As a way of introducing the data set and setting out from the onset the natural imperfections inherent in doing data analysis via a web based tool, I thought I would begin with a brief explanation of the intent of the survey and some preliminary charts showing the demographics provided for those who took the survey.
The survey was initially a test of the Church Engagement Survey promoted by Gallup. After doing my previous post based on the engagement survey we take where I work, ricke mentioned the book “Growing an Engaged Church” and I became very interested in this. The basic premise of the book is that members of a church who are emotionally engaged in the church are also more spiritually committed and are better able to achieve desirable outcomes. I decided to “experiment upon the word” and to run an analysis of this correlation myself. So, while a set of my questions measured basic demographics, there were a set of questions that measured engagement, an additional set which measured what I will call discipleship, and another set which measured desired outcomes. (short spoiler: The book was awesome! I bought one for each member of my bishopric.)
Secondarily, the local philosophy of the church has been greatly shaped by some recent visits by general authorities who have taught the need of the church to focus on what may be Boyd K. Packer’s most famous statement “The study of doctrine will change behavior quicker than the study of behavior” (citation unknown). The basic idea is that the membership should not worry about how the gospel is being applied by others so much, but focus on teaching correct doctrine(The Why), which leads to correct principles (The What), which then leads to correct application (The How). This is all a leads up to me saying that we are all generally well aware of some points that are discussed ad naseum throughout our community, and especially here on the blogs. I added a rather large set of “reductio ad absurdum” doctrinal questions to the analysis as a test case to see if those instances where there was doctrinal certainty (whether in the negative or positive direction) lead to increased engagement one way or another.
While I can now see where the second intent was somewhat misguided (more on that later) and proved to be ultimately untenable in that I would have been better served by only having questions which measured belief in baseline LDS concepts, I was able to gather some very interesting results in any case, which I will share as these posts progress. Further, this was much more fun.
Who took the survey?
First and foremost, 80% of the responses were from regularly attending (more than once monthly) and holding a calling. Another 10% also attend church regularly, but do not currently hold callings. The final 10% either did not consider themselves members of the church, or did not attend church regularly (more than once a month). This lines up pretty much perfectly with my previous demographics on engagement and so, like then, we have to accept that we are primarily dealing with active participants in the church and thus will be biased in that direction.
In the current survey, we attempted to break the barrier which was imposed on the previous survey of “That’s just what bloggers think.” And so we sent the survey not only via NCT, but also via personal connections through email, Facebook, and Twitter. (Anecdotally, while 20 people from Twitter did look at the survey, 0 filled it out. Apparently people who are in the mode for 140 characters or less are not interested in 60 question surveys!). We also asked that those we invited to take the survey spread the word. Here is what we ended up with:
After deleting duplicates, we were left with 1323 responses, 49% female, 51% male, which comes fairly close to the demographical 50/50 split in the church (according to A Statistical Profile of the Mormons ). While NCT was our primary draw for the church, we nearly collected as many responses from “unknown”. Unknown typically has 3 different causes. Either the user is opening the link in a new tab or browser, and thus its source is not detectable, or the url is being copied and pasted into a browser, or finally, an email tool like Outlook is being used. Anecdotally, we have reason to believe the 3rd option to be the most probable for the majority of these unknown responses, and not just any emails, but especially emailed messaged from Facebook. One surprise here was the high response rate from “Cougar Ute Forums” which I had never heard of before, so thanks for who ever linked the survey there.
To further break down how our respondents came in, here is the same data as above but in an area chart with time overlay:
The 4th and 5th of April marked our initial email campaign, post of Facebook, and post on NCT. The second spike was on the 8th, when Rico kindly posted for us at Mormon Matters. The 12th was when we officially started a group on Facebook. (Thus our anecdotal evidence relating unknown directly to Facebook. We have more group members on Facebook than responses marked from Facebook in the survey). We then had a lull until on the 23rd we posted our final request on Facebook and NCT, and were linked at BCC to the NCT post. The 26th was when Cougar Ute Forums picked us up and that basically carried us to the end of the survey on the 30th, at which point we closed the survey.
Most of the responses were married, with a good number of respondents also not having ever been married. Divorced and Widowed were not represented much.
Most of the respondents were the same age as me. This is one of the problems with sending surveys to your personal friends. We did have a very diverse group overall in terms of age though, so it worked out pretty good.
I used age at baptism and current age to calculate how many years each person had been a member. Only 17% of the dataset was considered a convert (person baptized over the age of 8) but the majority of converts had been members 10 or more years, so it is reasonable to assert they know their ropes.
How international is this study?
(Yellow is Female Responses and Green is Male)
The US was still over 93% of all responses, breaking down into metro areas as follows.
Unsurprisingly, SLC/Provo was our largest contributing area, but did get big pulls from San Antonio, TX (probably due to our being locals), San Diego, NYC, DC, Phoenix and…Kansas?
Two observations from the demographics:
1. When looking at Marital Status in conjunction with Membership Status, I did notice that the divorced members as a block were much less likely to attend church regularly. While I can not pretend to know the cause of this, this does initially indicate to me that the church probably could exert more attention into the status of divorced members.
2. In looking at the most active converts to the church (attending more than once a month and considered self a member), after we ignore the 9 year olds who join the church, it seems that the most active converts long term are between the ages of 18 and 21. So tell your missionaries to go hang out on college campuses.
However, see if you can find the “Rainbow Discussions” they used back in the 70s, as those got the best results.
Next- Measuring Discipleship