10.7: Surveys

Last updated
Save as PDF

Page ID: 65865

\( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \) \( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)\(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\) \(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\)\(\newcommand{\AA}{\unicode[.8,0]{x212B}}\)

Other than the fact that they both involve asking people questions, interviews and surveys are quite different data collection methods. Creating a survey may seem easy at first, but developing a quality survey can be quite challenging. When conducting a survey, you need to focus on the following areas: survey creation, survey testing, survey sampling, and distributing your survey.

Survey Creation: Length and Types of Questions

One of the keys to creating a successful survey is to keep your survey short and focused. Participants are unlikely to fill out a survey that is lengthy, and you’ll have a more difficult time during your analysis if your survey contains too many questions. In most cases, you want your survey to be something that can be filled out within a few minutes. The target length of the survey also depends on how you will distribute the survey. If you are giving your survey to other students in your dorm or classes, they will have more time to complete the survey. Therefore, five to ten minutes to complete the survey is reasonable. If you are asking students as they are walking to class to fill out your survey, keep it limited to several questions that can be answered in thirty seconds or less. Derek’s survey took about ten minutes and asked students to describe what they ate for a day, along with some demographic information like class level and gender

Use closed questions to your advantage when creating your survey. A closed question is any set of questions that gives a limited amount of choices (yes/no, a 1–5 scale, choose the statement that best describes you). When creating closed questions, be sure that you are accounting for all reasonable answers in your question creation. For example, asking someone “Do you believe you eat healthy?” and providing them only “yes” and “no” options means that a “neutral” or “undecided” option does not exist, even though the survey respondent may not feel strongly either way. Therefore, on closed questions you may find it helpful to include an “other” category where participants can fill in an answer. It is also a good idea to have a few open-ended questions where participants can elaborate on certain points or earlier responses. However, open-ended questions take much longer to fill out than closed questions.

Survey Creation: Testing Your Survey

To make sure your survey is an appropriate length and that your questions are clear, you can “pilot test” your survey. Prior to administering your survey on a larger scale, ask several classmates or friends to fill it out and give you feedback on the survey. Keep track of how long the survey takes to complete. Ask them if the questions are clear and make sense. Look at their answers to see if the answers match what you wanted to learn. You can revise your survey questions and the length of your survey as necessary.

Sampling and Access to Survey Populations

“Sampling” is a term used within survey research to describe the subset of people that are included in your study. Derek’s first research question was: “Are students’ eating habits at Purdue University healthy or unhealthy?” Because it was impossible for Derek to survey all 38,000 students on Purdue’s campus, he had to choose a representative sample of students. Derek chose to survey students who lived in the dorms because of the wide variety of student class levels and majors in the dorms and his easy access to this group. By making this choice, however, he did not account for commuter students, graduate students, or those who live off campus. As Derek’s case demonstrates, it is very challenging to get a truly representative sample.

Part of the reason that sampling is a challenge is that you may find difficulty in finding enough people to take your survey. In thinking about how get people to take your survey, consider both your everyday surroundings and also technological solutions. Derek had access to many students in the dorms, but he also considered surveying students in his classes in order to reach as many people as possible. Another possibility is to conduct an online survey. Online surveys greatly increase your access to different kinds of people from across the globe, but may decrease your chances of having a high survey response rate. An email or private message survey request is more likely to be ignored due to the impersonal quality and high volume of emails most people receive.