Kristi Leonard – kal@uga.edu 

Kristi is a Ph.D. student in Instructional Technology, a long-time IT professional with the College of Ed, and rather experienced at conducting experts. --MAF

Where to Begin 

• Define your research topic:  What do you want to know? 

• Define your audience:  Who can give me information about what I want to know? 

• Define how to collect your data:  What is the best way to reach your audience?   

  • Email 

  • Web Site 

  • Phone 

  • Face-to-face 

 

Developing Survey Questions 

• Based on what you want to know, what do you need to ask your participants? 

• Are you asking something that will help answer your question? 

• Be consistent with how you word your questions and with the terms you use in your questions. 

• Make sure your questions make sense.  Are they understandable?  Are you using slang or technological terms that participants may or may not understand? 

• Will the participant respond to the question or will it embarrass them and make them want to quit your survey? 

• Are you leading the participant to a desired answer? 

• Make sure you are asking only one question at a time. (no ands) 

• If you are using any short answer questions, make sure the participant has plenty of space to type their response. 

 

Ordering Your Questions 

• Start the survey with fact-based questions and end with opinion type questions. 

• Similar questions and questions with similar topics should be grouped together. 

• If using more than one Likert Scale for your questions, try to keep questions with similar Likert scales together.   

• If the first Likert scale used has the most positive response first, the rest of the Likert scales should as well. 

 

Review Your Questions 

Ask colleagues, friends and faculty to review your questions and responses for clarity and understanding. 

• Do you understand the question? 

• Is it phrased appropriately? 

• Is it grammatically correct? 

• Are there any spelling errors? 

• Are the questions consistent? 

• Are the responses appropriate? 

• Are the responses consistent? 

 

Survey Directions 

• Briefly explain the survey’s purpose to your participants.  

• Explain the confidentiality parameters of the survey. 

  • Are you going to try to keep their answers private? 

    • FYI, you can try to keep responses confidential, but you cannot guarantee it. 

  • Are you reporting summary data only? 

• Provide simple instructions for completing the survey. 

• Consider including a consent form at the beginning of your survey. 

  • e.g. Please click on "I agree to participate" to give your consent to use your data. 

 

Pilot Testing 

Pilot testing is one of the most important steps on launching a survey.  Here are the steps we use in OIT. 

1. Fill out the form several times.  The first time select all the first answers on the Likert scale and type a 1 in all short answer fields.   

2. Repeat this several times (increasing the entered number by 1 each time) until you have utilized all available responses included in your Likert scale. 

3. Retrieve your data and evaluate it to ensure that all fields and choices are saving properly. 

4. Purge your sample data. 

5. Ask a group of volunteers (similar to your final target audience) to complete the survey.  (If your target audience are teachers within a school district, then try to get volunteers from the different schools.  If they are downtown business owners, then ask them.)   

6. It is important to simulate your actual data collection process as much as possible. 

7. Ask these participants for feedback on the survey itself, the data collection process you used, and the technology they were required to use. 

   1. We often run into problems when deploying surveys in highly protected environments such as schools or medical systems due to their firewall.  Most likely, their network administrator is the only one who can resolve this problem.) 

 

Ask Yourself 

Does the sample/test information answer what your defined research question? 

1. Do you need to edit existing questions? 

2. Do you need to eliminate existing questions? 

3. Do you need to add additional questions? 

4. If editing or additions are needed, you should return to the “Review Your Questions” section and have colleagues or friends review the survey and conduct another pilot test. 

 

Soliciting Participation Process 

1. Send an explanatory email out to potential participants. 

   1. Define the purpose of the survey. 

   2. Explain when you will be distributing the survey. 

   3. Thank them for their time and future participation. 

   4. Provide them with contact information in case they have any questions? 

2. On the previously designated date, send out the “you have been invited to participate” email that includes: 

   1. The purpose of the survey. 

   2. A thank you for their time and participation. 

   3. Contact information in case they have any questions 

   4. Directions on how to access the survey 

3. When the survey is complete, you can send a follow-up email thanking them for their participation and you can share the data analysis with them. 

 

Keep in Mind 

• Keep the survey short.  If you ask too many questions, participants will drop out. 

• Don’t require questions to be answered.  If you force a participant to answer all questions without the ability to skip any, they are more likely to drop out. 

• Don’t go crazy with color and design.  Research has shown that participants are more responsive to a simple, black text on a white background environment. 

 

Where to Host Your Survey 

·         Survey Monkey - http://www.surveymonkey.com/ 

·         Zoomerang - http://info.zoomerang.com/ 

·         QuestionPro - http://www.questionpro.com/student-research/ 

o        Free student version (with caveats) or $15/month 

·         Survey Gizmo - http://www.surveygizmo.com/ 

·         Free Online Surveys - http://www.freeonlinesurvey.com/ 

 

Likert Scale Resources  

• Research Methods Knowledge Base: Likert Scaling 

  • http://www.socialresearchmethods.net/kb/scallik.php 

• Sample Likert Scales 

  • http://www.gifted.uconn.edu/siegle/research/Instrument%20Reliability%20and%20Validity/Likert.html 

• Likert Scales: Dispelling the Confusion by John S. Uebersax 

  • http://ourworld.compuserve.com/homepages/jsuebersax/likert.htm