Everyday Research Methods

Tilt your head down to appear more intimidating

According to the study, Captain Marvel's eyebrows have something to do with her dominant pose. Credit: Collection Christophel/Alamy Stock Photo

Captain Marvel looks tough. Perhaps one reason is the tilt of her head, which enhances the appearance of dominance, according to a series of studies that appears in Psychological Science. The studies were summarized by journalist Alice Walton for Forbes. 

Here are some excerpts from the journalist's description of the study:

...there’s a way to communicate dominance that doesn’t involve moving the facial muscles at all: just tilt your head downward slightly and maintain eye contact.

In one experiment, participants in the University of British Columbia study rated [several] avatars who were tilted upward 10 degrees, tilted downward 10 degrees, or level (neutral). Then they rated statements designed to gauge how dominant the person appeared, including “This person would enjoy having control over others,” and “This person would often try to get his way regardless of what people may want.”

Participants rated the downward-tilting heads significantly more dominant than either of the other two positions. When the experiment was repeated using photos of actual people rather than avatars, the same pattern was found.

a) The journalist labels this an experiment, which it really is. What is the independent variable? What are its levels? Is the IV manipulated as independent groups or within-groups? (By the way, the Forbes article shows photos of the three levels.)

b) What is the dependent variable in this study?

c) What's the experimental design: Posttest only? Pretest-posttest? Repeated measures? Concurrent measures? 

d) Sketch a bar graph of the results described here: "Participants rated the downward-tilting heads significantly more dominant than either of the other two positions."

e) What do you think of the construct validity of the dependent variable? Does this operationalization really get at "dominance?" Why or why not? What questions from Chapter 5 might you ask here? 

f) The journalist mentions that "the experiment was repeated using photos of actual people rather than avatars". It's an example of replication. Which type of replication study would you call it: direct replication? conceptual replication, or replication plus extension? (Chapter 14 defines these terms.)

g) By replicating the avatar study on actual people, which validity (or validities)  are these researchers enhancing? 

More Advanced Questions

After establishing the basic effect, the researchers conducted some additional studies to test the mechanism, or mediator. Read this paragraph (written by the journalist) to decide what the mechanism was, and think about how they tested it. 

In another experiment, the team found that the key to the effect was the eyes and eyebrows: when they showed participants faces with the eyes/eyebrows occluded, the tilted head/dominance effect disappeared. But then they showed participants pictures of just the eyes and brow portion of the face, the effect was present, which suggests that this part of the face is both necessary and sufficient—meaning it's not only required but is all that's necessary to elicit the effect.

The above argument suggests that the researchers' theory goes like this:

    Tilting head downward ---> changes in eyebrow position --> dominant appearance

h) To find evidence for their theory, the authors tested four conditions in a 2x2 factorial design. Downward tilt (vs. neutral), and face only vs. eyes/eyebrows only. You can see examples of the stimuli they used in Figure 4 of the original article in Psychological Science. Sketch the predictions of these two studies. Like this:

If changes in eyebrow position are important for tilted faces to appear dominant, then the results should show [sketch a result].

If changes in eyebrow position are NOT important for tilted faces to appear dominant, then the results should show [sketch a result].

 

Of head horns and cell phones

Did the study collect the right data to conclude that "screen time found to cause 'head horns'?"  Photo Credit: Nature Scientific Reports

Maybe you heard the story last week about how cell phones are making horns grow on people's heads. Take a minute to watch this  short video of the morning crew from ABC news discussing it. The link is captioned, "Screen time found to cause head horns" In this video, you'll notice that the journalist announces, "New research shows that constantly looking down at your screen could give you so-called 'head horns'" .

And here's another video that recycles the same material along with extra commentary under the title, "Tech neck causing skull horns"

There's so much to unpack about this story. If you read beyond the first journalists' coverage, you'll find some great analyses of this study and the journalism around it. Turns out, the study itself had problems, and the journalistic coverage didn't help. 

Take a moment to read one of these two critiques of the "head horns" moment: 

Smartphones aren't making millenials grow horns. Here's how to spot a bad study. or Hold the phone: Are you really going to grow a horn from slouching over your phone? 

Use the information in the two critical stories to research the following questions.

First, there are issues related to the "journal to journalism" cycle (Chapter 1), 

a) The original journalist changed the name of the protuberances to "horns." Every journalist thereafter copied this nickname in their own reporting.  What were they originally called? Why do you think the journalist changed their name?

b) In Chapter 1, you learn that journalists don't always pick the most important studies to summarize. Why do you think journalists picked this particular story? Do you think the study is important? 

Second, there are problems with the study itself, which many journalists did not detect--at least at first.

c) One criticism with the study concerns its sample. What are the criticisms raised about this study's sample of 1200 x-rays?

d) Another criticism has to do with the researcher's potential conflicts of interest (CoIs). What were these, and why are CoIs a potential problem in scientific research and publication? 

e) Another criticism has to do with the proposed link between cell phone use and the "horns."  In order to establish a link between having these protuberances and screen time, the scientists would have had to measure both variables and compute the correlation. Did the researchers do this? 

f) Finally, many of the news headlines used causal verbs ("Screen time found to cause head horns"). Even the scientists overclaimed here when they were interviewed by the press. What kind of study would a researcher have to do in order to establish such a causal link? 

g) Are there any additional problems you discovered?

The news cycle around the head horns study illustrates the benefit of cross-reading and triangulating. If you come across a journalist's story that's just too weird to be true, read about the same topic on other sources, or on snopes.com.