Skip to main content

CNN, exit polls, and chi-square examples.

CNN posted a whole mess of exit polling data that illustrates how different demographics voted last night. And through my "I teach too many stats classes" lense, I see many examples of chi-square.

I think they work at a conceptual level to clearly illustrate how chi-square looks at people falling into different categories, then measures whether the distribution of people is by chance or not.

If you actually wanted to test these using chi square, I would suggest you should problem delete the other/no answer column (or else they will all come out as statistical significant, I would wager).






EDIT (11/14/16); Daniel Findley made of video demonstrating how to use Excel to conduct chi-square tests on the marital status data. Check it out here.
Labels:

Comments

  1. Daniel FindleyNovember 12, 2016 at 4:11 PM

    Thanks for all your great suggestions. I put together a video based on your recommendations - because of the large sample size, I went into further analysis to show how the p-value changes as the sample changes.. If you have any thoughts/suggestions, I'd welcome any feedback you have.

    https://youtu.be/XMMw_TkTytQ

    ReplyDelete
    Replies
    1. Jessica HartnettNovember 14, 2016 at 7:24 AM

      Looks great. I'll share the link with my readers and add it to this post.

      Delete
    2. Jessica HartnettNovember 14, 2016 at 7:39 AM

      My only suggestion, and it is a little one, is label each table (is that expected data? Is it modified in some way?).

      Delete
    3. Daniel FindleyNovember 14, 2016 at 8:00 PM

      Good suggestion, I'll update the associated excel file that can be downloaded.

      Delete

Post a Comment

Popular posts from this blog

Ways to use funny meme scales in your stats classes

Have you ever heard of the theory that there are multiple people worldwide thinking about the same novel thing at the same time? It is the multiple discovery hypothesis of invention . Like, multiple great minds around the world were working on calculus at the same time. Well, I think a bunch of super-duper psychology professors were all thinking about scale memes and pedagogy at the same time. Clearly, this is just as impressive as calculus. Who were some of these great minds? 1) Dr.  Molly Metz maintains a curated list of hilarious "How you doing?" scales.  2) Dr. Esther Lindenström posted about using these scales as student check-ins. 3) I was working on a blog post about using such scales to teach the basics of variables.  So, I decided to create a post about three ways to use these scales in your stats classes:  1) Teaching the basics of variables. 2) Nominal vs. ordinal scales.  3) Daily check-in with your students.  1. Teach your students the basics...
Post a Comment
Read more

Using pulse rates to determine the scariest of scary movies

  The Science of Scare project, conducted by MoneySuperMarket.com, recorded heart rates in participants watching fifty horror movies to determine the scariest of scary movies. Below is a screenshot of the original variables and data for 12 of the 50 movies provided by MoneySuperMarket.com: https://www.moneysupermarket.com/broadband/features/science-of-scare/ https://www.moneysupermarket.com/broadband/features/science-of-scare/ Here is my version of the data in Excel format . It includes the original data plus four additional columns (so you can run more analyses on the data): -Year of Release -Rotten Tomato rating -Does this movie have a sequel (yes or no)? -Is this movie a sequel (yes or no)? Here are some ways you could use this in class: 1. Correlation : Rotten Tomato rating does not correlate with the overall scare score ( r = 0.13, p = 0.36).   2. Within-subject research design : Baseline, average, and maximum heart rates are reported for each film.   3. ...
Post a Comment
Read more

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 w...
1 comment
Read more