Preliminary Findings

After about a month or two, I collected over 5000 hymns. I started play around with the data in Excel to see what kinds of patterns appear. Turns out, they're pretty interesting and some of them are not exactly what you'd expect.

These results are based on the first 15 or so wards' hymns that I collected. Since each ward is represented by anywhere from just a few hymns to several years' worth, the results are certainly not a representative sample of the Church. Hopefully, when I get a couple hundred more wards here, that'll all even out.

What is the most common hymn?
From my data so far, it looks like As Now We Take the Sacrament (169) is the most common. The fact that it's a sacrament hymn doesn't come too much as a surprise. In fact, 20 of the top 25 hymns are sacrament hymns. Within them, I thought for sure that I Stand All Amazed (193) would be most common (turns out it's 3rd).

The following list shows the top ten most common hymns, excluding sacrament hymns, with their average frequency.
  1. I Know That My Redeemer Lives (136)—2.21 times per year per ward
  2. Because I Have Been Given Much (219)—1.86
  3. How Firm A Foundation (85)—1.44
  4. Come, Come Ye Saints (30)—1.36
  5. I Believe in Christ (134)—1.39
  6. Now Let Us Rejoice (3)—1.33
  7. The Lord Is My Light (89)—1.33
  8. Called to Serve (249)—1.30
  9. Lord, I Would Follow Thee (220)—1.28
  10. The Spirit of God (2)—1.25
There are about 15 other non-sacrament hymns that are sung an average of slightly more than once per year.

What hymns are the least common?
In my data, there are 56 hymns that have never been sung, with another 20 being sung only once. Those numbers should decrease as I gather more data. Note though, that that's almost 25% of the hymnal that we hardly ever use.

Why are some hymns more common than others?
Since starting this project, I have been able to learn some simple statistical models that can help answer this question. Just considering some superficial aspects about the music alone, there really is no way to predict if a hymn will be popular or not. There is a hint of a slight tendency for hymns that don't go as high to be sung slightly less often, but it doesn't reach any level of significance, so this is only speculation. I would need to collect much more data from many more wards to see if this holds true.

In conversations about this project, it seems like most people assume we only sing about 30 to 50 hymns. The number is surprisingly a bit higher. If a new organist learned the 50 most common hymns, they'd only be able to play only about 47% of the time. However, if we cut the hymnal in half, keeping the most common 171 hymns or so, most wards wouldn't notice a difference about 92% of the time.

Individual Wards' Trends
I have enough data from most wards that I can get a start to get a general idea about their individual tendencies. I've got a formula that will give each ward a score that quantifies how unusual their hymns tend to be. If you're interested in seeing how your ward ranks, just let me know.

Regional Trends?
While most of my data is from wards the United States, I do have information from one ward in England and another in New Brunswick, Canada. Given though that the information from each ward I have is unique in some way, there's no way to tell if this is due to regional trends or due to the personality of the music director. I hope to have lots of data from many wards so that I can spot more subtle regional trends that aren't unique to just the ward itself.

Annual Trends
There are some annual trends that are fairly significant, though not at all surprising. Using a formula in Excel, I can normalize the years so that I can look at which hymns are the most common during the first week of the year, second week of the year, all the way to the last week of the year. Using that, I'm able to quickly identify the hymns with obvious peaks around major holidays (Christmas, Easter, Thanksgiving, Mother's Day, Father's Day, Pioneer Day, and 4th of July). Hopefully, with more data, I can find some less obvious trends, like perhaps those that are more common on other holidays (Memorial Day, or Labor Day perhaps), Fast Sunday, and around church history events (though the April General Conference and Easter probably make singing restoration songs difficult around April 6).

Special Musical Numbers
Doing a detailed analysis of special musical numbers would be out of the scope of this project. However, I do have some superficial results. On average, there are 3.35 hymns per week. Taking Fast Sundays into about, that means that there's somewhere about a 50% chance there will be a special musical number instead of a intermediate hymn. Now, this is just an average—some wards have musical numbers almost every week while others have virtually none. It's also worth mentioning that there tend to be more special musical numbers (unsurprisingly) between September and December (thanks to primary programs, Thanksgiving and Christmas), and around Easter.

No comments:

Post a Comment