Newsletters

Recruiting a wide range of people and facilitating talk about concepts introduced by both the interviewer and the interviewee allows researchers to identify cultural models that represent shared patterns of thinking. Participants were recruited by a professional marketing "rm and were selected to represent variation along the domains of ethnicity, gender, age, residential location (inner city, outer city, and regional/rural areas up to three hours from city center), educational background (as a proxy for socioeconomic status/class), political views (as self-reported during the screening process), religious involvement, and family situation (married, single, with children, without children, age of children). #e sample included 15 women and 15 men. Nineteen of the 30 participants self-identi"ed as "white," eight as "black," one as "Asian," and two as "other." Eighteen participants described their political views as "Middle of the Road," seven as "Liberal" and "ve as "Conservative." #e mean age of the sample was 47 years old, with an age range from 27 to 69. One participant was a high school graduate, 10 had completed some college, 12 were college graduates, and seven had postgraduate education. #irteen of the 30 were married, and 18 were the parent of at least one child. Although we are not concerned with the particular nuances or differences in the use of cultural models between different demographic groups at this level of the analysis (an inappropriate use of this method and its sampling frame), we plan to consider these nuances in subsequent research through other methods that are better suited to address such concerns, such as a large, nationally representative survey experiment. To analyze the cultural models interviews, FrameWorks' researchers adapted analytical techniques employed in cognitive and linguistic anthropology to examine how participants understand issues related to health and housing. 3 First, researchers identi"ed common, standardized ways of talking across the sample to reveal organizational assumptions, relationships, logical steps, and connections that were commonly made, but taken for granted, throughout an individual's talk and across the set of interviews. In short, the analysis documents patterns discerned from both what was said (how things were related, explained, and understood) as well as what was not said (assumptions and implied relationships). In many cases, analysis revealed con$icting models that people brought to bear on the same issue. In such cases, one of the con$icting ways of understanding was typically found to be dominant over the other, in the sense that it more consistently and deeply shaped people's thinking. On-the-Street Interviews Cultural models interviews were supplemented with an additional set of 36 10-minute "on-the-street" interviews conducted in Boston, Massachusetts, and Frederick, Maryland, in October 2015. #ese interviews were conducted to con"rm the results from cultural models interviews and to test a small number of reframing hypotheses. FrameWorks researchers recruited participants who were passing on the street, asking them if they would be willing to participate in a short interview as part of a research project on "social issues." #e recruiting researchers paid particular attention to capturing variation in gender, ethnicity, and age. All informants signed written consent and release forms, and interviews were video- and audio-recorded by a professional videographer. Participants in on-the-street interviews were asked several open-ended questions about housing and healthy housing to elicit cultural models. #ey were then read one of several messages, a%er which they "A House, a Tent, a Box": Mapping the Gaps Between Expert and Public Understanding of Healthy Housing 10

• Cover