I love counterintuitive data that challenges commonly held beliefs. And there is a lot of counterintuitive health data out there (For example, data questioning the health benefits associated with taking vitamins  or data that lead to a revolution in how we put our babies to sleep AND cut incidents of SIDS in half ). This story by Aschwanden for fivethirtyeight.com discusses efficacy data for various kinds of cancer screening. Short version of this article: Early cancer screening detects non-cancerous lumps and abnormalities in the human body, which in turn leads to additional and evasive tests and procedures in order to ensure that an individual really is cancer-free or to remove growths that are not life-threatening (but expose an individual to all the risks associated with surgery). Specific Examples: 1) Diagnosis of thyroid cancer in South Korea has increased. Because it is being tested more often. However, death due to thyroid cancer has NOT increased (see figure below)...

Elahe Izadi, writing for the Washington Post, did a report on this article by Eichstaedt et. al, (2015) . The original research analyzed tweet content for hostility and noted the location of the tweet. Data analysis found a positive correlation between regions with lots of angry tweets and the likelihood of dying from a heart attack. The authors of the study note that the median age of Twitter users is below that of the general population in the United States. Additionally, they did not use a within-subject research design. Instead, they argue that patterns in hostility in tweets reflect on underlying hostility of a given region. An excellent example of data mining, health psychology, aggression, research design, etc. Also, another example of using Twitter, specifically, in order to engage in public health research ( see this previous post detailing efforts to use Twitter to close down unsafe restaurants ).

This article and figure from Harry Enten (reporting for fivethrityegiht) provides informative and horrifying data on the median last day of measurable snow in different cities in America. (Personally, I find it horrifying because my median last day of measurable snow isn't until early April). This article provides easy-to-understand examples of percentiles, interquartile range, use of archival data, and median. Portland and Dallas can go suck an egg.

Weber and Silverman's article for the Wall Street Journal has lots of good psychy/stats information  ( here is a .pdf of the article if you hit a pay wall ). I think it would also be applicable to health and I/O psychology classes. The graph below summarizes the main point of the article: Certain occupations have a greater likelihood of obesity than others (a good example of means, descriptive statistics, graphs to demonstrate variation from the mean). As such, how can employers go about increasing employee wellness? How does this benefit an organization financially? Can data help an employer decide upon where to focus wellness efforts? The article goes on to highlight various programs implemented by employers in order to increase employee health (including efficacy studies to test the effectiveness of the programs). In addition to the efficacy research example, the article describes how some employers are using various apps in order to collect data about employee health and...

io9.com Das and Biller, reporting for io9.com , shared several good examples of bad graphs. The graphs are bad for a variety of reasons. I have highlighted a few below. Non-traditional display of data that create the illusion that the opposite of the truth is true: Note the y-axis is flipped (0 at the top...huh?), so murders have actually INCREASED since "Stand Your Ground".  Cherry picking data: Confusing data presentation: I think that this could be fun to use in class as a discussion piece to pick apart bad graphs, so that your students 1) think critically about all graphs and figures they see and 2) learn how to make truthful graphs. Another fun way to use this in class would be to present these graphs to your students and then ask them to create APA style manual compliant graphs of the same data.

Taylor, writing for Mashable , describes what happens when carefully conducted public health research (published in the  Journal of the American Medical Association ) becomes attention grabbing and poorly represented click bait. Data published in JAMA (Case, Burwick, Volpp, & Patel, 2015) tested the step-counting reliability of various wearable fitness tracking devices and smart phone apps (see the data below). In addition to checking the reliability of various devices, the article makes an argument that, from a public health perspective, lots of people have smart phones but not nearly as many people have fitness trackers. So, a way to encourage wellness may be to encourage people to use the the fitness capacities within their smart phone (easier and cheaper than buying a fitness tracker). The authors never argue that fitness trackers are bad, just that 1) some are more reliable than others and 2) the easiest way to get people to engage in more mindful walking...

