I stumbled upon an interesting article in the Wall Street Journal which talks about how a recent study suggests that people exercise significantly less than they say they do.
It immediately caught my eye because this is the time of year when many Americans have made it their New Year’s Resolution to lose weight or get buff.
Researchers gathered 1,571 men and women, ages 18 to 84 and close to half were overweight or obese. Participants were mailed questionnaires that accessed low, moderate and intense exercise time, as well as sitting and resting time. They also received an activity monitor that recorded the actual exercise time.
Interestingly enough, but not too surprisingly, researchers found a wide variation between the answers in the questionnaires and the data from the activity monitors, within the week long study.
To give you an idea of what the discrepancies were, men reported 56 and woman reported 52 more minutes of moderate exercise in the questionnaires than actually recorded by the activity monitor. But it doesn’t end there. When it came to intense physical activity, men recorded a total of 20 minutes in the questionnaire and women recorded 12 minutes in the questionnaire as opposed to an actual 2.9 and 2.4 minutes recorded by the activity monitor.
I find this study to not only be humorous but also insightful. This study does a great job of illuminating the idea of social desirability bias or in layman’s terms the idea that people answer questions in surveys based on what is seen as favorable and not by what they believe or do themselves.
I don’t blame some of the participants for fabricating their exercise times; it can be a very personal matter, especially if they are overweight. But the fact of the matter is that the tendency of social desirability bias poses a huge problem with conducting research with self-reports, especially questionnaires, which is exemplified in this study.
The problem with traditional questionnaires is that they are written without taking this idea into consideration. So plain and simple, traditional questionnaires can have major flaws when attempting to accurately extract information from participants in a study.
I believe this article, though not directly, does a great job of articulating this notion. Using the example of exercise is a great way to display the idea of social desirability bias to Americans, who may typically find it to be dry subject matter when shown in political polls.