Oakwood in Red & Blue

The Big Sort, a somewhat new political book, has received some notice in unusual places. Dayton Most Metro parented a thread on it, which died a quick death, and the Louisville Courier-Journal did a somewhat favorable review in their Saturday book page.

The premise is that Americans are sorting themselves out into homogeneous communities, and this has political implications.

The concept isn’t that new. In the mid 1970s (1974) Herbert Gans introduced the notion of taste cultures as a more nuanced approach to socioeconomic status and class. Between 1976 and 1978 the somewhat similar PRIZM market segmentation concept was developed. Populations were refined into market segments, called clusters, and the clusters where mapped (by zip code and media market), so there was a geographic turn to clustering. A related concept developed around the same time (1978) was the VALS (Values/Attitudes/Lifestyles) which was somewhat less geographical .

The PRIZM approach was popularized via an 1988 book, “The Clustering of America”.

The Big Sort shows how politics reflects this geographic clustering of like-minded folks. One of the points of the book is that beyond the red/blue divide, political geography at a more granular scale reveals counties and cities that become more polarized over time, becoming more and more Democrat or Republican. The author uses the increasing number of “landslide counties” to measure polarization. If the Democrat (or Republican) vote was 20% over the vote for the opposing party (ie a “landslide” win) the county was deemed polarized.

For Montgomery County, in the 2000 & 2004 presidential elections, the vote was nearly exactly 50/50, with the GOP making a slight increase in 2004, so not polarized at the county level. Below that things get more interesting.

The Big Sort in Oakwood.

Oakwood is the case study as it is small enough to work with (and had a precinct map on the city website), and also due to the local stereotypes about the suburb as a sort of stuffy, conservative, blue-blood kind of place . This would imply a sort of conservative Republicanism.

The reality is more nuanced.

The following chart shows the Oakwood precincts by election in 2006. From left to right the Senate race, Governor and statewide offices, followed by two county commissioner races and a county auditors race. The blue-Dem/red-GOP graphic convention is used to show which party carried a precinct. The number shows the point spread for the precinct. In other words it shows by how many % points the Dem or GOP candidate carried the precinct (negative number shows a GOP win).

For example, for precinct V, Senate race, red color indicates the Republican carried the precinct, and the number shows the Republican won it by 26.5 percent points over the Democrat. Since the interest here is in political polarization….”landslide precincts”…the degree of partisan support is of interest , as well as how frequently a precinct goes Dem or GOP.

The bottom bar shows which party won the suburb by office, and just by this Oakwood is pretty Republican; the GOP carried nearly all the offices listed here



Oakwood Political Neighborhoods

Mapping the results of the chart by precinct, one can easily see how Oakwood divides up into ‘political neighborhoods’, with Far Hills Avenue being the obvious divide. Far Hills has long been a social divide, with “West of Far Hills” being the estate district. Yet it’s interesting to see this also map out politically.

Things get more interesting in northern Oakwood, where the Shantz Park area is swing precinct splitting its vote between Dems and Republicans. There is also a strong GOP presence in precinct B, east of Far Hills.


Let’s look at the Senate and Governors race to see if polarization is evident.

For the Senate race the GOP did well. Four precincts in northern Oakwood polarized towards the GOP, including a leaning GOP precinct. The Dem precincts in southeast Oakwood went Dem, but with tight point spread


For the governor’s races, the GOP apparently had a weak candidate as he didn’t carry any GOP precinct over 20%, and lost two of the swing precincts. In contrast two of the Democratic leaning precincts polarized, with point spreads 20% or over in favor of the Dem candidate.


Revisiting the chart, and looking at landslide wins one can see the areas that are GOP tend to deliver more landslides while the Democratic areas are less likely too. Precinct O seems to be the most reliable Democratic-leaning precinct.



Without having multiyear comparisons one can’t say if there is a Democratic or Republican trend in Oakwood. Given the suburbs’ reputation and the evidence of mostly Democratic and swing precincts, one can infer that this suburb, or parts of it, is more Democrat than the stereotype. What’s also evident is that the suburb isn’t that polarized. The point spread depends on the office and candidate, and there is considerable ticket splitting in at least two precincts.

Next we will look at a much more politically polarized suburb; the unincorporated parts of Washington Township