In preparation for the 2010 Census, the Census Bureau has conducted two rounds of cognitive testing on test versions of a bilingual, Spanish/English “swimlane” questionnaire (the first round tested the 05 version, the second round tested the 07 version with differences in content). The primary objective of the first round of testing was to test the Spanish translation contained in the questionnaire. The form was tested with only Spanish speakers in Round 1 (See Caspar, et al., 2006). In the second round of testing, we conducted more traditional cognitive testing that focused on comprehension in addition to navigation and issues of item non-response. The second round of testing included monolingual English speakers, monolingual Spanish speakers and bilingual Spanish speakers with limited English proficiency and also tested changes made to the Spanish translation based on the first round of cognitive testing. 

This report presents the results of the second round of testing and focuses on issues leading to comprehension difficulty and item nonresponse. Additional navigational issues are also reported here. Recommendations for form revisions and future research are provided.