1

Establish a Working Committee

2

Determine Needs

3

Identify Stakeholders

4

Identify Population to be Surveyed

5

Develop a Questionnaire

6

Collect Data

7

Analyze the Data

8

Report the Results

9

Next Steps

Report the Results

Now that you have collected information and drawn some conclusions, let people know. There are many ways to share this information with stakeholders. In addition to widely distributing a formal written report, explore venues and formats to share results at town meetings or at meetings of the various stakeholder groups involved in the survey. Explore presentation options, such as PowerPoints, social media pages and events, videos, flyers, factsheets, a listserv or email list, and news releases for local media outlets and/or community organization newsletters.

Organization of the Written Report:

  1. Executive Summary — includes a description of the population surveyed, the response rate and the results related to key issues, thanks to the contributors, and who to contact for more information.
  2. Introduction — describes the purpose or objective of the survey process, purpose of the report, and who was involved.
  3. Resource SectionSecondary data on the issue or the region in order to justify the project/survey.
  4. Project Section — description of why and how the questionnaire was designed, how the data were collected, the sample size, the response rate, and any limitations to interpretations.
  5. A description and interpretation of the results, and the most relevant or significant findings.
  6. Recommendations for committees to consider, issue by issue. A call to action on the part of local government or the sponsoring organization.
  7. Acknowledgments and recognition of persons and organizations which helped in the survey effort. Include the volunteers, individuals and organizations that contributed financially or logistically.
  8. Include a copy of the questionnaire, cover letter, and any other important documents referenced in the report.

Final Report Template

Develop a Questionnaire

Select a questionnaire from the Questionnaire Template page

Determine Needs

The working committee will identify issues to explore via the survey. While it is impossible to assess all relevant issues, it is important that the most important few issues are identified and agreed upon by the working committee. It is critical that community or organization members from a wide range of interest groups, income levels, and backgrounds are involved in choosing the issues to address.

As you identify the needs, use the following resources:

After looking at your county/community profiles, collecting information, and brainstorming issues, list your top issues below, with a brief description.

Issues Brief Description

There are sample questionnaires that cover various community issues, as well. See Questionnaires.

Identify Stakeholders

A stakeholder is anyone affected by the issues addressed in the survey. For many development issues and community projects, every resident is a stakeholder. For other issues, a few specific segments of the population are stakeholders. Stakeholders should be included in planning the survey process, developing the questionnaire, and disseminating the results. For example, if your survey includes questions related to education in the community, the planning group should include school administrators, teachers, parents, students, employers, and anyone else who cares about educational outcomes. This kind of buy-in will assure that results are shared with the appropriate interest groups, and that results are analyzed with a realistic mindset.

In addition to surveying a random sample of the general population, you may also sample a particular group to gather adequate information (retirees, elected officials, business owners, families with young children, etc.).

List the issues chosen for the assessment (from your decision matrix), and, under “Stakeholders,” list individuals who have a stake in these issues, who have a role in implementing decisions or policies, who might make decisions regarding the issues, and anyone else who may be affected by the outcomes.

Issues Stakeholders

For each stakeholder group, decide if (1) the questionnaire should include questions that identify them, to sort out their survey responses, (2) if they should be sampled in addition to the generalsample, and/or (3) if they should be consulted about items for the questionnaire.

Conducting a survey is one way to collect information from residents of your community or members of your organization. Through a survey, less vocal community members can share their ideas, opinions, preferences, and concerns. It is a way to gain a broad range of perspectives in the community, increase government transparency, and gather information and perspectives.

The survey process begins by establishing a working committee to provide direction and share responsibility for the project. Then follow the guide to determine needs, identify stakeholders, identify survey respondents, develop a questionnaire, collect data, analyze the data, report the results, and identify the next steps.

Our guide, in the form of a wheel, will guide you through this process. Feel free to create an account, so your online work is saved, and so you can log out and log in to continue where you left off.

Establish a Working Committee

The first step is to establish a working committee to guide the process. Broad representation of the community will enhance the credibility of the process, and will contribute to a comprehensive survey questionnaire. In the planning phase, don’t leave anyone out. If you do, you may hear from them later when they criticize the process or the outcomes. The committee should include a broad spectrum of perspectives, including:

  1. People who can represent the geographic area and population to be surveyed;
  2. People from organizations which may act on the results of the survey;
  3. Groups which can contribute resources needed in conducting the survey;
  4. Media professionals who can help disseminate information; and
  5. Service and social organizations to provide volunteers.

If stakeholders suspect this committee is controlling the direction of the project by pushing its own agenda, the study will have little credibility. That is why all interest groups, income levels, and backgrounds should be represented.

Try completing the following table:

  1. Brainstorm all of the groups or segments of the population make a list.
  2. For each group on your list, complete the following information.
Group or Organization Point of Contact Phone/Email Recruiter

Identify Population to be Surveyed

Population

What is the population from which you will choose the sample? Is this questionaire intended to obtain feedback from a special population -- high school students, elderly people, businesses, farmers, members of an organization? Or is this questionnaire intended for all the members of the community? How is the community defined? Does it only include residents who live "in town," or does it include everyone in a particular zip code? Clearly define the population and intended recipients of your survey. From this population, you will either draw a random sample to survey, or allow everyone the opportunity to respond.

Sample Size

How many people or households should we contact? First, decide how to select respondents:

  • Random sample from target population— from the target population, randomly select persons or households to participate in the survey. See the "magic number" for guidance.
  • Open participation— allow anyone in the target population to respond to the survey.

