claim they're incapable in the fields. The secret to your success is convincing them that they can and want to master these skills.
I know - I teach a stats module as part of my sophomore course for majors. They learn how to read, interpret and critique statistics in articles in their field of study. Did you know that most of them don't know how to read and interpret statistics? The number of students at the start of the course who tell me they don't stop to read the charts because "they'll never understand them" is staggering. Statistical literacy should be the bedrock skill you inculcate. Show them good and bad uses of statistics. Teach them to figure out when someone's playing fast and loose with figures, hoping to fool readers. That will build their confidence and their thirst for knowledge.
My students go on to create their own time series and other statistical outputs from a dataset that they all find fascinating. (I use the Old Bailey Online for this, a website with material in statistically manipulable format for almost 200,000 trials at London's major criminal court: almost everyone finds the history of crime at least a little bit intriguing and so they will persevere a bit more when they run up against problems or road blocks.) Don't waste a lot of the time throwing new theories at them - make sure that every new concept you introduce is tied to something they'll want to and be able to explore.
Sure, some won't want to try. They'll find the work too hard or uninteresting no matter what you do. But others will be able to master this if you make it clear both why they need to learn certain techniques and how while giving them some clear and jargon-free walk-throughs. Exercises they can tackle tied into the fields they already find interesting are a great way to keep them motivated.
Look at some of the textbooks that are out there for stats that are directed to your U's social science fields - see what elements they emphasize as important for the field of psych, poli sci, etc., and then decide how you want to incorporate those key elements into your own teaching. Avoid getting too tied into teaching a particular software package - make sure they understand how to generalize their application.
Good luck - you're tackling what many consider a thankless course but one which can help to change students from math-phobic and fearful to at least statistically literate and confident that they can understand and apply some basic skills in the field as they go on in life.
ancarett, historian and zombie gamer
You do it exactly the same. Psychologists take stats pretty seriously.
...especially as regards the use of mathematics in the interpretation of 'data' where the soft sciences have such a 'hand wavy' approach to cause and effect.
To me, economics is a prime example. Forgive me if I'm off base in in my belief that economics is both sociological and soft(headed), but tyring to measure human behavior in the absence of an accounting for political corruption within this purely human realm and leaving the so-called black market beyond it's consideration leaves the inclusion of economics within the realm of 'science' suspect.
I would haved greatly appreciated any attempt by a professor to explain the difference between soft science and hard science, especially if it included an math based explanation of the nuance between these different domains.
Soft sciences are typically about trying to solve 'wicked' problems, which are those that are generally impossible to completely solve (end poverty or health inequality, understand crime, migration, or human behaviour in general etc). Hard scientists typically try to solve problems that are relatively much easier because they have a simple concrete goal (put a man on the moon, make a bomb, cure some disease)
Soft scientists need a much stronger theoretical framework to interpret their data, because of the absence of any really testable mechanisms for the effects they observe. This can come across as 'hand wavy' but it really isn't. Your economics example isn't entirely fair, some economic models will include corruption and black markets etc and others wont, just as some physics models include relativistic effects and others don't. A good scientist has to choose the right model to approah any problem, regardless of discipline.
I've been working in an inter-disciplinary group and have had the opportunity to see medics and economists try to work together. The two cultures are very different in their scientific approach, both consider the other to be unnecessarily picky about some aspects of the work while not being rigorous enough in others. Eg economists spend a huge amount of their time trying to prove causation in observational data, while medics will typically wave this away if they think the causal effect is likely enough. On the other hand economists tend not to contextualise their results well enough, while medics will see the bigger picture in terms of building on existing science.
This might be the angle in for the original questioner's method.
Maybe he can reduce the raw theorems by 25%, and instead push harder on media and logical thinking issues.
Instead of too much push on the formal notation, what if he goes into a lot of "biased science" examples from the real media? Showing how slanted presentations produce emotional reactions, etc.
In a sense, "If I were in a position to hire", I'd rather have a smart thinker who's drilled cold on picking up sample bias than a book theoretician who can drill out 18 line proofs but folds the minute he/she gets into something about affordable housing studies and doesn't account for geo-social trends.
My first Journal Entry ever, in 8 years! http://slashdot.org/journal/365947/aphelion-scifi-fantasy-horror-poetry-webzine
True Story: -Engineers at my school had to take 15 units of Arts courses as part of their studies, and Economics 100 was one of the popular choices. We had an Econ 100 class that matched a hole in the 3rd year mech and EE schedule, as a result we had about 2/3 engineers and 1/3 Arts students.
One day the prof, a very smart man with a subtle sense of humor, drew a graph of some function on the board. He drew the x and y axes, a straight line with a 45 degree slope and labelled the x intercept "a" and the y intercept "b". One of the girls from Arts puts up her hand and says "I don't think it should be that steep". The prof erases the line, redraws it half as steep and labels the x intercept "a" and the y intercept "b". "How's that" he says."Much better" says Arts girl.
Every engineer in the place falls over laughing. We laugh even harder as we see the confused look on Arts girl's face as she tries to figure out what's so damned funny. The prof never cracks a smile.
