Sue Frantz

Do Fitness Trackers Work?

Blog Post created by Sue Frantz on Sep 21, 2016

When covering research methods or as a research methods boost in health psych in Intro, ask your students, “Do fitness trackers, like Fitbit, work? If you were a psychological scientist trying to answer this question, how would you do it?”

 

Give students a couple minutes to think about this on their own. Next, ask students to work in pairs or small groups to design an experiment that would help answer this question.

 

As you circulate among the groups, make sure the groups are answering the question “Do fitness trackers work… at doing what?”

 

As discussion is winding down, bring the class back together and ask for groups willing to volunteer their experimental designs.

 

There should be some sort of random assignment to wearing a fitness tracker or not. Ask students why? [Because if you compared existing users with existing non-users, the users may already be more motivated to engage in physical activity.]

Ask students to identify the independent variable [fitness tracker usage] and the experimental condition [fitness tracker] and control condition [no fitness tracker]. How long did students think participants should use/not use a fitness tracker to ensure a fair test? Why that amount of time?

 

Ask students to identify the dependent variable(s) [perhaps weight loss].

 

With class discussion on the design wrapping up, share with students the results of such a study (Jakicic, et.al., 2016). Participants (470 of them) were six months into a 2-year weight-loss study when they were randomly assigned to either wear a fitness tracker that included a website for monitoring diet or self-monitor exercise and diet via a website (74.5% completed the study; every six months, participants were given $100). Note that this study did not have a no-treatment control group.

 

Ask students to predict the results by a show of hands or via an audience response system.

A. Fitness-tracker users lost the most weight

B. Self-monitors lost the most weight

C. Fitness-tracker users and self-monitors lost about the same amount of weight.

 

Ready for the results? Those assigned to wear fitness trackers lost 3.5 kg (7.7 lbs). Those assigned to self-monitor lost 5.9 kg (13 lbs). There were no differences in the groups at the 6-month mark, the point in the study where they were randomly assigned to wear the fitness tracker or self-monitor. But at the next three check-ins (12 months, 18 months, and 24 months), the self-monitoring group had always lost more weight. Did your students guess right? Were they surprised by the results?

 

Some explanations for these results are offered in this NPR story. But before you share these with students, ask students to generate some hypotheses as to why the self-monitoring group lost more weight than the fitness-trackers. If time allows, give students a couple minutes to think on their own before sharing in pairs or small groups.  Ask student volunteers to report out their hypotheses. Write the hypotheses where students can see them. If you’d like to send students off with a take-home assignment, assign students to design an experiment (but not conduct it!) that would test one of the student-generated hypotheses. Students should identify their independent and dependent variables and anything else they would do that would eliminate confounding variables. You can either let students choose the hypothesis, or assign hypotheses by last name, e.g., “If you’re last name begins with A through F, you have hypothesis 1.” Students can submit as a written assignment, or if you have time at the beginning of the next class, give students an opportunity to share their designs with each other, and then take a few minutes to ask volunteers to share their designs.

 

Reference

 

Jakicic, J. M., Davis, K. K., Rogers, R. J., King, W. C., Marcus, M. D., Helsel, D., . . . Belle, S. H. (2016, September 20). Effect of wearable technology combined with a lifestyle intervention on long-term weight loss: The IDEA randomized clinical trial. Retrieved from http://jama.jamanetwork.com/article.aspx?articleid=2553448

Outcomes