People keep telling me, “You should teach!” I do seem to have a gift for explaining things in terms that can be understood. And, there are certain subjects — things people want to learn — that I know.
The readers of this blog include a lot of people interested in data. They want to know how to find patterns and determine whether they’re real, how to separate the trend from the noise, how to handle uncertainty. My readers are particularly interested in climate data, but probably want some skills they can apply to data in general.
So, I’m considering offering instruction online. There are two courses I have in mind, one for those who know only a little statistics or none at all, another for those who have some background but are interested in time series.
Stats 101 is to learn basic statistics in a way which will give you a genuine understanding of what it is and how it works. No prior statistics is assumed (but it won’t hurt) and you don’t need to know calculus. A bright high school student should be able to handle it easily. If you want to do more than just look at data, and especially if you want a solid foundation of understanding (not the “cookbook” approach too often used to teach statistics), this is for you.
Time Series will look at some of the specialized techniques of that discipline, including a deeper look at regression. It’s geared more to the physical sciences than is usual for an intro to time series. I’ll also share some of the notational conventions I’ve used which are a bit outside what’s usual in this field. A basic understanding of probability and statistics is needed, and yes we’ll use calculus.
If there’s sufficient interest out there, I’ll do it. I’ll start with two six-week courses, twice a week, of serious study — not “Oh my god another class” serious, but “Now I get it, I’m really ready to apply this stuff” serious. It’ll be night school so adults with day jobs (and students with day classes) can participate.
I’ll limit classes to keep things easier. Perhaps it’ll be one each of Stats 101 and Time Series, maybe it’ll be two classes of the same subject — it depends on how many want to learn these subjects.
Forsooth, ’tis better than thine
Maybe you’re one of those who would be interested in learing from someone with extensive knowledge, a gift for teaching, and lots of experience. You can do it online, in a setting that’s convenient but still allows the kind of personal attention and interaction that makes learning more fun, and much more effective. And it’ll be an online classroom, you won’t have to leave your house.
Maybe you know someone else who would be interested. Perhaps you’d like to make this happen for someone you value, who would value this knowledge — maybe a student who shows promise as a future scientist . I especially welcome the young, those who might be headed for science as a career — we can help them prepare for making sense out of the data that all scientists end up studying sooner or later (I think it would be a wonderful gift for some youngster you want to help). Just be sure they’re ready to be serious about it, because most of their fellow students are likely to be adults.
Teach thee I shall, but charge thee I must
My time ain’t free (except what I donate here on the blog). I’m thinking of charging $200 per student per course. With class size limited, you’d reserve your place by paying the tuition.
I’m soliciting comments. Would you be interested? Someone you know? Which course piques your interest? Are there other possibile subjects? What time of day/night would be best — early enough that east-coast folks don’t fall asleep but late enough that west-coast folks can still join?
Nothing is carved in stone yet. But if this comes to pass, and if you take one (or more) of my classes, I promise you this: I will not spend my time showing off how smart I am. I intend to show you how smart you are.
Open Mind 
Great idea.
Also, you may be underestimating the impact of your reputation – Have you studied the ad revenue if you moved off WordPress and started your own blog?
I’ll donate $50 toward a scholarship fund, if you want to consider applicants who honestly (however determined) don’t have the money. Young people who given time and knowledge could develop some leverage, for example.
—-
Possibly relevant:
http://slatestarcodex.com/2016/04/15/links-416-they-cant-link-our-dick/
—-excerpt—-
Maybe the most important article I’ve read this year: When Confounding Variables Are Out Of Control. A new PLoS paper argues …. many studies that claim to have gotten a result after “controlling for confounders” but which haven’t used complicated statistical techniques that nobody uses are now potentially suspect. I’ve always noticed that correlational studies that control for confounders get confirmed by experiments much less often than I would expect, and now I finally understand (some of) why.
Related: Andrew Gelman: “Let’s just put a bright line down right now. 2016 is year 1. Everything published before 2016 is provisional. Don’t take publication as meaning much of anything, and just cos a paper’s been cited approvingly, that’s not enough either. You have to read each paper on its own. Anything published in 2015 or earlier is part of the “too big to fail” era, it’s potentially a junk bond supported by toxic loans and you shouldn’t rely on it.”