Consider the opportunities and limitations of each:

Respondent Selection Opportunities (Pros) Limitations (Cons)

Random Sample: From the target population, randomly select persons or households to participate in the survey.

Information from a randomly selected group can be generalized to the population.

Usually, fewer questionnaires need to be processed; fewer data to input.

May not capture information from small segments of the population.

May be difficult to randomly select individuals or households to participate.

Anyone who did not receive the questionnaire may feel left out, and may question the credibility of the results.

OR

Open Participation: Allow anyone in the target population to complete and return the questionnaire.

Allowing everyone to participate eliminates questions and suspicion about who was invited to provide information.

Voices are heard that may not have been picked up by a random sample.

Eliminates the challenge of randomly selecting the sample from the population.

Cannot generalize to the population, unless you have 100% participation.

May result in many repsonses to input and analyze.

May not gather all the information or perspectives.


Then, determine the sample size. The following factors influence sample size decisions:

  • Size of population
  • Number of sub groups within the population
  • Level of accuracy you seek
  • Cost
  • Time

Read additional information on how to determine sample size.

Target Survey Respondents

To determine sample size, you should know (1) the population of the general sample, (2) the sample who carries the attribute you seek (youth, seniors,etc.), and (3) the variability of the attribute.

(3a) If there is a normal distribution of the attribute across the population of your community, then you would use your community’s population to calculate your sample size. For example, if everyone in your community has access to festivals and events, you can survey the entire population, but sort users from non-users with an opening question. In this case, you would only use a general sample.

(3b) If you have a variable distribution of the attribute, the sample size should be larger – at least 50% of the population. For example, to assess the use of a public facility, such as a swimming pool, you can oversample the general population to capture an adequate number of respondents, or simply target users at the facility. If you want to know about why some residents do not use the facility, select your sample from the population that has access and ask a question or series of questions about why they do or do not use that facility.

(3c) If you have an unknown population (you do not know who carries the attribute), aim to sample the entire population.

If you have a population of 200 or less, sample the entire population.

Response Rate

A low response rate may not give you accurate information about the population. Low response rate from a large sample is as problematic as a low response rate from a small sample. To improve the response rate, send another request to the original sample to complete the survey. Use post card reminders, letters and phone calls to encourage responses. Send another questionnaire. The cost of reminder notes and letters must be incorporated into the cost of your survey from the beginning.

The number of completed surveys you will need for a 95% confidence level with a +/-5% sampling error is:

Population Size Returned Responses Needed
500 345
1,000 525
3,000 810
5,000 910
10,000 1,000
100,000+ 1,100

Distribution Plan

Now that you know your population, target population, samples, and response rate, you must decide your how you will distribute the questionnaire. Consider your time constraints, deadlines, how many people you have available to work on this project, how much money you have allocated for the project, and other criteria. Then, consider the following delivery options:

  • Telephone Questionnaire
  • Mailed Questionnaire
  • Face-to-face Interviews (door-to-door or intercept)
  • Drop off
  • Online Questionnaires

Review the following table to consider the criteria and delivery options:

Criteria Mailed Questionnaire Telephone Questionnaire Online Questionnaire Drop-off Questionnaire In-Person Questionnaire (door to door; intercept)
Convenience      
Ability to cover a random sample  
Cost-effective      
Ease of Implementation        
Time Constraint    
Literacy and Language      
Response rates      
Control over who responds      

As you develop your distribution plan, use the following budget template to assess costs:

Survey Budget

Collect Data

Questionnaire Distribution and Response Rate – However you distribute the questionnaire, it is important to have a good collection strategy. Whether you have a recurring email or any easy way to the mailed or hand-delivered questionnaire, more contact with respondents improves the response rate.

Data Collection and Tabulation – Most surveys require the construction of a database / data set. Responses are either entered by hand or directly into a data file (if administered online). Spreadsheets, databases, or other software can be used to tabulate the survey data, and the data entered should be checked for data entry errors. Some online survey programs automatically collect, compile, sort, and create reports with the data. To assure quality of your data set, compare the composition of your respondents to the entire population –in other words, check for subgroups (with certain demographics or other attributes) and see if it is representative of your target population. For example, if you have sent the questionnaire to everyone in your community, your responses should mirror the demographic of your community.

Analyze the Data

Make a table with number of responses, the totals, and percentages. See the examples below:

Q1: Do you have a pet?
Response # Responses % Total Responses
Yes 37 41%
No 54 59%
TOTAL 91 100%
How satisfied are you with your services?
Response # Responses % Total Responses
1 – Very unsatisfied 25 25%
2 – Unsatisfied 35 35%
3 – Neutral 15 15%
4 – Satisfied 10 10%
5 – Very satisfied 15 15%
TOTAL 100 100%

Analyze subgroups – Identify subgroups among respondents, whether based on demographic groups (gender, age, race), or based on an attribute (the subgroup of the set that answered “yes” to a specific question). Comparing the responses of a subgroup to the larger set, or comparing two subgroups may be useful to you. For example, if you are interested in why some people are not using the swimming pool, sort them with an initial question, and then delve into responses to questions about what they seek in recreational activities v. what the swimming pool provides.

Compare results to a benchmark or goal, and report results in relation to that benchmark. For example, you may have a certain level of satisfaction that you are hoping to gauge on a public service. You may want to assess attitudes to influence decision-making or to report on the success of a project.

Survey Tool User Feedback