The 2010 Census was the first decennial census to include a bilingual English / Spanish questionnaire. The goal behind the 2010 bilingual stratification was to reach the maximum number of housing units that may require Spanish language assistance to complete their 2010 Census questionnaire. We used American Community Survey data from 2005 through 2007 to identify the specific areas that received a 2010 Census bilingual questionnaire (“data-identified”). The criteria used for the stratification was that at least 20 percent of the occupied housing units in a tract were in need of “Spanish Assistance”, i.e., households in which at least one adult (age 15 or older) in the household speaks Spanish and does not speak English "very well." In addition, eight of the twelve Census Regional Offices provided input (“local knowledge”) and some additional tracts were identified to receive bilingual questionnaires. (Due to operational and processing requirements, and the need to accommodate field assignment area delineations, the selected tracts were then linked to the appropriate collection blocks.) All housing units in these blocks were selected to receive a bilingual English/Spanish initial questionnaire. This assessment focuses primarily on the results for the initial questionnaire returns in the bilingual delivery area, since the replacement questionnaire for these areas was English-only.

The 2010 Census bilingual questionnaire is a booklet containing questions and response options in both English and Spanish in a side-by-side, “swim-lane” format. To make use of the two additional pages that result from the design of the booklet, the bilingual questionnaire collected full information for Persons 1 through 6 as well as Persons 7 and 8, with an extended roster for Persons 9 through 1 2. This is a different format from the English-only questionnaire, which is a tri-fold sheet and collected full information for Persons 1 through 6 and used an extended roster for Persons 7 through 12.

The Census Bureau tested a bilingual questionnaire in the 2005 National Census Test. Results from the 2005 National Census Test showed that the bilingual questionnaire significantly increased the self-response rate nationally and, more specifically, in areas where there was a high concentration of non-White or Hispanic populations.