—–end excerpt—–
[Response: A “scholarship fund” for deserving but poor, is a great idea.]
To market this you should explain why your paid course will be better than the many free online MOOCs from Coursera, Edx, etc. I don’t think that would be difficult to market though. eg. Udemy and Lynda offer paid courses or subscriptions that you can find for free at other places, but they have enough tailored courses that people choose to pay rather than wait for the next MOOC (or even find the next MOOC that offers something similar to their tailored course).
Would you also be recording your lectures for people who may miss a class (or classes)? [Response: Yes.] It will be difficult, I suspect, finding a time suitable for folks across a wide geographical area. Some may prefer a more flexible option to view the courses in their own time as work and life can often prevent people from signing in to see the original live presentation.
Definitely explore doing this though; you are a good teacher. I have Understanding Statistics and liked how you explained things (chapter 12 on Uncertainty cleared up a misconception I’d long had…Ch. 13 on Hypothesis Testing is another chapter I keep referencing when I’m explaining some basic bit of stats to folks via email–don’t want to mislead them esp. when I’m rusty, which I am now unfortunately-haven’t had many contracts lately that demanded much in the way of analysis).
[Response: For those interested, you can buy my book here.]
Interesting idea. I would have some interest in this. Not sure which (if any) of the proposed courses would be appropriate. Perhaps you could advise. Some background: B.S. from MIT in chemistry. Took calculus and differential equations as well as physics courses (and, of course, chemistry, biochemistry, etc.). But that was a long time ago (graduated in 1967) and I no longer have any proficiency in these mathematical techniques. Then PhD in organic chemistry (U. of Vermont), followed by post doc in organic chemistry (U. of Rochester) and finally a career as a medicinal chemist for a major pharma company doing drug discovery research (for 35 years). I used various statistical techniques to assist in interpreting biological data (t tests, ANOVA, correlation analyses, dose response curve analysis, etc.) and used various multivariate techniques (PLS, PCA) for other purposes (analyzing large data sets, compound design, etc.). But time series analysis is not an area where I have the needed expertise. I’ve done it using standard linear regressions, but without taking into consideration things like autocorrelation (learned about the need for that on your blog). I am now retired, but retain an interest in retrieving data and analyzing it. I have done this with some climate data (and also with marathon winning time data and other things), and would like to have more proficiency at it. Just not sure what the best next step for me would be. Since I have no actual need to do this (more of a hobby, I suppose, and has been useful in informing friends and relatives of what’s going on), the cost would be a factor, particularly in light of the free courses available (Coursera, edX, etc.) some of which I have taken on climate science, but none yet on data analysis (many are available).
It would be interesting. I have a medical interest (job). So I guess the Stats101 would be of more use, personally.
I too would be happy to give a little more $50-100 for you to offer a free course to someone else…
HAH
I’d be interested in both classes, but if I had to choose, I’d go with the time series (I am familiar with calculus and I’ve had some basic statistics)
I am not young nor do I expect to be able to pursue anything like a career in climate science. I’m just not that smart. I’m smart, just not that smart.
I would be studying for the sake of my own development and for citizenship.
The $200 price tag is too steep for me. I understand the need for you to be able to afford to offer this service. I’m not suggesting you should sacrifice too much for an abstract good. I’m just letting you know how it stands with me.
I have enjoyed your posts and learned mutch from them. I have no doubt that any course you offer with be well worth the trouble and expense.
I hope you get enough positive response to go ahead.
Meanwhile, I hope you will be able to continue the good work you have done in the past for the rest of us, even if not as frequently.
I needed this 60 years ago, but I think it is a little late for me now. :(
Still unemployed, or I’d take the time series course.
I would likt to participate in the Time series class.
I have a math and physics background, so it really sounds interesting.
But what does night school mean when living in Europe? :-)
I’m interested in Time Series, esp. if you’d focus on climate data.
the time series class sounds terrific. if you offer it, I’ll happily enroll.