The 2010 Content Reinterview Survey is an evaluation of the reliability of the census data items collected in the 2010 Census. The Census Bureau has conducted a Content Reinterview Survey for every census since 1950. The purpose of a Content Reinterview Survey is to evaluate the consistency of responses to the census questionnaire, covering self-response and enumerator-response. Assessing response error to questionnaire items aids both census planners and data users. Measuring response error for specific items helps census planers improve the quality of the items through additional testing.

This report addresses the following research questions:

• How consistent was the reporting of the census data items between the Content Reinterview Survey and the 2010 Census? What percent of responses to the 2010 Census changed in the reinterview for tenure, sex, age, relationship, Hispanic origin, and race?
• How did the inconsistency in the 2010 Census compare to the inconsistency in Census 2000?

For the 2010 Census Content Reinterview Survey, a systematic sample of households was drawn from the Universe Control and Management file prior to the 2010 Census data collection. The initial sample consisted of 11,000 households in the United States and 950 households in Puerto Rico. It was estimated that 10 percent of these households would be vacant or non-existent units. After accounting for noninterviews, the final number of cases sent to the content reinterview was 8,747.

Households selected for content reinterview were contacted over the phone using computer assisted telephone interviewing from June 1 through July 31, 2010. Each household with a valid address was mailed a letter prior to being called. During the call, respondents were re-asked the data items from the census questionnaire in order to assess the reliability of the 2010 Census data. The participation rates for the reinterview were 70.9 percent for stateside cases and 71.4 percent for cases in Puerto Rico.

Gross difference rates are a measure of the percentage of responses that differ between the original interview and reinterview. Overall gross difference rates for the data items ranged from 1.0 percent for sex to 6.0 percent for race for stateside cases. The highest gross difference rate for Puerto Rico was 22.2 percent for race.