These survey results are really striking in how different they are in 2010 compared to 2022. In 2010, the direction of the lines is unmistakable: down. That means that as household incomes increase, the likelihood of attending religious services every week drops. Those drops are not that small either. Among those with a college degree, the dip in weekly attendance from the bottom end of the income spectrum to the top end is about eight percentage points. It’s about the same among those with some college. For those with a high school diploma or less the impact of income is more modest, with a three percentage point drop.

The 2022 results are just completely different compared to those from 2010. The lines clearly aren’t pointing downward at nearly the same rate. For instance, the drop in weekly attendance from the left to the right side is just two percentage points among those with a college degree. That is statistically significant but not substantively large.

For those with some college education, the trend line is completely flat. That means that income has zero impact on religious attendance. Definitely a big change from the 2010 results. For those at the lowest end of the education spectrum, an increased income actually results in a higher likelihood of attending religious services. It’s a modest increase: less than three percentage points. But it’s statistically significant.

Something happened in the last decade or so when it comes to religious attendance and household income. That’s very apparent to me from these regression results. It used to be a negative relationship. Now it’s likely no relationship in the aggregate, with a little variation based on education.

But here’s something I haven’t mentioned to this point — look at the red line in the 2022 graph for college educated folks. They attend religious services at higher rates than those with less education, regardless of income. That was also the case among college educated respondents in 2010, as well.

I am going to keep working at this, by the way. I don’t think I have a fully formed understanding of the interaction between income, education and religiosity (in it’s various forms). I am thinking that there has to be a racial aspect I’m missing. Or something to do with generations. Or maybe age.

Rest assured — I don’t think that I will ever run out of questions to answer when it comes to income, education and religion.

This piece was originally published in Ryan Burge’s “Graphs About Religion” Substack.