Although the Obama administration announced in June that officials would look for ways to accurately count same-sex marriages, the coding process and the software that reads the forms when they come back still reflect the former policy.
A Commerce Department aide said Census preparation takes almost 10 years and the software would not be able to be altered in time for the once-a-decade count......
As a fix, officials decided to release the raw data on married same-sex couples before that data is processed and tabulated by the software.
.....
"Later in 2011, the Census Bureau will release detailed tabulations from the 2010 Census, including counts from the relationship question," said a statement from the Commerce Department. "A footnote will indicate that there are no same-sex couples included in the husband/wife relationship category. At the same time, the Census Bureau will release counts from the relationship question, by state, that show the unedited data [that is, which do not recode same-sex couples who report themselves as husband/wife].
"The Bureau will start producing reports on the data in 2012, and according to the statement, "the Director has determined that one of these special reports will focus on the question of how same-sex couples report their relationships and what the unedited data reveal about this issue."
There was a booth at our local pride parade that encouraged us to be accurate: only if you are legally married (not DP'd or unioned), make sure you tick off the box for "married".
DPs and Unions do not appear on the form.
How can it take 10 years to build software????
As the LA TImes commented :
The Obama administration has delivered a logic-defying interpretation of the act, which prohibits the federal government from recognizing same-sex marriage: It does not stop the Census Bureau from reporting the number of married gay couples, the president's lawyers said. Yet isn't counting these marriages the purest form of recognizing them?.....
Neither the census nor the survey will differentiate between same-sex couples who are legally married and those who consider themselves so. In the 2000 census, before any states recognized gay marriage, one-third of the people in same-sex households identified their partners as spouses. But the census already takes the word of heterosexual couples as to whether they're married, so the new count will be fair. It just won't be very informative.
The 13-year-old Defense of Marriage Act has always been discriminatory, and now it is out of sync with the realities of a changing society. With same-sex marriage legalized in six states, the District of Columbia recognizing such marriages performed elsewhere and an estimated 18,000 married gay couples living in California, what's needed aren't convoluted interpretations of the federal law but a push from President Obama for Congress to repeal it.
We'll see how those DOMA cases go....
2 comments:
IT, I've never collected and reported data on this scale. I'm guessing you have had oceans more experience, and more recent to boot. But I recall in grad school (groan, DOS-based SPSS) that the reporting is hardly the most difficult part on a major survey. And tweaked a lot, to boot.
I had a neighbor who worked for U.S. Census on the business side. I never had the impression that things moved quickly, but no one has a huge set of wonderful data without working it a lot of ways.
And now that I'm thinking back to grad school, re-categorizing this way is poor form because it's known to be inaccurate. Bet the raw files will be used for demographic research for consumer goods.
Sorry I don't drop by more often!
Thanks, Lynn!
Post a Comment