Community Library Survey Tutorial
Based on a document created by Doyle Marketing Communications, Inc., Rochester, NY 14607, and funded by a 2003-04 Library Services and Technology Act grant to Pioneer Library System, Canandaigua, NY
Conducting a Community Survey
A community library survey can be a valuable asset in establishing a public library district/special legislative district. A survey allows residents to provide input into the development of library services, and guides libraries in allocating resources and prioritizing services. The results of a carefully crafted community survey also offers valid and compelling arguments for approving budget proposals. The following steps will help guide you through the process.
Step 1 – Determine Survey Purpose
What is the purpose of the community survey? Perhaps you want to know how satisfied library users are with current services, why some in the community don’t use the library, and what the demand is for creating new services or improving current services. Clearly defined goals for your survey will result in data that will assist you in making important decisions about future services and programs. You must also consider from whom you need to obtain this data, if you wish to track changes in library usage or satisfaction levels over time, and how will the results be used.
Step 2 – Sample Size & Characteristics
Who will be interviewed (the target population) and how many respondents will be needed to provide a representative sample? Your sample is considered representative if it accurately reflects the population from which it has been drawn. Other factors in determining the sample size will be the size of the survey budget, the amount of time available to gather and compile the results and the importance of having a representative sample. The method of surveying your target population will also impact the response rate. If the response rate of your sample is poor, survey results will not accurately represent the general population. A higher percentage of responses will result in a more representative sample.
Sample size calculator from Creative Research Systems
Criteria that help determine a representative sample include: size of the population in the region served, age, gender, education level and income level. Also be aware that some segments of the population may be statistically significant, such library users vs. non-users. Use census data for your service area to establish the representative characteristics needed for the sample.
Step 3 – Survey Methodology
How will you survey your target population? The method used will depend on a number of factors, including the cost of distributing and collecting survey responses, the amount of time needed to carry out the survey and the effectiveness in reaching the desired representative sample. Most libraries will find that personal interview or pen-and-paper surveys are one of the best methods for getting community input. A close second choice is the mailed survey. While both methods are time-consuming and require a substantial amount of work by library staff and volunteers, they are among the least expensive ways to obtain valid input. Survey formats are explained below. Click on the link for each type of survey to see an example.
Personal Interview/Pen & Paper Survey:
This type of survey is administered face to face, where the survey-taker interviews the respondent and records his/her responses, or where the respondent is asked to fill out a paper survey. Survey locations should include sites that will ensure the participation of both users and non-users, for instance, the library itself, grocery stores, schools, community centers, malls, etc. To ensure an unbiased sample, the interviewer should ask everyone who approaches/enters the site to participate OR should establish a random selection process (for example, every 3rd person). Take into account the need to obtain data from specific sub-groups (youth, senior citizens, students, etc.) when identifying survey locations, times and days.
Pros: Conducting in-person interviews can provide immediate feedback, allowing the interviewer to probe further on certain questions. Participants will usually tolerate a longer interview than in a telephone or mail survey.
Cons: In-person interviews are typically labor intensive, requiring recruitment and training of survey-takers, scheduling of times and places, and the materials and staffing needed to administer the survey. Be aware that different locations may result in characteristics that differ from the desired target population. For instance, interviewing at different area malls may result in survey participants with different demographics, i.e, geographic location, income, age, etc.
Click here for a sample Peronal Interview (off-site) / Pen & Paper Survey
Many libraries use mailed surveys to collect data because it can be designed, printed and distributed in-house using a mailing list obtained from their local tax assessor or school district. The library may also choose to distribute the survey via an internal newsletter or local newspaper. Results can be compiled by library staff, which can reduce costs. Many surveys offer incentives to increase response rates, for example, a small gift such as a pen or keychain; a dollar bill as a thank-you, or the opportunity to be entered into a drawing for movie tickets or dinner at local restaurant. Make sure to include a cover letter explaining why you are conducting the survey, how the results will benefit the survey respondent and/or community and that you need their help. Provide a postage paid envelope in which to return the completed survey.
Pros: This is usually the least expensive survey method, and the least intrusive since respondents can fill it out at their leisure.
Cons: Mailed surveys require the greatest amount of time to coordinate. The library must design and print the survey, obtain a mailing list and provide several weeks for the survey to be distributed, filled out and returned. A much greater number of surveys must be mailed in order to get the minimum needed (depending on the expected response rate, you may need to mail 20,000 surveys to get a return of 2,000 completed surveys). Distributing the survey by newsletter or using the local newspaper makes it even more difficult because survey participants are required to take it upon themselves to mail back or drop off the survey. This method also increases the chance of getting a non-representative sample. A significant percentage of low literacy or non-English-speaking groups within your target population can also affect the response rate. Also beware that using assessment rolls will leave out non-property owners and renters.
Telephone surveys are frequently used because they can ensure a statistically valid and representative sample in a short amount of time.
Pros: Telephone interviews can be conducted quickly so the survey time period overall is significantly shorter.
Cons: Telephone surveys require a dedicated telephone bank, an accurate listing of telephone numbers, computers and software (or a supply of numbered survey copies) and personnel to survey participants and enter the data. For most libraries, this is means hiring a research firm or call center to conduct the survey instead of doing it themselves. The establishment of “do-not-call” lists also has increased the cost of telephone surveys and may limit access to the desired target population.
In this version of a written survey, participants would answer online by accessing the library website or by receiving an e-mail survey.
