You may remember news last summer that researchers had conducted a nationwide Amish population survey.  I just had a chance to read an interesting paper detailing their findings.  Here are 5 of them:

1. New settlements  On average a new settlement is founded every few weeks.  This rate has been increasing. From 1990-2000, a settlement was founded, on average, every 5.33 weeks.  2000-2010 saw this number decrease to 3.66 weeks.  In the first 2.5 years following 2010, the authors found it to be just 2.55 weeks.

2. Many small settlements  Over half of the 450+ Amish settlements are just one church district in size.  However these one-district communities only make up about 13% of the total number of church districts.

3. “Mega-settlements”  The top 4 Amish settlements (Holmes County, Lancaster County, Elkhart-Lagrange Counties, Geauga County) account for 1/3 of the total Amish population.

Lagrange County Large Amish

Will Lagrange County one day be majority-Amish?

4. “Most Amish” counties  The two “most Amish” counties are Holmes County, Ohio, and Lagrange County, Indiana, with around 42% and 38% of their populations being Amish, respectively.  The authors suggest that Holmes County could become a majority-Amish county within the next 15-20 years.

5. Amish clusters  If you want to visit many Amish settlements in a short amount of time, head to Western Pennsylvania. In the three contiguous counties of Crawford and Mercer in PA, plus Ashtabula County in Ohio, you’ll find 22 Amish settlements.  This represents almost 5% of all Amish communities in North America.

You can read a draft version of the paper (“The Amish Population: County Estimates and Settlement Patterns of the Old Orders”, by Joseph Donnermeyer, Cory Anderson, and Elizabeth Cooksey) here (page moved).

Here are three thoughts I had while reading this paper:

1. Deceptive one-district settlements  A one-district settlement can have as few as 3 families or as many as 30-40.  So a mature one-district community could be 10 or more times as large as a brand new one-district community.  This introduces uncertainty when comparing the one district settlements.

2. Large vs. Medium vs. Small If you selected a random Amish person, there is a 1 in 3 chance they would live in one of what we could call the 4 “mega-settlements”.  There is a real difference between living and raising children in a community of 100+ church districts vs. one of just a handful.  For example it is essentially impossible for such a person to know all the people in his or her community, each of which contains thousands of Amish households.

Hicksville Ohio Small Amish

Hicksville, Ohio: 99 years old, 1 district in size

When writing about a community I sometimes pause when having to describe its size.  Is such-and-such community “large”, or just “medium”?  When does a settlement stop being “small”?  There is a big drop in size from #4, Geauga County, to the next largest settlement of Adams County, Indiana, though I’d still think of Adams County with its 45-50 churches as “large”.  Does a settlement become “large” when a  middle-aged adult can no longer place all the families in it?

For what it’s worth, I have tended to think of “small” communities being from 1-4 districts in size, “medium” those from 5-20, and “large” those above 20 churches.  This designation is arbitrary, though there may be something to the name-your-neighbor theory.  I need to ask some Amish people what to them feels like a large community, etc.

3. Community clustering Amish may be drawn to an area by the presence of other Amish, resulting in unrelated settlements living in close proximity to one another (as with the 3 PA/OH counties mentioned above).  Other areas where this has been seen include New York’s Mohawk Valley and the southern Indiana counties of Orange and Washington.

This makes sense in that a lot of the footwork has already been done–the local ground has been plowed, so to speak.  The land is probably decently-priced.  You have a sense of how English take to Amish in the area, such as how they tolerate buggies and their leave-behinds on the road.  And the economic prospects for survival have been at least somewhat tested as well.

Tags: