Creating a Strong Survey

By Patricia Steele
HigherEd Insight

Surveys are a great way to get information to inform your work. Perhaps you want to know what additional services your students need. Maybe you need to learn more about what employers think of your program. You may want information on program graduates and what they’ve done since exiting. Perhaps you need a before and after measure for an intervention.

Conducting a survey is one way to get such information from a group of people without talking to each and every one. In our next blog post, we’ll look at promising strategies for getting optimal results from your survey respondents.

Hot Tips and Engaging Examples

An important key to getting useful and reliable survey data is to develop a tightly focused survey with clear questions that ask about what you really want to know.

Focus. Decide what you most want to learn from your survey respondents and stick to those topics. A survey should not be used to ask a laundry list of everything it would be interesting to know. Decide what is most important and ask questions that you think are likely to yield information that will help you adjust your program to serve the needs of your stakeholders.

Focus Your Questions

Topic:  What are local employers’ perceptions of our job training program?

On-topic Question:   How many graduates of our job training program have you hired in last 12 months?

Not So On-topic Question: Have you ever personally participated in a job training program to get a job?

Keep it simple. Respondents are more likely to answer surveys that are brief, to the point, and easy to respond to. They should be visually appealing and published in an easy-to-read, good-sized font (ex. Times New Roman 11 pt). Leave some white space on your survey pages. (If you are using an online survey service such as Survey Monkey or Zoomerang, as discussed in a previous Evaluation Corner post, it will leave the white space for you).  Try to ask no more than 10 questions; fewer would be even better.

Open or closed? Surveys can ask both open-ended and closed questions.  A closed question is one where you provide respondents with answers to select from, like multiple choice or a rating scale. An open-ended question requires survey takers to type in their own answers. These questions can raise issues you hadn’t previously considered and may yield anecdotes that can be engaging to the people you plan to share the data with.

Survey respondents are more likely to answer closed questions than open-ended, and you are likely to end up with more usable data if you use mostly closed questions. Too many open-ended questions early in a survey can lead to respondents abandoning the survey. Use closed questions when you know most or all of the possible answers respondents can give. If you know most of the possible responses to a question, list them, and add an “other” category where respondents can tell you something you may have missed. Open-ended questions are useful for collecting information on topics where you want respondents to tell you about their thoughts in more depth. An open-ended question is sometimes a good way to wrap up your survey; you might ask something like, “Is there anything else about [the topic] that you would like to share?” One downside of asking open-ended questions is you have to review and organize the data, which can be time-consuming. 

Closed Question: 

Which of the following strategies do you use to communicate the cost of textbooks to students? (Select all that apply)

a. Students’ counselors discuss this with them when they select courses

b. Our program website has a section discussing textbook costs

c. Students receive an email when they enroll discussing textbook costs

d. We do not communicate with students about the cost of textbooks

e. Don’t know

f. Other (please specify):________________________________

Open-Ended Question:

Please describe your policy for communicating the costs of textbooks to current and prospective students.

Ask for some demographic data. Think about what characteristics of your respondents are important. For example, do you want to know their age, how many college credits they already have, if they are veterans, their gender, race/ethnicity, or native language?  Consider how you might use such information later when analyzing your data. It is usually best to place demographic questions at the end of your survey and to offer a “prefer not to respond” option for sensitive questions such as income or race.

Find out more.  The internet has many helpful survey-related resources that can be accessed for free. These sites provide practical, user-friendly information on many aspects of good survey design:



Tell Us More

How do you use surveys in your work?  Do you have any examples you’d like to share?



Post new comment

The content of this field is kept private and will not be shown publicly.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <img> <div> <p> <br>
  • Lines and paragraphs break automatically.