Boston San Francisco

Let's talk about stats (Part 2 of 2)

Metrics matter, and can have policy consequences.
By Constantine A. Valhouli  |  January 13, 2018 3:42 PM

Castro, San Francisco by Flickr user torbakhopper

In connection with a recent study, we were trying to compare the size and percentage of LGBTQ populations in major cities. The challenges we had in finding and parsing accurate information merited their own post.

San Francisco was an epicenter of the gay rights movement; the first city to elect an openly gay mayor, Harvey Milk; and the Castro neighborhood remains an icon of gay culture. As a result, San Francisco is one of the cities that is often referenced when discussing LGBTQ statistics.

What percentage of San Francisco's population is LGBTQ? Well, it depends.

The U.S. Census is a natural place to begin. However, the questions that are – and aren't – asked are going to shape the nature of the data that is gathered. For example, the census only asks questions pertaining to same-sex households, rather than to individual sexual orientation.

This approach skews LGBTQ data lower from the outset. And it can appear even smaller as a percentage of total population, depending on what data set it is being compared to.

At NeighborhoodX, our data models of cities are built from the bottom up, consolidated from market reports of individual neighborhoods. As a result, we are working with discreet data sets that collectively form a picture of the geographic borders with which we're working.

However, the U.S. Census defines "urban areas" as not just the core city, but the adjacent suburbs and even towns much further out.

Why does this matter?

Let's work with small numbers to make it easy to grasp. Let's assume there are 1,000 people in the city, and 5,000 in the broader "urban area" which includes the suburbs. Let's also assume that there are 250 LGBTQ people in that region. However, only 150 of those are same-sex households; the rest are single or merely dating.

The actual LGBTQ percentage of population in the city proper (250 people out of 1,000) is 25%. However, because of the nature of the census questions (150 people in same-sex households out of 1,000), it only reflects 15% of the population as LGBTQ.

Now, let's assume that the LGTBQ population is clustered in the city and is less represented in the immediate suburbs of the "urban area." If the census compares LGBTQ data to the city population of 1,000, it represents 15%. However, if the census compares LGBTQ data to the "urban area" population of 5,000, it represents only 5% of the population.

From a Census training seminar: "These estimates of same-sex couples are in no way meant to stand in for estimates of the LGBTQ population as a whole ... Anyone who's using them as a proxy for a larger group of people we just that [it is] ill-advised." However, these numbers are often used in that manner – and this can have significant policy implications. Without an informed view of the true size of a significant segment of the population, it is difficult to know, for example, how to allocate funds and resources to the appropriate social programs. According to NPR, "Reliable data about the lesbian, gay, bisexual and transgender community are hard to come by, and advocates say policymakers need that information to make informed decisions."

To compound matters, the U.S. Census will no longer even collect information about LGBTQ households beginning with the 2020 census, according to The Hill. Praveen Fernandes, writing in the New York Times, put it succinctly:

"In my roughly 20 years working in the federal policy arena, few things have become clearer to me than the importance of data. If something is not counted, it is neither seen nor understood. For all intents and purposes, it does not exist ... Failing to collect good data on sexual orientation and gender identity allows policy makers and elected officials to hold the utterly false belief that no L.G.B.T. people use their services and that no L.G.B.T. people live in their electoral districts. It robs policy makers of the ability to understand us and it makes evidence-based policy more difficult. It puts L.G.B.T. Americans as a group back into the closet."

- - -

Check out the Let's talk about stats (Part 1) – measuring population density in NYC.

Constantine A. Valhouli is the Director of Research for NeighborhoodX.