HLSC 2P27 Lec. 10 - Questionnaire Development

Systems thinking

Identifies underlying causes of complex problems for sustainable solutions

Potential cofounder

For example alcohol consumption may affect the relationship between smoking and liver disease

Closed-ended questions

Allow a limited number of responses and easier to statistically analyze (may not capture true opinions)

Open-ended questions

Allow unlimited responses and enable participants to explain, provide multiple answer, or express unanticipated views (take longer to answer, risk of irrelevant response, time consuming, harder to analyze)

Date and time questions

Calculate duration between events

Numeric questions

Gather numerical data

categorical questions

Questions answered by selecting the proper category

Dichotomous categorical response

Two category options (i.e yes or no)

Multiple options categorical response

Can have many category response choices

Anonymity

The condition of being unknown (important for surveys involving topics like illegal activities, or personal opinions)

Questionnare problems to avoid

Big words, undefined abbreviations, ambiguous, vagueness, double negatives, faulty assumptions, two in one, recall bias, sensitive, hypothetical, unspecific, overlapping answers, poor scale

Habituation

Error when participants become accustomed to giving the same response

Filter/contingency questions

Determines respondent eligibility for subsequent questions (i.e participants who have never used tobacco can skip questions about smoking habits)

Paper-based surveys

Provide clear, written instructions for skipping questions

Computer-based surveys

Use skip logic codes to automatically hide irrelevent questions based on filter question responses

Reliability (precision)

Consistent answers to similar questions (same outcome when an assessment is repeated)

Validity (accuracy)

Responses or measurements are correct and accurately reflect what is intended to measure

Internal consistency

Items in a survey measure different aspects of the same concept

Stability

Some surveys repeat questions or ask similar ones to confirm participant responses

Validity in surveys

Content, face, construct, convergent, discriminant, and criterion validity

Back translation

Translating material to a foreign language and then back to the original language to compare discrepancies in meaning

Independent translation

Two translators independently translate the survey and a consensus process is then used to ensure clarity and accuracy of meanings.