Date: Thursday, 5/16/2019 04:30 PM

Room: Civic Ballroom North Concurrent Session A

Daniela Glusberg (RSS), Alisú Schoua-Glusberg (RSS), Erika Martinez-Picazo (RSS), Kerry Levin (Westat), Jennifer McNulty (Westat).

Cognitive testing of translated questionnaires for U.S. national surveys is invaluable to uncover unforeseen linguistic and comprehension issues that may affect how language minorities respond to surveys in non-English languages. The cognitive testing process gives researchers the opportunity to discover unanticipated problems with question interpretation as well as breakdowns in the response process that stem from respondents living in cultural realities of immigrants' lives that may differ from the dominant culture. Questions that work well in English, even if perfectly translated, may not work well for non-English speakers because of distinct cultural contexts in which the monolingual users of the translated version operate.

Our presentation will explore this phenomenon with a discussion of recent findings from cognitive testing of a draft Spanish version of the Bureau of Labor Statistics’ (BLS) Household Survey of Occupational Injuries and Illnesses (HSOII). As part of BLS research evaluating the possible collection of workplace injuries and illnesses from workers, the testing focused on assessing understanding of this topic as well as respondent employment experiences surrounding the time of the illness or injury. Results uncovered significant challenges for monolingual Spanish speakers’ ability to understand and respond to questions, most often unrelated to translation accuracy. Many of the cognitive testing participants were answering about jobs that offered limited or no benefits and/or jobs that were often unreported or “under the table.” These situations impacted understanding of some key topics. We will discuss participants' understanding of what it means to be “self-employed” and familiarity with workers compensation and ideas about who is entitled to receive it.

We will also discuss the wider implications of this use of cognitive testing of translations and how findings from testing with monolingual Spanish-speakers in the US, who are often marginalized, can help uncover issues that could apply to non-Spanish speakers as well.