... what we drive and who we vote for
The Political Bumpers project has closed. It ran from September 20 to October 31, 2004 and collected a total of 1342 bumpers. This is what came of it.
- Project Summary
- Final Analysis: Unscientific Observations
- Final Analysis: A More Statistically-Correct Look
- The full list of bumpers
- About the Spotters
This project exists for one reason: to detail the results of an extremely unscientific research project about the correlation between the car someone drives and their political leaning. All data is collected informally by a group of people I know.
- Though Nader didn't run on Green Party ticket in 2004, any Nader bumper stickers (for past or present elections) were classified as "Green." Other third parties or unclear political leanings will be classified as "Other."
- Generally, if a submission could not be determined to be a specific type of car (for instance, a model reported that could be either a sedan or a wagon), it wasn't entered the database. The same goes for submissions that were "unclear" as to the type of vehicle for any other reason.
- Cars without a specific make/model were included as long as they could be classified as a certain type of vehicle.
- Spotters were allowed to count their own cars if they were displaying a bumper sticker.
- There were no fatalities due to overzealous bumper spotting. Thank goodness... my insurance wouldn't cover that.
by Ryan MacMichael
It feels kind of weird to be done spotting political bumper stickers. One spotter asked if there would be a recovery program for spotters that just can't stop taking note of bumper stickers. My response is that now the roads are a lot safer without the lot of us tailgating everyone else just to try and make out what model your car is.
This project came about fairly quickly. I conceived the idea one day and two days later, the site and database were designed and a crew of spotters was assembled. While I would haved liked to have more of a variety in terms of location, I feel like we managed to quite a good sampling. With over 1300 entries, we had a balance of 51% left-leaning stickers, 49% right-leaning stickers, and 1-2% "other" (which comprises Greens, Libertarians, and people who we just couldn't figure out... like IDs #571, 384, and 60).
I came into the project with some thoughts about how things would shake out. SUV owners would be primarily right-leaning, while hybrid owners would be entirely left-leaning. I figured minivan and pick-up owners would be right-leaning, while wagon owners would be split evenly. For the most part, my predictions were correct, but there were some areas where I was wrong.
For instance, the Small/Midsize SUV crowd is somewhat right-leaning, but not significantly (59%). However, that number jumped up quite a bit when looking at Large SUV owners: 83% (of an admittedly smaller sampling) are right-leaning. Indeed, Hybrid owners are entirely left-leaning and Pick-Up drivers are mostly right-leaning (69%).
What surprised me the most was that Economy car owners are significantly left-leaning (71%), and that's a number that dropped from over 80% at the halfway point. Also surprising is the leftie nature of Wagon drivers (67%).
Other car types divvied up relatively evenly. Minivans, Sports cars, Sedans, and Vans were all under the 60-40 range. (The sampling size of trucks was way too small.)
Among our spotters, we had eight left-leaning spotters and five right-leaning spotters. Political beliefs do not seem to have affected reporting as six of the 13 spotters reported more stickers from the opposite political leaning. There were no significant swings in either direction, though, even taking into account the spotters' locations.
Late in the process, I wrote up a few new reports that parsed through the data and broke out makes and models of cars and examined their respective political leanings. These reports are not as precise as the other reports since there is some room for error in terms of the parsing algorithm, but the margin of error is still small enough to make some notable observations about certain makes and models.
For instance, 2/3rds of Chevy owners are right-leaning. A staggering 77% of Honda owners are left-leaning and Toyota owners are almost twice as likely to be left-leaning as right-leaning.
Breaking it down further, we see that all ten of the Chevy Tahoes reported sported right-leaning stickers. Same with Dodge Dakotas. As mentioned, Honda owners are strongly left-leaning. Accord owners are 72% left, Civics come in at 80%, CRVs at 73%, and the Odyssey at 86% while the small Elantra sampling was split 3-2.
Stronger numbers appear on the right side for Ford F150s (75%) and Ford Expeditions (76%). All of the Toyota models leaning strongly left (ranging from 69-78%) except for the small sampling of Highlanders (which broke even) and the two trucks, Tacoma and Tundra. Volkswagen Beetle owners are left-leaning while the Volkswagen Passat seems to be more evenly split.
So, there you have it. And now I'll turn it over to Katherine who will explain things in terms that I probably don't understand.
A More Statisically-Correct Lookby Katherine Shaver