MEMORANDUM
To: City Council
From: David J. Deutsch, City Manager
Subject: Police Survey
Date: October 27, 2004
Attached is a proposed police survey as requested by City Council at the Strategic Planning meeting on October 12th. These questions reflect staff’s understanding of the desire of Council to gauge the perceptions of residents as to how safe they feel in the Bowie community and to get some assessment as to how satisfied citizens are with current police services in the City.
The format and questions for this survey came primarily from larger community
surveys from the cities of Ft. Worth, Texas, Columbia, Missouri, and Victoria,
British Columbia. The survey has been designed without open-ended questions
to facilitate evaluation. A database would be set up to compile and evaluate
the results. We anticipate using City staff to enter the responses into this
database. This is similar to the approach used in evaluating the diversity
survey.
Staff feels that repeating the Public Administration Service (PAS) survey
from 2000 is not going to yield the data that Council is looking for regarding
community feelings. The questions on the PAS survey were primarily centered
around encounters with police (see Attachment 2). It is staff’s understanding
that Council is looking for a broader reading of the community regarding
citizen perceptions of public safety and citizen perspectives regarding the
effectiveness of current police services. Therefore, the survey has been
tailored accordingly.
Staff would also recommend not asking in this survey whether residents would be willing to support the development of a City police department at a particular cost level. Until the police analysis, which is due at the end of January, is completed, it will be very difficult to determine the benefits that would be associated with additional costs for a separate police department. Therefore, it would be extremely difficult to properly frame questions regarding costs and benefits at this juncture.
Process
Staff would also recommend doing this survey in-house. The cost for hiring a consultant would be considerably higher than costs to do this survey in-house. For example, the International City and County Management Association (ICMA) offers a survey mechanism that would cost $8,200 to implement (see Attachment 3).
Staff contacted the Schaefer Center for Public Policy at the School for Public Affairs at the University of Baltimore regarding conducting a telephone survey. A phone survey would cost approximately $14,000 to $15,000. This cost includes survey design, telephone survey, preparation of a report and making a presentation. According to the University of Baltimore representative that staff spoke with, in order to get a 95% confidence level, you would need to conduct 400 phone surveys. Once the survey is designed, it is anticipated that it would take six weeks from start to finish. Survey design time would be added on top of this time. The University of Baltimore representative did not recommend doing a mail survey because of concern about reaching the 400 survey response level. However, since we reached in excess of this amount with the PAS survey, staff feels that this 400 target number is achievable. See Attachment 4 for the types of survey services that could be offered by the University of Baltimore.
City costs to conduct this survey in-house would be considerably less than the above amounts. If we send out 2,000 mailings (which represents 10% of the 20,000 households in the City) at .37 cents per mailing, our initial mailing costs would be $740.
PAS had a 25% response rate to their survey. The City would set up a postage paid account and put a return address on the back of the survey, at a cost of .81 cents per response, to facilitate people returning the survey. Setting up a postage paid account as described above means that the City would only pay for the actual mail that is sent back and would not have to put postage on all of the return documents. If we had a similar response rate of 25%, our costs for the return mailing would be approximately $405. Costs for envelopes and paper would be approximately $100.
Therefore, total costs for postage and materials would be approximately $1,245. Additional labor costs, for extra hours to enter data as described above, are not anticipated to be greater than $500. Total in-house costs are thus anticipated to be approximately $1,745.
In 2000, PAS sent out surveys to every tenth household on a random basis by street. The City's Information Technology Department randomly selected every tenth address using data that was sorted by each address on a particular street for the PAS survey. This way we ensured a representative sample that was evenly distributed throughout the City. Because we developed the random mailing list in house for the PAS survey, we foresee no problem in developing a similar list for the proposed survey using the City’s Information Technology Department.
Another advantage to doing this in-house is that the survey can be produced and mailed in a more timely manner.
Timeline
Given the approach described above, staff would propose the following time frame for implementing this project.
1. Obtain Council approval for the approach outlined above and the attached survey at the City Council Meeting on November 1st.
2. Have Information Technology prepare randomly selected mailing list by November 8th.
3. Mail surveys to 2,000 randomly selected residents by November 19th.
4. Have Information Technology design a database to compile and analyze information by November 22nd.
5. Request that surveys be returned by December 6th.
6. Evaluate surveys and compile results by December 31st.
7. Communicate results to City Council by early January 2005.
This is a tight, but doable schedule. Also, it is considerably less expensive than hiring a survey firm.
STAFF RECOMMENDATION
For the reasons put forth in the above memorandum, staff recommends that City Council approve the attached survey for distribution. The survey is similar in format to the PAS survey in that every tenth resident of the City would be sent a survey on a random basis. We also recommend that this survey be handled in-house for a considerable cost savings over what a consultant would charge to carry out this task.
DJD/JLF