Pros: This is good way to attain input from an important and growing type of library user. It is fast and easy to fill out, and by using software that instantaneously compiles results, the library can track the response rate and data trends.
Cons: If you wish to reach non-users you will need to work with a research firm that does online surveying, which adds significantly to the cost. Online surveys draw only those who use the library’s website, which may not accurately represent the entire library customer base, and the community at large, or have an active email account. Those who do not use computers will not participate. Also, you need to be able to screen survey participants in order to avoid skewing the results with multiple responses from the same individual or by people who live outside your service area.
Step 4 – Develop Questionnaire
Use survey goals to determine the questions you should ask. For example, if the survey is intended to tell you what services are most important to patrons, you would ask them which services they are currently using, how often they use these services, if they are satisfied with these services, and what, if any, services they would like to see the library offer in the future. The frequency of a particular response to a question, out of the total number of responses to the question, will determine how important this specific service is to users or nonusers.
Limit the number of open-ended questions (those that allow respondents to write in their answer). Instead, try to provide yes/no or multiple choice responses that are easily compiled. Example: A total of 750 respondents out of 1,000 (75%) chose “be open more evenings; 100 respondents (10%) chose “be open more weekends” and 150 (15%) did not respond.
Make sure to test the questionnaire with several members of your target population in advance to ensure that questions are easily understood and instructions are easy to follow.
Step 5 – Compile Responses and Analyze Data
The simplest way of compiling data is to add up the number of responses to each question. Then compare that number to the total number of those answered that question or participated in the survey. Example: 72 people out of a total number of 125 respondents, or nearly 58%, said they have used the library in the past six months.
Compile demographic data to establish how statistically representative survey respondents are to the target population. Example: Out of 500 adult respondents, 270 (54%) were women and 230 (46%) were men. Census numbers for your population area show that the gender breakdown among adults is: 52% women and 48% men, so this indicates that your survey response rate is representative of the adult men and women in your target population.
Demographic data can also be used to qualify specific responses. This is done by isolating the frequency of a response by a demographic characteristic such as age, gender or income level. Example: If you separate survey responses by gender, you may discover that a higher percentage of men than women said they use the library to borrow materials. You may want to examine if the survey responses indicate that men and women use the library for different purposes. This information would be helpful in understanding the impact of various services affects the satisfaction levels among demo groups within your target population.
Step 6 – Create Report
Your report should include the following sections:
- An executive summary that showcases significant findings from the survey and overall conclusions
- A section for each question that summarizes the conclusion drawn as well as a table detailing the responses.
- A section on demographics that illustrates the characteristics of the survey respondents and how well the sample representatives the overall population. This section would also provide details on any findings pertinent to specific segment of the population.
- If appropriate, a section listing all of the written-in responses to various questions.
- A copy of the survey used, along with any other significant materials
Once the report is completed and shared internally, it is time to let the community know the results. There are several methods for sharing the results with the public.
- Schedule presentations with important constituencies such as library trustees, Friends board members, elected officials and other major supporters/donors.
- Create a press release announcing significant results for distribution to the local media.
- Post the press release and/or report on the library’s web site.
- Prepare a brief summary of the report and distribute to mail survey participants.
Things to Remember:
- Keep the survey questionnaire as simple and short as possible.
- Avoid asking about different issues in the same question (i.e., do you attend adult or children’s programs with a yes/no response).
- Place a story in the local media announcing the schedule date of the survey and encouraging people to participate.
- If questions arise after seeing the responses to a particular question or if the overall response is still unclear, consider a focus group. This is a small group consisting of members of the target population who would meet with a facilitator to discuss in more detail their reactions and feelings on the issue.
Costs can vary widely depending on a number of factors: Will the survey be produced and conducted in-house or by hiring an outside firm? Conducting a personal interview survey using staff/volunteers may reduce hard costs but would require a significant amount of staff time to coordinate and oversee, as well as volunteer time in carrying out the project. Mail surveys incur costs for the additional materials involved such as the cover letter, return envelope and postage. If you are compiling the results in-house, you will need to dedicate staff and computer resources to the project.
The size of the community does not necessarily affect the overall survey cost. A small valid sample can be representative. And while increasing the number of responses will increase the likelihood of a good representative sample, the rate of improvement does not change substantially whether you have surveyed 1,000 or 2,000.
Hiring an outside firm varies depending on your geographic area, the number of local firms capable of doing the work, the scope of the survey project and the amount of time the library has allocated for obtaining results. Hiring a consultant will require additional time for staff to create an RFP, advertise and accept bids, and conduct the selection process.
Depending on the scope of the survey project, you should expect to spend at least a minimum of 3-4 months on a personal interview or mailed survey project.
|Task Description||Time Needed|
|Establish staff team to oversee project||1 week|
|Develop RFP for Consultant Services*||1-2 weeks|
|Distribute RFP & select finalist *||4-6 weeks|
|Determine survey objectives||2-3 weeks|
|Create questionnaire and identify target population, sample size and sub-group quotas||1 month|
|Create/obtain mailing list **||3-4 weeks|
|Choose survey locations/times/dates and assign staff or recruit volunteers as survey**||4-5 weeks|
|Conduct survey||1-2 weeks for personal interviews; 6-8 weeks for mail survey|
|Compile and analyze results||2 weeks|
|Create and distribute report||2 weeks|
* Not applicable if you choose not to work with a research consultant