This preview shows page 1. Sign up to view the full content.
Unformatted text preview: Questionnaire Development Questionnaire
The Instrument Questionnaire design includes: Questionnaire
Writing questions Drafting Drafting and organizing a questionnaire questionnaire Constructing a Good Questionnaire Constructing Includes: Thinking Thinking through your research questions and objectives before writing questions objectives your analysis before you write questions questions yourself if each question is necessary and its contribution to your analysis Planning Planning Asking Asking Question Wording Question Question When When wording should be unambiguous writing questions consider the literacy level of the R words that everyone can understand. Avoid slang, technical language and colloquialisms UNLESS they are appropriate UNLESS to the audience to Use Use Examples of Problematic Words Examples Fair Fair – may mean “average,” “pretty good,” “favorable,” or “not bad” “favorable,” Anybody, Anyone, Anything
mean “every,” “some” or “only one” Any, May Avoid Avoid using words such as “all,” “always,” “never” “never” Questionnaire Construction Questionnaire Start Start with easy questions that all respondents can answer with little effort can questions should also be non-threatening First Don’t Don’t start with knowledge or awareness questions questions questions should be directly related to the topic as described in the introduction or advance/cover letter advance/cover First First Ordering the Questions Ordering Segment questions by topic Complete one topic before moving onto the next Complete Ask about related topics together Salient questions take precedence over less salient ones Ask recall backwards in time in chronological order (job Ask histories) histories) Ordering the Questions Ordering Use transitions when changing topics – give a sense of Use progression progression If your questionnaire contains a lot of skips, provide the If Rs with flow chart of the filter questions Rs Always consider the saliency of the topic to the Rs Always Rs before lengthening your questionnaire before Put objectionable questions at end Put demographic questions at the end Physical Format of SelfPhysical Administered Questionnaire Questionnaire Questionnaire formatting is necessary to decrease errors and increase R’s R’s motivation motivation needs must always take priority over researchers, followed by those of the interviewers and data processors interviewers Rs Format Checklist Format Number all questions sequentially Use large, clear type; don’t crowd Give questionnaire a “Title” General completion instructions included: General
“Unless otherwise specified, please circle the code number that Unless best represents your answer.” best Include “white space” – place more blank space between Include questions than between subcomponents of questions (a, b, c) b, Format Checklist cont. Format List answer categories vertically instead of horizontally Avoid double/triple banking of response choices Be consistent with direction of response categories Be consistent with placement of response categories Don’t spit questions across pages. Don’t Format Checklist cont. Format If question requires more than 1 page to complete, If restate question and response categories on next page restate Special instructions for R should be included in the Special questionnaire and next to the question questionnaire Distinguish directions/instructions from question (text Distinguish boxes) Rather than use check boxes precode the questionnaire Social Desirability
Social Desirability is a Rs’ desire to present self in a Rs’ positive light As questions became more threatening Rs are more Rs likely or under/over-report behavior likely For socially desirable behavior, ask EVER before current For socially undesirable behavior ask about current than For usual usual Remember. . . Remember.
Self-administered questionnaires reduce Self-administered question threat and improve reporting on sensitive questions sensitive Longer questions reduce sensitivity when Longer obtaining information on frequencies of socially undesirable behavior undesirable Train your interviewers well, a professional Train attitude is essential attitude Testing the Questionnaire Testing It It is preferred that you test your questionnaire with Rs similar to those Rs included in your main study population included the questionnaire in the same mode that you plan to administer the instrument in in cognitive pretesting Test Test Consider Basic Questionnaire Design Steps Basic
Decide what information is needed 2. Search for existing questions 3. If beneficial, conduct focus groups 4. Draft new questions/Revise existing ones 5. Order questions 6. Have questionnaire evaluated by peers 7. Revise and test on self/others 8. Consider conducting think-aloud interviews
1. Design Steps cont. Design
9. After think-aloud interviews revise and eliminate questions 10. If interviewers are involved in study: a. Prepare interviewer training manual b. Prepare interviewer Q by Qs (standardized instructions for each question) c. Conduct training; include mock interviews 6–20 cases d. Pretest instrument e. Debrief interviewers e. Revise/eliminate questions based on pretest R’s and interviewer comments Design Steps cont. Design
11. Pretest again, if necessary 12. Prepare final interviewer instructions 13. Conduct main study ...
View Full Document
- Summer '09