Welcome to the Gayborhood: mapping queer neighborhoods

@[email protected] : @shoshana.bsky.social

I think that place forms an important part of our lives. It determines a lot of our community and I think that's especially important for LGBT folks. So, I decided I wanted to map LGBT communities.

I think many queer folks might be thinking about where they want to move over the next 5-10 years. Laws in many states are getting scary for trans folk, women, and the queer community generally. Even in the absence of that, there are always people looking to move and I want to create something useful for people making decisions about their lives.

I've lived in Boston for a while and I know where lesbians are here, but I don't kow what lots of cities are like. Lesbians in Philly tend toward West Philly. Cities like Burlington, Albany, Ithaca, and Baltimore are a lot more lesbian than I knew. Raleigh-Durham? Durham seems to have a lot more lesbians than Raleigh. New York City? Lesbians seem centered around Bed-Stuy to Park Slope in Brooklyn while Manhattan has a lot more gay men.

I think there's useful insights that can come out of this like "where's a good place for lesbians that isn't one of the 5 most expensive cities in the US?" or "I'm gay but hate big cities, where can I go?" or "I got a job in Chicago, where will my community be?"

These maps aren't meant to be a be-all and end-all of queer communities, but I think they offer a useful visualization and a useful baseline when thinking about our communities.

How it works

The census has data on where same-sex couples live on many different levels like by city/town, by state, by metro area, and by census tract. These maps are based off the census tract data.

Initially, I worked off the city/town data. However, that created a bias toward small towns. Because borders are somewhat random, smaller places were more likely to randomly have a high percentage of queer folks. It would ignore neighborhoods in larger cities like Boston or Atlanta which were highly queer while highlighting smaller jurisdictions which were less queer than those neighborhoods. The way I think of it is that larger areas are more likely to regress toward the mean. The area of the United States is literally the mean. A city like Atlanta might include both very queer neighborhoods and less queer neighborhoods while a tiny place like Avondale Estates (population 2,960) only includes a queer area.

So I worked off the census tracts. However, those also had some problems. They're more uniform in size than cities/towns, but still have variance. They can also be so small that they're susceptible to noise in the data. I decided that I could use the census tracts to essentially create pseudo-neighborhoods of around 15,000 households.

Basically: take a census tract and find the percentage of same-sex households. Then look at neighboring census tracts and the percentage of same-sex households in those tracts. Keep going until you've got yourself a pseudo-neighborhood of around 15,000 households. It also puts more weight to people closer. Same-sex couples living within the tract itself count stronger than couples who are farther away, but within the same pseudo-neighborhood.

It also cuts off a neighborhood size at around 20 square miles (50 square kilometers). This was to solve what I called the Westhampton Problem. Westhampton is a small town next to Northampton. Creating a pseudo-neighborhood of 15,000 there meant bringing in a large portion of Northampton whose lesbian population is overwhelming large - both because it's a much larger city and because it's extremely lesbian. If a census tract in a city isn't that queer, but the surrounding tracts are very queer, it's probably a couple blocks walk. With Westhampton, it was pulling in census tracts that were many miles away.

Limitations

Because this is based off census data, it isn't going to account for single queer people. It also won't account for some trans people who might be in a queer relationship that would get counted as heterosexual by the census - for example, a trans man married to a woman would simply show up as heterosexual in the data.

Still, I think that queer couples often end up in similar areas as queer singles and trans folks who might not get counted correctly likely live in similar areas as those who are counted correctly by the census.

As I mentioned previously, this isn't meant to be the be-all and end-all of queer communities. However, I think the map provides a useful visualization of the center-of-gravity of many queer communities across the US.

Future Work

There might be some refinements that I should make to the calculations and groupings. There might be other sources of data that I should incorporate beyond just the census data. I'm definitely open to ideas on how to make it better.

I think there are two things that stand out to me for future work. First, I think it would be useful to correlate this data with housing affordability. I love the queer community in Boston and I so want Boston to be affordable for everyone who wants to live here. At the same time, I think it's important to understand what is actually affordable today.

Second, I think it would be interesting to understand how queer communities have grown over the past decade. I think that a thriving queer community is one where the community can grow and welcome more people.