Designing the Questionnaire

Designing the Questionnaire

Designing the Questionnaire Logical Steps to Develop a Good Questionnaire Recall the research objective, the research questions and hypotheses Identify the variables to be measured Formulate questions (items in the questionnaire) Order and wording of questions and the layout of the questionnaire Test for omissions and ambiguity Correct the problems (pretest again, if necessary)

Exercise – Identify the variables

Questions must meet 4 requirements You must ask the right questions Respondents must understand the questions Respondents must know the answers Respondents must be willing to tell you those answers.

From Variables to Survey questions

Types of Questions 1. Open-response question: People look for different things in a job. What would you prefer most in a job? 2. Closed-response question: People look for different things in a job. What would you prefer most in a job? – – – – Work Work Work Work that pays well that gives a sense of accomplishment where you make most decisions by yourself that is steady with little chance of being laid off.

Open Ended Questions Advantages – Gain insight into the problem – Too many options to list – When verbatim responses are desired to give the flavor of the problem – When behavior to be measured is sensitive or disapproved – Interviewer / questionnaire structure influence can be minimal Disadvantages – Inarticulate respondents – Interviewer’s skill in recording quickly and summarizing accurately – Time consuming, subjective judgments while tabulating, adds to cost

Closed-response Questions Two Basic Formats for Closed Ended or Structured Questions Choice from a list of responses Appropriate single-choice rating on a scale

Closed-response Questions What type of fast-food restaurant do you visit most often? Burger Chicken Seafood Don’t know Mexican Pizza Chinese Other (please specify)

Closed-response Questions What is your overall satisfaction with McDonalds Hamburgers? Very satisfied satisfied Quite Satisfied Very satisfied satisfied Somewhat satisfied 7 6 Not at all 5 4 3 2 1 Not at all

Closed-response Questions Advantages – Easy to understand, quick responses possible – Micro-differences in responses can be captured – Easier tabulation and analysis – Answers are directly comparable from respondent to respondent Disadvantages – Neutral category may attract more responses than warranted – Information between categories may be lost (extreme case – dichotomous categories)

Issues in Questionnaire Design Mutually exclusive choices

Issues in Questionnaire Design Order of response categories

Issues in Questionnaire Design Meanings of response labels

Issues in Questionnaire Design Range of response categories

Issues in Questionnaire Design Respondent uncertainty – Should respondents be provided with a Don’t know or No opinion option? – When it is important to differentiate between ambivalence and ignorance, both options should be provided

Issues in Questionnaire Design Question Wording - Vocabulary – Simple, easy to understand, commonly used language – Avoid technical words and jargon (unless sample is technically qualified) – Words meaning something else in different languages and cultures (e.g. Nova meaning “no go” in Spanish; “mist stick’ meaning manure in German, etc.)

Question Wording Ambiguity

Question Wording Are any questions "double-barreled”?

Question Wording Are any questions loaded or leading? Both skew responses in the desired direction Questions which threaten respondent selfesteem e.g. occupation question produces more “executives”

Issues in Questionnaire Design Question Wording – vocabulary – “double-barreled” questions – leading or loaded questions – Instructions Complicated or lengthy instructions confuse and bias respondents

Question Wording Is the question applicable to all respondents? Why do you like fast-food? Assumes respondents like fast foods Better strategy would be to ask a filter question first.

Question Wording Question length – Should be short – Longer questions confuse and fatigue respondents Sensitive questions – Questions on information perceived to be embarrassing, like personal income, criminal activities, alcoholism, smoking, drugs habits, social desirability issues, etc. – Creativity rules (assurances of confidentiality, anonymity, slipping it in sideways, open-ended questions, asking in third person, etc.)

Sequence And Layout Decisions Opening questions – easy and non-threatening Flow – smooth and logical – avoid jumps Broad to specific Critical questions – placed in the middle Appealing and interesting Order bias – the possibility that subsequent responses are influenced by preceding responses e.g. fewer people will say that their taxes are too high after being asked whether govt. spending should be increased in certain areas. Demographic questions - last

Pretesting and Correcting Problems Pretesting Specific Questions For Meaning Task difficulty Respondent interest and attention Pretesting the overall Questionnaire Flow of the questionnaire Skip patterns Length Put yourself in the respondent’s shoes and answer the questionnaire.

Examples – spot the problems in the questions

Examples – spot the problems in the questions

Examples – spot the problems in the questions

Examples – spot the problems in the questions

Examples – spot the problems in the questions