In the American Community Survey, same-sex couples living together may describe their relationship as either "husband or wife" or "unmarried partner." Some of these couples report their relationship differently from others, despite living in the same objective conditions. Past literature finds that how same-sex couples define and report their relationship depends on both legal and sociological factors; as a result, far more same-sex couples report their relationship as "husband or wife" than are legally married. Using 2009 ACS 1-year internal data and logistic regression we test three hypotheses: same-sex couples are more likely to report as "husband or wife" (as opposed to "unmarried partner") when they (1) reside in a state recognizing same-sex marriage, (2) live with their own children, and (3) respond to the initial mailout form.
WORKING PAPER | OCTOBER 26, 2011
Evaluation of ACS Mail Materials and Strategy During the 2010 Census
Describes results of testing messaging and design changes to the ACS mailing materials to minimize impact of 2010 Census activities on ACS mail response rates.
WORKING PAPER | NOVEMBER 02, 2011
Evaluation of Multiple Responses
Assesses respondents' multiple responses to questions requiring a single response and if ACS should continue to blank multiple responses in the mail mode.
WORKING PAPER | JANUARY 17, 2012
Respondent Characteristics: Mandatory and Voluntary Response Methods
Examines how the distribution of respondent characteristics differ between the two methods and whether any difference is reflected in the final estimates.