Everyday Research Methods

Critical thinkers eat a good breakfast.

My morning cereal box carried some helpful hints for back-to-school. The photo shows a section of the back of the box. Let's discuss the second point in the photo below.

In my opinion, this cereal box is doing a pretty good job of reporting on science here. For one, the "facts'' they state are attributed to a source. (However, they don't say what NHANES is--you should Google that acronym to make sure it's not the national association of breakfast cereal sellers!)

Another thing the cereal box does well is that the second Fuel Up Fact, about having a healthy weight, is not turned into a causal claim. It's just an association claim. Good job, cereal box!

Even though it's stated as an association claim, it seems possible that people will infer a causal claim from this association.It's even possible that the cereal company hopes that people will make this causal inference, and buy more breakfast cereal!

a) What is one possible causal claim people might (erroneously) infer from the association between skipping breakfast and having healthy weight?

b) Given that there is a correlation between skipping breakfast and having a healthy weight, what are three possible causal paths that could be responsible for this association?  (Hint: Think about temporal precedence and internal validity, as outlined in Figure 7.16)

 

What's so funny?

This piece, transcribed from last week's NPR radio news show, illustrates how scientists can study almost anything--even what people find funny.

Although the researcher profiled in this article identified himself as an anthropologist, the methods that the researcher used are identical to those a psychologist would use.

The journalist makes the research interesting and illustrates the science of humor with specific examples. However, it's worth focusing carefully on how the term experiment is used by the journalist in part of the story (not to mention the use of the word, "prove," which is not one a scientist would use).

The journalist is explaining how the jokes people laugh at can say more about the person than about the joke itself.

He recently conducted an experiment that proves this. He had volunteers listen to an edgy, stand-up comic named Bill Burr.

"He has a joke about why men make more money than women for doing the exact some job," Lynch says. "The punchline is, 'I'll tell you why. In the unlikely case we are both on the Titanic and it starts to sink, you get to leave with the kids and I get to stay. So call it a dollar-an-hour surcharge.'" 

Lynch also gave the volunteers a psychological test that measured their unconscious gender attitudes. What he found was that volunteers with traditional gender views — people who believed women ought to stay home, rather than go to work — laughed harder at that joke than volunteers with more progressive views.

a) Is the study described in the second and third quoted paragraphs really an experiment?  Why or why not?

b) How about this next example--is this study really an experiment, as the journalist suggests? 

In another experiment, Lynch sought to understand the connection between laughter and the psychological trait of self-deception.

Self-deceivers are people who don't see their own values, motives and beliefs clearly.

"I simply gave people a self-deception test and measured their facial expressions in response to a stand-up comedian," he says. "And there was a very strong association between the two."

Self-deceivers were less likely to laugh.

How might you revise this journalist's sentences to more accurately describe the type of research the anthropologist did?

Some intuitive statistics

Appendices A and B in the text summarize descriptive and inferential statistics. Recently, NPR ran a story about this summer's extremely high temperatures. I love how it describes the standard deviation as a "unit of wierdness." And I also like how they describe rare events--statistically significant events--as "a lot of snake eyes in a row."  Take a look:

http://www.npr.org/2012/08/06/158215252/are-recent-heat-waves-a-result-of-climate-change

This article is simply a great example of how to reason statistically.