I will try to inform you a little about economics (speaking as the holder of both a BSc and PhD in Economics):
The key difference is that economics and social sciences are mostly non-experimental (people don't take kindly to you arbitrarily changing their parents, education, or wealth - which is the 'experimental' way of establishing cause and effect). This means that the statistical issues are orders of magnitude larger than those that exist in experimental sciences. In an experimental science you can go off and get new data where you have controlled for most everything except the effect you are interested in and a simple regression will generally be all you need. In a non-experimental science you are stuck with the data that nature has given you. As a result you need to be very careful to get meaningful results. But, in case you are doubting, you can get meaningful results if you are careful enough.
Thus, my second point: Economics is not soft headed. In fact, it is very hard headed because you need to be when you are dealing with data that are generally speaking - crap. There are so many ways you can be mislead by non-experimental data and you need to be very hard-headed to avoid this. I won't claim mistakes haven't been made, but those mistakes are the reason economics has gotten much better at dealing with this than many people might realise. But, there is only so much you can do when the data are the way they are.
So don't assume the difficulty of getting solid results in economics reflects the ability of the practitioners rather than the raw materials you are dealing with.
If I had a dollar for every paper published in a peer-reviewed social sciences journal which totally abused statistics, I'd retire and use my extra cash to fund organizations directed at basic logic and math education, trying to help with the situation.
Most social studies students I knew had little understanding of the statistics they were using. It was basically a magic incantation for giving them results and making their conclusions sound more credible to other people who likewise didn't understand statistics. The result is bad statistics and bad science. Yes, these people aren't idiots, but they've become used to being rewarded without having to think rigorously.
The impression I get is that the pattern persists even among those few who make it into the field. There are some psychologists etc who are really trying to do real science- a difficult task since the basic concepts are even more up in the air than the basic concepts of chemistry were in the days of the alchemists. As far as I can tell, however, quite a lot are quite happy to be able to find ways of running a study so it will inevitably vindicate their preexisting biases and will fudge the statistics to match.
For the OP: You're right to be concerned. Students for the GE stats class are usually woefully underprepared. Rather than giving them the rigorous preparation in logic, multivariate calculus, etc they really need to understand statistics, the GE stats class does the equivalent of the Wizard's favor to the Scarecrow.
"I can't give you a brain, so I'll give you a passing grade! Now you understand statistics! Go back to your department now, please. (Phew, they're gone at last. That kind of work may pay the bills here in the Stats dept. but it doesn't do wonders for my sense of academic integrity as an educator.)"
If you don't understand statistics you simply cannot work in the Social Sciences. Ever. You are not allowed to do the experiments necessary to isolate variables properly, and even if some sociopath (for example) traumatized three groups of ten people exactly the same way, and tried three different forms of psychological treatment on them to see what happens you'd run into the fact that all 30 of victims are individuals who will respond to each treatment differently.
Which means RL Psychologists are stuck doing a sophisticated study of people who just happened to get traumatized, and then chose a course of treatment; and the only way to get useful data from that is do lots of statistics. But not too much statistics or you risk over-fitting.
OTOH you can be a perfectly good chemist without understanding the difference between correlation and causation.
Stephen Greenfield, the best professor I ever had, happened to be one who mainly taught undergraduate math to math, physics, and engineering students and graduate math. However, he had a passion for teaching unlike I'd ever seen and he worked on a course at Rutgers to teach math to non-sciency types.
The last paragraph on this page has a description of the course.
The course diary has tons of material in it.
If you browse Stephen Greenfield's homepage, you'll find a wealth of teaching that might be able to be applied. He's since retired, but his page is still up, so make use of it!
"Nature doesn't care how smart you are. You can still be wrong." - Richard Feynman
I taught Introductory Astronomy to a bunch of non-science majors looking to fulfill their science requirement. It was fun, that the kids were good at it despite lacking the physics and mathematics background for it. Statistics maybe isn't as interesting as astronomy, so keeping them interested is probably your biggest challenge. That would be true no matter who was taking the class.
Teaching to science and engineering students too often results in off topic discussions which put me off my lecture schedule (that's my problem, but it makes those classes more stressful to teach). Enthusiasm and detail are good, but lectures have a time limit. Pre-meds (a totally different category from science and engineering students) rarely show more than a passing interest in physics. Social scientists were really a joy to teach. They were interested in the material, the historical context and particularly the differences between astronomy, 'movie astronomy' and astrology. There's more than enough historical and current relevance to statistics to pique their interest, but you'll have to point it out.
I think the main difference is likely to be that sociological students are more used to questioning fundamental assumptions. I suppose this it because hard logic is a lot less useful when a large proportion of your reasoning is based
intuition. So be prepared to explain just about anything you consider "obvious", and to having your pedagogical skills tested to the limit.
I myself came to mathematics at university as an outsider; I found that my peers would simply accept most of what the teachers said, but I had a hard time adjusting to many of the viewpoints. Another thing I found difficult was spotting what it was I was supposed to learn - in the first years I would work hard on applying the major theorems to all exercises, and it was not until after my bachelor that one of my toturs exclaimed, with some exasperation: "Why don't you use the techniques that you have been shown in the proofs, like you are supposed to!?" - So the second thing you will need to do is, point out explicitly what you expect your students to learn.