Monday, September 26, 2016

Hyperbole and a Half's "Boyfriend doesn't have ebola. Probably. "

I've been using this example in class for a few years but never got around to blogging about it until now.

It seems that the first chapter of every statistics class provides a boring explanation of what a variable is, and examples of variables, and operationalizing variables, and quantifying the abstract for the purposes of conducting statistical analyses.

I try to make that boring topic funnier and applicable to real life via this post entitled "Boyfriend doesn't have ebola. Probably." from Allie Brosh, editor of Hyperbole and a Half.

In this posting, she rips apart the good old FACES scale after a trip with her boyfriend to the ER.


Monday, September 19, 2016

If your students get the joke, they get statistics.

Gleaned from multiple sources (FB, Pinterest, Twitter, none of these belong to me, etc.). Remember, if your students can explain why a stats funny is funny, they are demonstrating statistical knowledge. I like to ask students to explain the humor in such examples for extra credit points (see below for an example from my FA14 final exam).

Using xkcd.com for bonus points/assessing if students understand that correlation =/= causation



What are the numerical thresholds for probability? 





How does this refer to alpha? What type of error is being described, Type I or Type II?





What measure of central tendency is being described?



Dilbert: http://search.dilbert.com/comic/Kill%20Anyone
Sampling, CLT



http://foulmouthedbaker.com/2013/10/03/graphs-belong-on-cakes/
Because control vs. sample, standard deviations, normal curves. Also,"skewed" pun. If you go to the original website, the story behind this cakes has to do with a section of crappy students...so that is kind of funny and therapeutic for us teachers.

NOTE: The website the cake example comes from contains a lot of NSFW language. Which I, personally, have no problem with, but you might.




http://www.phdcomics.com/comics.php?f=1793
Because bar graphs, error bars, and understanding the joke behind this graph.



http://smbc-comics.com/index.php?id=3732
What kind of error, Type I or Type II?




http://wilwheaton.net/2015/11/wil-cant-draw-teachable-moment/
 Reliability, n-size


https://mathwithbaddrawings.com/2016/04/27/symbols-that-math-urgently-needs-to-adopt/
What does correlation give us? What does it not?


http://xkcd.com/1725/

What does the r^2 here indicate? Why would it be difficult to guess the direction of the relationship?



What is the joke here? For more rigor: What does et al. stand for? What are the APA rules for when to use et al.?


xkcd's Linear Regression

http://xkcd.com/1725/



This comic is another great example of allowing your student to demonstrate statistical comprehension by explaining why a comic is funny. What does the r^2 indicate? When would it be easy to guess the direction of the correlation? More on that via this previous blog post.

Monday, September 12, 2016

Kristoffer Magnusson's "Interpreting Confidence Intervals"

I have shared Kristoffer Magnusson's fantastic visualizations of statistical concepts here previously (correlation, Cohen's d). Here is another one that helps to explain confidence intervals, and how the likelihood of an interval containing true mu varies based on interval size as well as the size of the underlying sample.

The site is interactive in two ways. 1) The sliding bar at the top of the page allows you to adjust the size of the confidence interval, which you can read in the portion of the page labeled "CI coverage %" or directly above the CI ticker. See below.


2) You can also change the n-size for the samples the simulation is pulling.

The site also reports back the number of samples that include mu and the number of samples that miss mu (wee little example for Type I/Type II error).

How to use in class:

Students will see how intervals increase and decrease in size as you reset the CI percentage. As the sample size increases, the range of the intervals decrease and the sample means converge upon true mu. As confidence interval decreases, the range also decreases.

This is useful for teaching good ol' confidence intervals (as a compliment to the mean) but also to teach confidence intervals as they become an increasingly popular alternative to NHST.

Monday, September 5, 2016

Matt, Rali & Rhonda's Statistical Test Flowchart.

Take a look at this interactive, statistical decision making flow chart. I think that almost every statistics text includes a flow chart, but the interactive piece of this, and its ability to immediately provide the reader with information on the appropriate analysis AND software assistant is something your students can't get from paper versions of same.

The flow chart is based on Andy Field's work.

I discovered this tool via Reddit. I'm including that Reddit thread because the person that created the thread (commentor4) states that they also created the flow chart.

So, you are lead through a series of questions (read this from the bottom up).





After you provide the necessary information, the page provides you with a quick definition of the test you should conduct as well as links to instruction using popular statistical packages.