Designing the Questionnaire

In order to have data from your survey to analyze, you must have respondents complete and submit the survey. A foolproof way to achieve a high response rate to your survey is to keep it short and simple. Only ask questions whose answers are important to your research. Knowing how you plan to use the data collected will help you determine which questions to include.

Attempt to make a personal connection between you and the respondent. Introduce yourself and state the importance of the survey and the data you will be collecting and analyzing. If the person will benefit from your research, tell them how. This information may compel the respondent to take whatever time is needed to complete the survey.

Two other techniques used to increase the likelihood of respondents to complete surveys are to put the easiest or least controversial questions at the beginning of the survey and arrange questions of the same subject matter together. Within the subject areas, group questions of the same type together.

Use clear, specific wording when creating your questions. Don't ask leading questions or make assumptions. You also don't want to ask questions that require long answers.

Spend time considering the answers you provide. Several tips are:

• Make sure you have anticipated every appropriate answer
• Make categorical answers clear and logical
• Offer a "Don't Know", "Not Applicable", "None", or "Other" answer where appropriate
• Answers should be provided in a consistent order throughout the survey, i.e., if Yes is listed before No it should always be that way
• Include a comment section where respondents can personalize and/or explain their answers.
  

Assure the respondent of the privacy measures you're taking. Who will see his or her responses? Will their responses be used in data analysis and not individually reported? Reassure them of the data transfer security measures you're using.

Types of Questions

The most popular types of questions researchers use on surveys are multiple choice, ratings, scales (agree/disagree, excellent/poor) and short answer questions allowing either numeric or text answers. Some examples are:

Multiple choice:

Which electronic mail program do you primarily use?

Netscape
Eudora Pro
Outlook Express
Microsoft Outlook
Web based e-mail (for example, Hotmail, Yahoo, or webmail.utexas.edu)
Other (Please specify:

from UT Connect Subscriber Survey

Rating Scales:

Please rate the following components of the UT Connect CD in terms of their importance and value to you?

1 - Not important or valuable

5 - Very important and valuable

Configuring my computer to connect to the Internet	1	2	3	4	5
Anti-virus software	1	2	3	4	5
Web browser	1	2	3	4	5
Eudora Pro	1	2	3	4	5

from UT Connect Subscriber Survey

Agreement Scales: (agree-disagree or always/never scale - include "No Opinion" option)

Are you satisfied with the UT Connect CD ROM?
1 - Very Dissatisfied                    5 - Very satisfied

1       2       3        4       5       No Opinion

from UT Connect Subscriber Survey

Numeric open end:

What is your age?

from UT Web Central Survey

Text open end:

If you could do one or more things to improve UT Connect or the process of connecting to and using the Internet here at UT Austin, what would it be?
          

from UT Connect Subscriber Survey