The 2013 survey closely resembles the previous 3 surveys in format and content which enabled NRC to compare survey results from previous years. The survey and its administration are standardized to assure high quality research methods and directly comparable results across The National Citizen Survey™ jurisdictions. City of Richmond staff did augment the survey by developing six (6) unique policy questions which enhanced the 2013 survey. Similarly to previous years, NRC mailed out surveys to 3,000 randomly selected and geographically dispersed Richmond households. Each household received three mailings with self-addressed and postage-paid envelopes and over 700 of those households also received a follow-up phone call. These methods provided a reminder for residents and gave them more than one chance to participate. In addition, City staff publicized the survey during the recent FY 13-15 budget meetings, in the City Manager’s weekly report, and on KCRT.
2013 Response Rates
A total of 408 completed surveys were obtained, providing an overall response rate of 14% with a margin error of plus or minus five percentage points. Typically, response rates obtained on citizen surveys range from 20% to 40%. This is compared to a 21% response rate in 2007, a 20% response rate in 2009, and a 17% response rate in 2011, and is the lowest response rate received thus far.
With previous surveys, response rates were lower in neighborhoods with lower socio-economic status, higher rates of non- or limited English speakers, and higher rates of rentals. In an effort to address this issue and try to increase the response rate in these neighborhoods, City Manager’s Office staff asked NRC to oversample households in central and south Richmond. For previous surveys, NRC selected households using random sampling without over sampling in any specific neighborhoods. While our attempt to increase response rates in some of our neighborhoods that have historically not responded well to the Community Survey was laudable, it was not as successful as staff would have hoped for. Staff believes that this over sampling attributed to the low response rate.
Margin of Error
The margin of error around results for the 2013 Survey (408 completed surveys) is plus or minus five percentage points. This is a measure of the precision of the results. A larger number of completed surveys gives a smaller (more precise) margin of error, while a smaller number of surveys yields a larger margin of error. With our margin of error, one may conclude that when 60% of survey respondents report that a particular service is or “good” or “excellent,” somewhere between 55-65% of all residents are likely to feel that way.
2013 Supplemental Web Survey Results (Non-Scientific Survey)
As part of its participation in The National Citizen Survey™ (The NCS), the City of Richmond conducted a mailed survey of 3,000 residents. Surveys were mailed to randomly selected households in April 2013 and data was collected through early June 2013. After the official data collection period was over and the data were reported, the City made available the web-based survey to its residents through a link on the City’s web site. Visitors to the site were able to complete the survey from June 2013 through December 2013, and 282 surveys were received. The results from the non-scientific survey are included in the 2013 Supplemental Web Survey Report.
The 2013 survey results are contained in four separate reports:
1) A comprehensive report of local results that includes the survey background, methods,
and analysis of local responses presented in tables and graphs (Attachment 1);
2) A report analyzing responses based on demographic questions relating to number of
years living in Richmond, household income, race and age (Attachment 2);
3) A report analyzing survey responses based on geographic areas disaggregated by police beats (Attachment 3); and
4) A comparison report of local results to other jurisdictions with median annual household incomes from $49,000 to $59,000
(Richmond’s median income is $54,012 per the 2006-2010 American Community Survey - Table ID: DP03, Dataset: 2010 ACS 5-
year estimates), as well as to other jurisdictions in the database (Attachment 4).