DataCenter logo
home programs research tools reports donate search
web resources internet tips campaign research research planning

Creating Surveys


STEP 2: Designing The Survey

Use the list created in Step 1 to develop questions that target what your group wants to know. Look through other surveys designed for communities or groups similar to the one you want to survey to get ideas of ways a question can be asked. It can take multiple questions to get to the answer you are looking for.

Design Tips (5)

  • Keep your language direct, use common words, if an uncommon word is used, include a definition.
  • Close-ended questions receive the highest response rate. Close-ended questions include: Yes/No, True/False, Multiple Choice
  • Providing close-ended questions instead of open-ended questions narrows the range of answers, helps avoid un-useful information and makes it easier to compare your responses later. However, you can learn a lot from open-ended questions and often times people will write in something you never anticipated, letting you learn something extremely important about the issue or community.

    for example:

Instead of asking:

Where do you live? ____________


What neighborhood do you live in? _________________

Which borough is this?

Staten Island

  • Be specific. For complex issues you may need to create multiple questions.

    for example:

Instead of asking:
You could also ask:

Do you get hurt on the job a lot?

Has your employer ever committed the following:

Hit me
Yelled at me
Pushed me
Raped me

Check the following injuries that have occurred while on the job:

Slipped and fell
Exposed to cleaning agents that made it difficult to breathe, see or hurt your skin

  • Fill-in-the-blank questions are good to use when you don't need specific answers, you don't want to limit the types of responses, or when you want information that is testimonial.
  • People often tire quickly when taking a survey. To avoid "survey boredom" it is good to mix up the types of questions you ask. For example, place true/false questions after a multiple choice question, etc.
  • Avoid questions that can be answered with "I don't know"
  • Do not combine two questions into one.
  • Stay away from questions with multiple answers. (For example: Circle all that apply) These will be more difficult to work with when you are analyzing your data.
  • Adding a "What else do you want me to know?" is a useful way to end a section or survey.

Length of survey

Try to keep the survey as short as possible—30 minutes at most for individual surveys. You will need to allow more time for interview surveys, but keep in mind that a respondent may get tired and not complete the interview.


Survey Format

Avoid long introductions. The introduction should be short and name the organization doing the survey. It should also include how the information gathered will be used and let people know it is anonymous and/or confidential.

The First Questions
The first questions will set the tone for the survey. The person should feel they have information to contribute. By making the first few questions relatively easy to answer, you may have a higher success rate of getting surveys completely filled out.

The Last Questions
Don't leave the most important questions for the end. Many surveys never get completely filled out.

Immigration Questions: Develop Trust Up Front
If you must ask questions about immigration status place them later in the survey. Use the first part of your survey to develop the trust of the respondent and then place these questions near the end of the questionnaire.

5- Developing A Downtown Survey: Basic Principles of Good Surveys

A project of the DataCenter's Economic Justice Program, May 2004.


Download "Creating Surveys" Toolkit
(18 pages requires free Adobe Acrobat®Reader)

Bajar "Creando Encuestas"
(Documento de 20 paginas de formato PDF; se necesita el programa Adobe Acrobat®Reader para ver o imprimir este documento.






Surveys to Support Campaigns

Survey Planning:

1. Why survey?
2. Time & Resources
3. Type of Survey
4. How many surveys?
5. Language
6. Safety
7. Reaching respondants

Creating Surveys:
1: Create Info List
2: Design Survey
3: Test Survey
4: Train Interviewers
5: Analyze Data

Evaluating the Survey Process

Sources & Resources

About the DataCenter

DataCenter, 1904 Franklin St., Ste. 900, Oakland, CA 94612, USA
Ph: (510) 835-4692 | Fax: (510) 835-3017 | Email:
Designed by CheneyWhite WebDesign 2001
Graphics by Rini Templeton