Data Visualization Fall 2018

A Visualization of Faulty Survey Data

D.C. 2017 Public Safety Survey

In January 2017, Washington, D.C. Mayor Muriel Bowser launched a public safety survey to gather input from D.C. residents on issues relating to public safety and interactions with the Metropolitan Police Department. This survey was made available online and in paper form at recreation and senior centers.

By early February of that last year, the Office of the Deputy Mayor for Public Safety and Justice received over 7,000 responses to the survey. Just under 4,000 responses were deemed valid and used by the Mayor's office to draw conclusions on resident's opinion of public safety in the District, qualities of a police chief and issues that would be beneficial for MPD to prioritize.

The Findings:

Taking the survey results at face value, it would seem that District residents generally have a positive opinion of the Metropolitan Police Department. When asked to rate their interactions with MPD, almost 60 percent of respondents (who provided both the ward in which they reside and their race/ethnicity) felt "very positive" or "positive" when interacting with members of the police department. In comparison to police departments nation wide, survey results show that D.C. residents think MPD is more effective than other departments across the country.

The Big Issue:

The Mayor's office recognized that there were flaws in their methods of conducting this survey. Bowser's report concluded that more data would need to be collected in order to make "precise statements" about the results. The Office of the Deputy Mayor for Public Safety and Justice acknowledged that their "initial analysis presents little statistically significant variation in responses across subpopulations."

This is evident by the decision to summerize the survey opinions by channel used to fill out the survey (i.e online, senior or recreational facility) rather than by ward or race/ethinicity. The problem with this is that respondents who filled out the survery online greatly outnumbered respondents who filled out the survey at the centers. Data in the survey also showed that online respondents were overwhelmingly Caucasion and senior center respondents were majority African American.

African Americans make up the majority of D.C.'s population at 45 percent. However, when comparing the actual District population with a race/ethnic break down of survey respondents, Caucasians make up the majority of the survey at 54 percent. That's an overrepresentation of about 18 percentage points. Asian Americans were the only subpopulation who's number of survey respondents most closely matched their percentage of the whole population.

When further breaking down the differences in survey representation between African Americans and Caucasions, in almost every ward Black residents were grossly underrepresented and White residents were overrepresented. This was discovered by finding the difference between percentage points of the race population in each ward and their percentage of total survey respondents.

For example: African Americans make up 93 and 90 percent of the population in Wards 7 and 8 respectively, but in the survery results these wards not only produced the least amount of respondents, African Americans were underrepresented in poll results by almost 19 and 12 percentage points respectively. In contrast, Caucasions only make up less than two percent of the population in Ward 7. Yet, in the survey they make up almost nine percent of the respondents from that ward. This is an overrepresentation of seven percentage points.

Conclusion:

This survey had the potential to give D.C. residents the chance to voice their worries and concerns about MPD, especially in light of many local reports about tensions between MPD and the African American community in Wards 7 and 8. But probably due to the short period of time that the survey was available and the method of filling it out, the Mayor's office missed the mark on accurate representation of opinion from many races/ethnicities within the District, especially African Americans.