Methodological Notes:
- Survey Population and Sample. The respondents whose views we are measuring—the survey population—are the adult residents of condominium and rental apartment buildings in Downtown East and Downtown West. We do not know the exact number of people in these buildings or their demographic characteristics and therefore cannot calculate the differential response rates of demographic groups within the survey population or their precise levels of representation in the sample. Because the downtown residential population has been growing rapidly in recent years, there are no definitive demographic reports available. The best available census data suggests that the downtown population (including children) is 62.5% white and 37.5% people of color (POC). In our sample, 86% of respondents identified as white and 14% as POC. To put this distribution in perspective, the City’s Transformation of Community Safety survey has reported that “BIPOC residents were underrepresented and white respondents were overrepresented” in its survey. This statement would appear to be true for our sample as well. One important difference between the two surveys, however, is that the City has not released any specific details of its survey results for individual questions or the demographic characteristics of its sample, while our survey has released all of this information.
- Measurement and Analysis of Race/Ethnicity. We used a standard survey question that asks respondents “What is your race or ethnicity” and offers nine response categories. Demographic questions are typically optional, and some people prefer not to answer them; in our sample, 1,055 people (out of 1,129) answered this question, including 904 white and 151 POC respondents. In this analysis we examine the relationship between race/ethnicity and public safety views in two ways: by comparing the views of people of color (POC) with those of white respondents or the full sample, and by providing results for specific race/ethnicity categories where relevant and statistically feasible. Three categories (Middle Eastern/NorthAfrican, Native American/Alaska Native and Native Hawaiian/Pacific Island) have too few people (a total of 7 in the three categories) to express results in categorical percentages, although these respondents are included in POC calculations. The other five POC categories range from 20 to 40 respondents (average=29). Percentages calculated for these specific groups suggest shared viewpoints within the groups, but we do not make categorical generalizations based on these numbers. For narrative efficiency, we use abbreviated category names in the analyses; the actual response choices listed on the questionnaire are: Asian; Black or African American; Hispanic or Latino; Multiracial or Multiethnic, and Another race or ethnicity.