Monday, March 28, 2016

Pew Research Center's "The strong relationship between per capita income and internet access, smartphone ownership"

This finding is super-duper intuitive: A positive, strong correlation exists between national per capita income and rates of internet access and smartphone ownership within that nation.

Because it is intuitive, it makes a good example for your class when you teach correlation to your baby statisticians. This graph is  more engaging than your average graph because the good people at Pew made it interactive. You can see which country is represented by which dot. You can also see regional trends as the countries are color-coded by continent/region.

For more context and information on this survey, see this more extensive report on the relationship between smartphone/internet access and economic advancement. This report further breaks down technology usage by education level, age, individual income, etc.

This data is also useful for demonstrating the distribution of wealth in the world and variability that exists among countries in the same region/on the same continent,

Monday, March 14, 2016

Kennedy's "'Everybody Stretches' Without Gravity: Mark Kelly Talks About NASA's Twins Study"

In addition to being an astronaut, Scott Kelly is one-half of a pair of twins and a lab rat for NASA researchers studying the effects space travel on the human body. This NPR story details how NASA has been using twin research in order to learn more about the side-effects of prolonged time in space as the agency prepares to go to Mars.

Scott and his twin, Mark (also an astronaut!), have been providing all manner of bio data to researchers. In particular, researchers are interested in the effects of weightlessness as well as exposure to space radiation on aging.

This story provides a good example in class, as you can discuss twin AND longitudinal research. I think you could also use this example to introduce the concept of paired t-tests.

UPDATE 2/9/2017:

Preliminary research is available if you want to flesh out this example. 

Monday, March 7, 2016

Granqvist's "Why Science Needs to Publish Negative Results"

This link is worth it for these pictures alone:

I know, right? Perfect for teaching research methods and explaining the positivity bias in publication.

These figures also sum up the reasoning behind the new journal described in this article. New Negatives in Plant Science was founded in order to combat the file drawer problem. It publishes non-significant research. It is open access. It publishes commentaries. It even plans special issues for specific controversial topics within Plant Science. Which absolutely, positively are NOT my jam. However, the creators of this journal hope that it will serve as a model for other fields. Given the recent flare up in the Replication Crisis (now Replication War?), this new journal provides a model for on-going, peer reviewed, replication and debate.

I think this journal (or the idea behind this journal) could be used in a research methods class as a discussion piece. Specifically, how else could we reduce the file drawer problem? The Open Science Framework offers a new model for transparency in research. Retraction watch provides a highly visible platform for tracking research that makes it through peer review but is later pulled by a journal. What else could we be doing to better vet our research?