Skip to main content

Posts

Geckoboard's "Data fallacies to avoid"

Geckoboard created a list of common statistical fallacies , including cherry picking, Simpson's paradox, gerrymandering, and many more. Each fallacy comes with a brief description of the fallacy, references, a printable card for review/display, and drawing. They are kind of gorgeous and to the point and helpful. https://www.geckoboard.com/learn/data-literacy/statistical-fallacies/sampling-bias/ Here is the downloadable card for the Regression Toward the Mean: https://www.geckoboard.com/assets/regression-toward-the-mean.pdf They even present all of their graphics as  a free, downloadable poster . My only peeve is that they use the term "Data Dredging" where I would have said "HARKing" or "Going on a fishing expedition". And that is just the tiniest of peeves, I think this is a good check list filled with images and concise descriptions that would look beautiful in a college professor's office, a stats class room, or anonymously ...

Mathisfun.com's Least Squared Error calculator

Mathisfun.com bills this as a Least Squared Error calculator , but I don't think it is a calculator. I think it is a nice visual aid that demonstrates how the regression line/equation change as your data changes. The static photo below doesn't do this interactive website justice. You can drag and drop any of the dots on the scatter plot and watch as the regression line and regression line equation are recalculated to best predict Y based on X. It doesn't explicitly show the math going on behind the scenes, but it is a nice compliment to your LSE lecture. https://www.mathsisfun.com/data/least-squares-calculator.html

Chi-square via The Onion's "Saying ‘Smells Okay’ Precedes 85% Of Foodborne Illnesses Annually"

Once again, The Onion publishes satire research (which should be, like, a submission category for JPSP) claiming to study phrases uttered before food poisoning happens . https://www.theonion.com/report-saying-smells-okay-precedes-85-of-foodborne-1819579726 I've turned this fake research into fake data to conduct an actual chi-square test of goodness of fit. Here is data that will give you a significant chi square, with 85% of participants falling into the "smells okay" category. Did sick person say aid "Smells Okay" before eating leftovers? No Yes 19 106

Press Roger's statistics infographics

Press Rogers is a very generous person who has created a number of documents that illustrate the thinking/math behind a variety of statistics. Here is his illustration for ANOVA, which includes one way, two way, repeated measured, and mixed. https://pressrogers.com/documents/anova-overview/ PS: I found out about this resource via the Society for the Teaching of Psychology FB group , which is active and filled with great advice about teaching statistics as well as other psychology topics. 

Crash Course: Statistics

Crash course website produces brief, informative videos. They are a mix of animation and live action, and cover an array of topics, including statistics. This one is all about measures of central tendency: Here is the listing under their #statistics tag , which includes videos about correlation/causation, data visualization, and variability. And, you know what? This is just a super cool web site, full stop. Here are all of their psychology videos .

Stein's "Troubling History In Medical Research Still Fresh For Black Americans"

NPR, as part of their series about discrimination in America , talked about how it is difficult to obtain a diverse research sample when your diverse research sample doesn't trust scientists. This story by Rob Stein is about public outreach attempts in order to gather a representative sample for a large scale genetic research study. The story is also about how historical occurrences of research violations live on in the memory of the affected communities. The National Institutes for Health is trying to collect a robust, diverse sampling of Americans as part of the All of Us initiative. NIH wants to build a giant, representative database of Americans and information about their health and genetics. As of the air date for this story, African Americans were underepresented in the sample, and the reason behind this is historical. Due to terrible violation of African American research participant rights (Tuskeegee, Henrietta Lacks), many African Americans are unwilling to partic...

Dozen of interactive stats demos from @artofstat

This website is associated with Agresti, Franklin, and Klinenberg's text Statistics, The Art and Science of Learning from Data ( @artofstat ), and there are dozens of great interactives to share with your statistics students. Similar and useful interactives exist elsewhere, but it is nice to have such a thorough, one-stop-shop of great visuals. Below, I have included screengrabs of two of their interactive tools. They also explain chi-square distributions, central limit theorem, exploratory data analysis, multivariate relationships, etc. This interactive about linear regression let's you put in your own dots in the scatter plot, and returns descriptive data and the regression line, https://istats.shinyapps.io/ExploreLinReg/.  Show the difference between two populations (of your own creation), https://istats.shinyapps.io/2sample_mean/