Now I know what statistical heaping is. Very cool. goo.gl/oAJZBN
I occasionally write functions in R to help me with some aspect of teaching, usually teaching undergrad stats. I always intend to share them with others, but I don’t actually know many other people who use R for teaching. So why not share in blog posts? I’ll start now.
Disclaimer: This functions is untested by anyone but me, it might not work, and I provide no guarantee of accuracy. I mean, I try, but you know…
An R function for visualizing single-sample z-tests, a function I named zvis, produces graphs as below, using (usually) pretty simple options:
zvis(x=104, mu0=100, sigma=15, n=30)
I won’t paste the code here because it’s kind of long, but anyone interested can have it from this link –> zvis. The whole set of arguments can be seen after the break. If, by chance, anyone is interested in this function and wants more explanation of these options, or has suggestions for improving the function, please let me know. Continue reading R for Teaching: Visualize a z-test
I’m really fascinated lately by how the concept of “trigger warnings” has divided various North American communities–academics, political pundits, students, abuse victims, civil rights advocates, etc. Mostly I’m fascinated because the issue doesn’t divide cleanly along traditional battle lines like “left/right,” “progressive/conservative,” “male/female,” “individualist/collectivist,” etc.
There is now (2015, Fall) a compelling description of how this issue, arguably run amok, basically shut down one professor’s course. This piece is notable because it’s not the standard rant about hypersensitive students, written by people who have very little sociocultural vulnerability–it’s a narrative from a professor attempting to expose and celebrate minority sexuality in film, a prof who doubles as a rape crisis counselor, a professor (sadly, this seems relevant) who is female and an ethnic minority. Demands from a few (apparently highly fragile) students derailed what appears to have been a relevant, informative, iconoclastic, and edgy class. In other words, the things college should be about, in any “good liberal’s” fondest daydreams, were shut down by other liberals.
So it’s getting pretty interesting. It’s liberals against liberals up in here. Educators against educators. Students against students. I’ve got more pat phrases like that, but I’ll stop now.
Notably, not everyone is convinced that trigger warnings are destroying America. And my own experience, at a “public liberal arts” university in the Northeast/Midwest, has not included any “PC tyranny;” but, then again, I’m a middle-aged white male, so perhaps I’m not going to bear the brunt of this kind of thing even if it is happening around me. Some of my colleagues report increasing emotional fragility of students in regards to classroom content, but I have mainly noticed increased fragility in regards to reading between class periods and finding test questions based on lecture content that was not in the PowerPoint slides.
I was slightly surprised to find that AAUP has published an official statement about trigger warnings, calling them “infantilizing and anti-intellectual.” I don’t know that I’m comfortable going quite so far down that road, but I do recognize that the desire to protect the psychologically vulnerable can rub uncomfortably against other core values, particularly free speech. Continue reading Trigger warnings fail to divide along classic lines
In “How college sold its soul to the market”, the estimable William Deresiewicz eloquently describes my own thoughts about the spiral of higher education toward becoming a servant of financial economies instead of a place where we learn to be thoughtful, analytical, useful citizens in our societies.
This is one of the many key points for me: “This is not inevitable. It is the result of choices we have made, driven by an ideology that we have allowed to impose itself upon us.”
Because this essay is so awesome, I’m going to post a bunch more quotes I thought were particularly great after the break. Continue reading Excellent piece about the “commodification” of higher education
Lately, I’m all about leveraging the mostly-unused bandwidth, memory, and disk space I pay for from my webhost every month. So I’ve got ownCloud up and running with few problems, my moodle installation is working nicely, I’ve revived my LimeSurvey installation so I can do class surveys, polls, etc., and–brand new–I’ve got DokuWiki installed and running.
The last one is something I’ve been meaning to do for quite some time. My undergrad stats content has morphed and grown and (I hope) become somewhat refined over a decade or two and half a dozen textbooks. I have thought of typing up some “professor’s notes” for the students, but then a couple of years ago it occurred to me that a wiki would be a great way to maintain that. So that’s what I did. It has very little content right now, that I’m slowly changing that. It might or might not be something like a full set of course notes during this Fall semester. If not, I hope Spring will see a full, robust wiki for the students.
Oh, and speaking of that, OpenIntro has released version 3.0 of their nice, industry-disrupting (OK, not much, but they try) undergraduate statistics textbook. It’s not the most student-friendly text out there (especially for psychology students), but it’s free and it’s accurate.
But I really need to get the wiki content filled out so the students have a second source for the information.
Last fall, my university had a tiny kerfuffle: a student working for the university’s social media team retweeted another student’s tweet. It’s been deleted, but it said, “[university], where the weather is more confusing than the women.” Somebody cried foul, one thing led to another, and the President of the university issued a mea culpa about this event that “…quite simply shouldn’t have happened,” agreeing that the message was “sexist,” with “offensive implications,” and “hurtful.” As a result of this incident, the President has asked for “new policies on oversight of messages on the website and social media.”
I currently study men’s beliefs about women–particularly men’s beliefs that women are “fundamentally unknowable” (i.e., approximately as confusing as the weather), so this caught my eye when I (belatedly) found out about it.
TL;DR: I don’t think it was “sexist” in the way we usually use that term, and I think, from a logical point of view, the university administration overreacted. However, from a realpolitik perspective any university President in her right mind would have reacted the same way.
Unsung Hero of the Conference: The Elevator Algorithm! (Also receiving award for “most inscrutable entity”). #aps15nyc
Simine Vazire: “Be prepared to un-believe your favorite results. Even your own results.” #aps15nyc
Amazing session with statisticians (and Gelman) pointing out big current problems. Suggestions for teaching stats? *crickets* #aps15nyc
(1) Not-so-Bayesian. (2) Not-so-Bayesian. (3) Gelman. OK, but… shouldn’t there have been be a Bayesian? #aps15nyc
Mayo, then Senn, and now Gelman. This isn’t as polite a statistics fight as I initially thought. #aps15nyc
Damn an entire field for ignoring science, then defend sweeping assertions with bad anecdotes? Live by the data, die by the data #aps15nyc
#aps15nyc If you’re not listening to the “philosophy of statistics” presentations, you’re missing the very polite fight of the year.
Has anyone ever tried purely random assignment of presentations to time slots? Call me crazy, but this seems doable. #aps15nyc
I found this very frank expression of joint corporate humanitarian and environmental values in my shower. #aps15nyc http://t.co/qZaX7Sqqkj
I found this very frank message about corporate humanitarian and environmental values hanging in my shower. #aps15nyc http://t.co/fDIqZeJhmW
James Pennebaker offers to reduce tenure-track & adjunct jobs even more with SMOCs. His data are pretty hard to argue with. #aps15nyc
I think I speak for all those who spent the last week grading papers when I ask, “Thann what?!” #aps15nyc http://t.co/zow0hoHYUL