Browse By

Religious Affiliation Doesn’t Move People To Exercise

One of the sales pitches churches and temples make is that belonging to a religious organization is a healthy choice, getting people out of their homes and in contact with other human beings who can support them in times of need. Some religious organizations go further, and claim that they can keep their members more healthy by assembling special prayer teams that target health problems.

Recent data show, however, that in one measure strongly correlated with population health, religious affiliation doesn’t seem to be beneficial for people’s health. A survey by the Centers for Disease Control and Prevention gathered information about the amount of exercise Americans had engaged in during the previous month. Another source, the American Religious Identification Survey, measured Americans’ religious affiliations. When combined, these two surveys can show if there is a relationship between religious identification and exercise.

The data show a correlation coefficient of about .6, meaning that among the states, there’s actually a weak negative correlation between the rate of religious affiliation and the amount of exercise. States with higher populations of people who are not affiliated with any religion have a slight tendency to have higher rates of exercise.

This doesn’t prove that lack of religion is causing people to engage in more healthy levels of exercise. It does argue against the idea, however, that high rates of affiliation with religious organizations brings a population into greater physical fitness.

One thought on “Religious Affiliation Doesn’t Move People To Exercise”

  1. Bill says:

    I don’t have an opinion regarding the subject, but as a lover of fine science reporting I’m moved to offer the following (I hope constructive) critique.

    1. A graph without labeled axes is literally meaningless. If it’s important enough to graph, it’s important enough to label.

    2. You indicate that your analysis “combines” the CDC and ARIS data. How’s that? Ordinarily, in order to test a correlation, a survey would record each individual’s exercise activity and religious affiliation. One could then test for meaningful correlations (or lack thereof). But combining two unrelated surveys into one data set? Doesn’t work. Judging from the cryptic phrase “among the states” I’m guessing you related each state’s average exercise level with each state’s religious affiliation.

    3. Just as when you have a hammer, everything looks like a nail, so too when you have a survey everything looks like a correlation. But to do a test such as you have attempted, one would have to match the two populations (religious and irreligious) with respect to other factors likely to affect either exercise activity or religious affiliation (age, income, education, employment). For example, the ‘religious’ population might be, on average, older than the ‘irreligious’ population, and this might well be a factor tending to reduce the religious population’s average exercise level…but match the test populations for median age and you might well find exactly the opposite. Or not. We’ll never know. Similarly, because the data are grouped at the state level (I think?) you’d have to take into account climate, as well…I suspect the irreligious coasts have more exercise-friendly weather than the religious fly-over states.

Leave a Reply

Your email address will not be published. Required fields are marked *

Psst... what kind of person doesn't support pacifism?