Warning: This isn't an easy story to listen to, as it is about life expectancy and terminal cancer (and how doctors can best convey such information to their patients). Most of this news story is dedicated to training doctors on the best way to deliver this awful news.   But Aronczyk, reporting for NPR, does tell a story that provides a good example of high-stakes applied statistics . Specifically, when explaining life expectancy to patients with terminal cancer, which measure of central tendency should be used? See the quote from the story below to understand where confusion and misunderstanding can come from measures of central tendency. " The data are typically given as a median, which is different from an average. A median is the middle of a range. So if a patient is told she has a year median survival, it means that half of similar patients will be alive at the end of a year and half will have died. It's possible that the person's cancer will advance quic...

http://www.washingtonpost.com/blogs/the-fix/wp/2014/09/29/ believe-it-or-not-some-members-of-congress-are-accountable-to-voters/ Philip Bump (writing for The Washington Post) illustrates the linear relationship between a U.S. House of Representative Representative's politics and their home district's politics. Yes, this is entirely intuitive. However, it is still a nice example of correlations/linear relationships for the reasons described below. Points for class discussion: 1) How do they go about calculating this correlation? What are the two quantitative variables that have been selected? Via legislative rankings (from the National Journal) on the y-axis and voting patterns from the House member's home district on the x-axis. 2) Several outliers' (perhaps not mathematical outliers, but instances of Representative vs. District mismatch ) careers are highlighted within the news story in order to explain why they don't align as closely with their distric...

This report from Pew  highlights the differences in opinions between the average American versus members of the American Association for the Advancement of Science (AAAS). For various topics, this graph reports the percentage of average Americans or AAAS members that endorse each science related issues as well as the gap between the two groups. Below, the yellow dots indicate the percentage of scientists that have a positive view of the issue and the blue indicate the same data for an average American. If you click on any given issue, you see more detailed information on the data. In addition to the interactive data, this report by Funk and Rainie summarizes the main findings. You can also access the original report of this data  (which contains additional information about public perception of the sciences and scientists). This could be a good tool for a research methods/statistics class in order to convince students that learning about the rigors of the scientif...

Kamenetz (reporting for NPR) talks about her book , Test , which is about the extensive use of standardized testing in our schools. Largely, this is a story about the impact these tests have had on how teachers instruct K-12 education in the US. However, a portion of the story discusses alternatives to annual testing of every student. Alternatives include using sampling to assess a school as well as numerous alternate testing methods (stealth testing, assessing child emotional well-being, portfolios, etc.). Additionally, this story touches on some of the implications of living in a Big Data society and what it is doing to our schools. I think this would be a great conversation starter for a research methods or psychometric course (especially if you are teaching such a class for a School of Education). What are we trying to assess: Individual students or teachers or schools? What are the benefits and short comings of these different kinds of assessments? Can you students come up with...

I'm an SPSS girl. I sit in my Psychology Department ivory tower and teach Introduction to Statistics via SPSS. SPSS isn't the only way to do the statistics. In fact, it is/has been losing favor among "real" statisticians. I recently had a chat with a friend who has a Ph.D. in psychology and works as a statistician. She told me that statsy job postings rarely ask for SPSS skills. Instead, they are seeking people who know R and/or Python. In order to better help our data-inclined students find work, I've gathered some information on learning R and Python. This probably isn't for every student. This probably isn't for 90% of our students. However, it may be helpful for an outstanding undergraduate or graduate student who is making noise like they want a data/research oriented career. Alternately, I think that an R class could be a really cool upper-level undergraduate elective for a select group of students. Also, if anyone is brave enough to teach thei...

Khan has been providing high-quality videos explaining...indeed...everything for a while now. Among everything are Probability and Statistics. Recently, they reorganized their content and added assessment tools as part of their #youcanlearnanything campaign in order to create self-paced lessons that are personalized to the user and include plenty of videos (of course) and personalized quizzes and feedback. 1) It requires the creation of a free account and selection of a learning topic (the screen shots below are from the Statistics and Probability course). 2) When you start a topic, you take pre-test to assess your current level. This assessment covers simple chart reading, division, and multiplication required for more advanced topics. If you struggle with this, Khan provides you with more material to improve your understanding of these topics. 3) After you complete the assessment, you receive your lesson plan. It includes the topic you select plus an additional introductory ...

UPDATE: The link listed below is currently not working. I've talked to Ariana Giorgi about this, and she is working to get her graph up and running again via Bloomberg. She was kind enough to provide me with a provide me with alternate URLs to the interactive scatter plot  as well as a link to the original text of the story . Ariana is doing a lot of interesting work with data visualizations, follow her on Twitter or hit up her website . _______________________________________________________________________________ This scatter plot (and accompanying news story from Bloomberg News)  demonstrates what a non-existent linear relationship looks like. The data plots CEO pay on the x-axis and stock market return for that CEO's organization on the y-axis. I could see where this graph would also be useful in an I/O course in discussions of (wildly unfair) compensation, organizational justice, etc. http://www.bloomberg.com/bw/articles/2014-07-22/for-ceos-correlation...