Census Data Mined for Gay Family Stats

The 2010 Census did not ask America's gay and lesbian families to identify themselves, but through some cross-referencing and detective work information on the country's same-sex families can nonetheless be inferred.

The Williams Institute at the UCLA School of Law announced in a June 14 press release that their think tank would do just that. The release also noted that same-sex families might have underreported themselves by a considerable margin because of how the Census questionnaire was worded.

Census data is used to determine which federal and state programs are funded, and at what level. By not specifically asking about gay and lesbian families, the 2010 Census failed to address the needs of taxpaying American citizens who do not fit a predetermined -- but incomplete -- profile of hetero-normative families. As a result, federal dollars that might have gone to programs benefiting gay and lesbian families are likely to be funneled elsewhere.

An anti-gay 1996 federal law, the so-called "Defense of Marriage" Act (DOMA), specifically excludes gay and lesbian families from any form of federal protection or recognition. The Census Bureau, as part of the federal government, is subject to the provisions of the act.

But researchers are able to subject the results to analysis that reveals, with some degree of accuracy, how many same-sex families there are in the country, and what their geographical distribution is. This is exactly what the Williams Institute intends to do.

"Historically, US Census Bureau data on same-sex couples have been a critical resource for informing the many national, state, and local debates about lesbian, gay, bisexual, and transgender (LGBT) rights," the Institute's press release said.

"The Williams Institute has been a leader in disseminating and analyzing these data from Census 2000 and will continue to do so with the release of data from Census 2010."

The release went on to explain, "Same-sex couples are identified in households where Person 1 describes his or her relationship with another adult of the same sex as either a 'husband/wife' or 'unmarried partner,' and clarified how America's gay and lesbian families might, unwittingly or otherwise, have underreported themselves.

"It is possible that some same-sex couples may be unwilling to identify themselves as such on the Census due to concerns about confidentiality," the release noted. "Same-sex couples may experience stigma and discrimination and consider it too risky to identify as spouses or unmarried partners on a government survey like the Census. Instead, they may choose to call themselves roommates or unrelated adults.

"A second reason for an undercount is that only couples where one partner is Person 1 can be identified on the Census since identification relies upon knowing the relationship between Person 1 and others in the household," the release continued. "For example, a younger couple residing in the home of a parent would not be identified if one of the parents was Person 1 in the household."

The problem of same-sex couples being under-counted also occurred eleven years ago, the release said.

"The Williams Institute conducted a survey of same-sex couples just after Census 2010 and found that about 15% either identified themselves as something other than spouses or unmarried partners or were in a household where neither partner was Person 1.

"Another concern about the accuracy of same-sex couple data involves the possibility that a small portion of different-sex couples miscode the sex of a spouse or unmarried partner and are incorrectly counted as a same-sex couple," added the release. "Since there are more than a hundred different-sex couples for every same-sex couple, even a very small amount of errors among different-sex couples result in relatively large numbers of misidentified same-sex couples."

In all, the release added, "between 15 and 20 percent of identified same-sex couples may, in fact, be miscoded different-sex couples. Given that most different-sex couples are married, the bulk of these miscodes likely occur among same-sex couples identified as spouses."

Because of changes from decade to decade, it's not possible to generate reliable data about trends among gay and lesbian families using Census data, the release noted.

"In 1990, same-sex couples who identified a partner as a spouse were not identified as same-sex couples at all," said the release. "In 2000 and 2010, the Census Bureau did count same-sex spouses among same-sex couples but the procedures used to process same-sex spouse couples differed. As a result, it is not appropriate to compare the same-sex couple tabulations across the three decennial Census years."

The Institute has scheduled a release over time of Census data showing, on a state-by-state basis, where gay and lesbian families are, and what their concentration is. The first release is set for this month.

"Data from several states are distributed each week," the release said. "These data will include official tabulations of same-sex male and female couples and information about the presence of children in their homes at a variety of geographic levels (e.g., state, county, and Census tract)." The release also said that all same-sex families would initially be tabulated as "unmarried partners," regardless of whether they live in the five states where marriage is legal and have availed themselves of the opportunity to wed. Currently, thanks to DOMA, even where marriage equality is legal, it is limited to state-level rights, protections, and recognition. But a separate tally later will follow, looking at the question of married same-sex families.

"Later in the year, the Census Bureau will release tabulations on its American FactFinder website and a technical note about same-sex couples that will include national and state counts of the number of same-sex couples who designated themselves as spouses versus unmarried partners," the release said.

"The Williams Institute will publish two-page research briefs on each state as the SF-1 data are released," added the release. "These briefs will highlight the geographic distribution of same-sex couples across the state, rankings of cities and counties, and information about the number of male and female couples and the number of couples raising children."

Local and national GLBT equality advocacy organizations will be able to reference the data supplied by the Institute in arguments relating to protecting the rights and legal status of GLBT individuals and their families.