Tuesday, April 03, 2007

Christian church attendance rates over time

(Part 2 in an 8 post series on Christian church attendance)

How have church attendance rates changed over time? Are Christians attending church less regularly over time? Since the GSS has been collected for several decades now, we can use it to examine these trends. Using the recoded, four-part attendance variable, here are attendance rates among all Christians by decade:

1970s 1980s 1990s 2000s
Weekly 40.7% 39.2% 37.7% 38.1%
Monthly 17.2% 18.9% 19.2% 19%
Yearly 33.6% 33.2% 33.7% 32.5%
Never 9.5% 8.7% 9.4% 10.4%

Since I have no idea how to create a proper table in blogger, and I like pictures, here is a graph of these data:

Overall it looks like there is a small drop in weekly attendance and a slight increase in monthly- and never-attendance, but overall attendance rates among Christians have remained mostly stable since the 1970s. So, are Christians becoming more nominal over time? Doesn't appear so.

Next: Attendance rates by denomination

5 comments:

Paula Harrington said...

Hey,
Glad to find your site. I have a question. I'm working on a book and wondered how accurate your results were. :)

Thanks,
Paula

Brad Wright said...

Good question...

Let's see, the data I analyze, from the General Social Survey, are widely accepted as a reasonably accurate reflection of American society.

I'm pretty sure that I have analyzed them correctly in terms of using a statistical package.

Probably the most potential for inaccuracy has to do with interpreting them. E.g., the label of "nominal" Christian is purely arbitrary, so in that sense any statistics associated with it are likewise arbitrary (or, at least, dependent upon an arbitrary assumption).

Thanks for asking.

Paula Harrington said...

I hope I didn't offend you by asking. I have a gift for that :)

Thanks for answering.

Brad Wright said...

No at all... in fact, people should ask that question of sociologists a whole lot more.

Brad Wright said...

You've got the numbers right.

As you point out, the threshold and measurement used here of being nominal is oversimplistic. But, given the assumptions made, you give a correct interpretation.