Sometimes, you just need a new, simple example for a homework question or a class warm-up.   I eyeballed and entered the   data here  ( r   = -.55). Enjoy. I use this little example to explain to use the regression formula to make a prediction. Here are my slides .

A few weeks ago, I posted about how you can share data in such a terrible way that one is not breaking the law, but the data is completely unusable. This makes me think of all the times I am irked when someone states a problem but doesn't offer a solution to the problem. Instead, they just talk about what is wrong and not how it could be. So, as a counter piece, let's cheer on organizations that ARE sharing data in a way that is readily accessible. You could use this in class as a palate cleanser if you teach your students about data obfuscation. You could also use it as a way of helping your students understand how data really is everywhere. Or even challenge them to brainstorm an app that uses readily accessible data in a new way to help folks.  Pro-Publica This website lets you check how often salmonella is found at different chicken processing plants. All you need to do is enter the p-number, company, or location listed on your package of chicken: https://projects.propubli...

Your students have heard of probiotics. In pill form, in yogurt, and if you are a psychology major, there is even rumbling that probitotics and gut health are linked to mental health. But this is still an emerging area of research. And NPR did a news story about a clinical trial that seeks to understand how probiotics may or may not help eliminate GI problems in children who are on antibiotics . Ask any parent, and they can tell you how antibiotics, which are wonderful, can mess with a kid's belly. When they are already sick. Science is trying to provide some insight into the health benefits of probiotics in this specific situation. They spell out the methodology: How to use in class: 1) I love about this example is that the research is happening now, and very officially as an FDA   clinical trial . So talk to your students about clinical trials, which I think you can then related back to why it is good to pre-register your non-FDA research, with explicit research m...

This is an excellent video for introducing t -tests AND finally getting the story straight regarding William Gossett, Guinness Brewery, and why Gossett published under the famous Student pseudonym. What did I learn? Apparently, Gossett DID have Guinness' blessings to publish. Also, this story demonstrates statisticians working in Quality Assurance as the original t-tests were designed to determine the consistency in the hops used in the brewing process. Those jobs are still available in industry today. Credit goes to the RStats Institute at Missouri State University.  This group has created many other tutorial videos for statistics as well.

New York Magazine's  The Science of Us made a brief, funny video that investigates the long running issue of the dangers of double dipping.  It is based on a Scientific America report of an actual published research article  about double dipping. Yes, it includes the Seinfeld clip about George double dipping. The video provides a brief example of how to go about testing a research hypothesis by operationalizing a hypothesis, collecting, and analyzing data. Here, the abstract question is about how dirty it is to double dip. And they operationalized this question: Research design: The researchers used a design that, conceptually, demonstrates ANOVA logic (the original article contains an ANOVA, the video itself makes no mention of ANOVA). The factor is "Dips" and there are three levels of the factor: Before they double dipped, they took a base-line bacterial reading of each dip. Good science, that. They display the findings in table form (aga...

Fivethirtyeight provides lots of beautiful pictures of spurious correlations found by their own in-house study. At the heart of this article are the limitations of a major tool use in nutritional research, the Food Frequency Questionnaire (FFQ). The author does a mini-study, enlisting the help of several co-workers and fivethirtyeight.com readers. They track track their own food for a week and reflect on how difficult it is to properly estimate and recall food (perhaps a mini-experiment you could do with your own students?). And she shares the spurious correlations she found in her own mini-research: Aschwanden also discusses how much noise and lack of consensus their is in real, published nutritional research (a good argument for why we need replication!):  http://fivethirtyeight.com/features/you-cant-trust-what-you-read-about-nutrition/ How to use in class: -Short comings of survey research, especially survey research that relies on accurate memories -...

UPDATE, 9/22/22: Here is a non-paywalled link to this information:  https://www.nytimes.com/2017/10/09/learning/whats-going-on-in-this-graph-oct-10-2017.html This article from the NYT is based on a survey . That survey asked a bunch of nutritionists if they considered certain foods healthy. Then they asked a bunch of everyday folks if they considered the same foods to be healthy. Then they generated the percentage of each group that considered the food healthy. And the NYT put the nutritionist responses on a Y-axis, and commoners on the X, and made a lovely scatterplot... Nutritionists and non-nutritionists agree that chocolate chip cookies are not healthy. However, nutritionists are far more critical of American cheese than are non-nutritionists.  ...and provided us with the raw data as well.

I love research that is counterintuitive. It is interesting to me and makes a strong, memorable example for the classroom. That's why I'm recommending Carroll's piece  from the NYT. It questions the conventional wisdom that breakfast is the most important meal of the day. As Carroll details, there is a long standing and strong belief in nutrition research claiming that breakfast reduces obesity and leads to numerous healthy outcomes. But most nutrition research is correlational, not causal. AND there seems to be an echo-chamber effect, such that folks are miss-citing previous nutrition research to bring it in line with the breakfast research. Reasons to use this article as a discussion piece in your statistics or research methods course: -Highlights the difference between correlation and causation -Provides an easy to understand example of publication bias ("no breakfast = obesity" is considered a fact, studies that found the opposite were less likely to...