In 1993 the Census Bureau implemented the Spanish Forms Availability Test (SFAT). The SFAT was part of a research and development program by the Census Bureau designed to assist in formulating policy and design options for the Year 2000 Census of Population and Housing. The purpose of the test was to determine the effect of the availability of mailing Spanish forms to targeted areas with high concentrations of persons who speak Spanish and who do not speak English well or at all. Specifically, the test sought to determine whether making Spanish forms available would increase response rates in target areas, the reaction of non-Hispanics living in those areas to receiving Spanish forms and the effect on item nonresponse. The test consisted of two strata divided into three panels: a control panel in which an English language form was mailed to each household; a dual forms panel that included an English and a Spanish language form and a bilingual panel in which each household received a bilingual booklet questionnaire with English and Spanish questions back-to-back.

This study reports some of the results from the test, including item non-response effects and the distribution of responses for the Spanish and English forms within and across panels. Using log linear analysis the demo graphic characteristics of persons choosing a Spanish or English language questionnaire by treatment and by language of form returned is examined.