Chapter 5 : Survey Research (Constructing a Questionnaire)

Chapter 5 : Survey Research (Constructing a Questionnaire)

1. 1. Selecting Measure

   • i) Best is to use measure already shown to be reliable and valid in previous research
   • ii) If no suitable measure is found, create a questionnaire/measure but creating a reliable and valid questionnaire is not easy

2. 2. Important First Steps

   • a) Decide what information should be saught
   • b) Decide what type of questionnaire should be used (e.g. self administered?)
   • c) Write a first draft of the questionnaire
   • d) Have experts review questionnaire and then revise it based on their suggestions
   • e) Pretest the questionnaire using sample and conditions similar to the planned administration of the survey
   • f) Review results and edit the questionnaire

3. 3. Next Steps: Establish Reliability and Validity

   • a) Reliability
     • Test and retest questionnaire using sample and conditions similar to planned survey
   • b) Validity
     • i) Convergent: Administer questionnaire with measures of theoretically related constructs
     • ii) Discriminant: Administer questionnaire with measure of theoretically unrelated constructs
   • 4. Guidelines For Writing Survey Questions
     • a) Choose how participants will respond
       • i) Free-response (open ended)
         • ii) Closed-response (multiple-choice, true/false)
           • Responses are quick and easy
           • Easy to score
           • BUT may not accurately describe individual's responses cause they are restricted
         • Greater flexibility in responses
           • BUT difficult to code/grade/mark

4. 5. Write Clear and Specific Questions

   • a) Use Simple, Familiar Vocabulary (ex: few, recently)
   • b) Avoid Double-Barreled Questions (2 questions in one: do you have headache and nausea?)
   • c) Be as Short as Possible (20 or fewer words)
     • Also be aware of the length of your questionnaire (good questionnaire have +-30 items)
     • More items is good but can let participants get tired
   • d) Place Conditional Phrases at the Beginning of Sentences
     • Ex: "if you were forced to leave the job, what..." compared to "what type of work... if you forced to leave your job"
   • e) Avoid Leading Questions
     • Ex: Most people favor the use of solar energy. What do you think? (Bias may occur)
   • f) Avoid Loaded (emotion loaded that evokes emotion) Questions
     • Ex: do i look fat in this dress?
   • g) Avoid Potential Response Bias
     • Happens when respondent only use extreme points on rating scale or midpoints
     • Solution - Reverse Item (Q1 - i feel happy today, Q5 - i feel sad today) [if both if 5, participant not reading properly]
   • h) Check your Survey's Readability
     • Ask people familiar with topic to read
     • Help reduce ambiguous and confusing words and sentences

5. 6. Ordering of Questions

   • a) Self-administered Questionnaires (on the internet)
     • Most interesting questions first, at last only put demographic info
   • b) Personal and Telephone Interviews
     • Demographic questions first to warm them up
   • c) Use Funnel Questions
     • Start with general questions and move to more specific questions on a given topic when creating questionnaire
   • d) Use Filter Questions
     • i) General questions are asked to see if specific questions needs to be asked
     • ii) Direct respondents to appropriate questions
     • iii) Example
       • General: do you read book every day?
       • Specific: how many books do you read a day?