Chapter 18 Flashcards
What is a questionnaire?
A tool for systematically gathering information from study participants.
A good questionnaire is:
Carefully crafted for a specific purpose.
Questionnaire design usually works best when:
1) Starts with the identification of the general and specific content to be covered by the survey instrument.
2) Progresses to choosing the types of questions and answers for each topic to be assessed.
The questions within each section and the section themselves should be _____. The formatting of the document should be _____. The survey instrument should be _____.
In logical order; visually appealing and easy to read; pre-tested and revise as necessary.
What are the nine steps of a questionnaire design plan?
1) Identify general question categories
2) Select specific question topics
3) Choose question and answer types
4) Check wording
5) Choose order
6) Format layout
7) Pre-test
8) Revise
9) Use

The first step in designing a questionnaire is:
To list the topics that the survey instruments must cover (exposure, disease, and population areas that are the focus of the study question).
What are potential confounders?
Factors that might influence the relationships between key exposures and outcomes (the questionnaire must include them).
What are some question areas?
1) Demographics
2) Key exposures
3) Key diseases/outcomes
4) Related exposures and outcomes
The questionnaire must include questions confirming what?
That participants meet the eligibility criteria
The questionnaire must be able to accurately place participants into:
Key categories. (Cohort study may ask about exposure and disease status; Case control studies may ask to confirm case and control definitions)
A survey that is too short will _____. A survey that is too long may _____.
Miss potentially crucial information; yield a low response rate.
Decisions about the types of questions to ask must include considerations for what?
Statistical tests the researcher wants to be able to run on the collected data
What are closed-ended questions?
Questions that allow a limited number of possible answers.
Give one pro and one con for closed-ended questions.
Pro: They are easier to statistically analyze
Con: They may force respondents to select answers that do not truly express their status or opinions.
Give one pro and one con for open-ended questions (Free response).
Pro: Allow participants to explain their selections and qualify their responses, give multiple answers, and provide responses not anticipated by the researchers
Con: Take longer to ask and answer, and they may result in irrelevant answers; recoding answers into objective and meaningful categories for statistical analysis is time consuming and imprecise.
When are open-ended questions most useful?
When they are used to capture initial impressions or to clarify responses to closed-ended questions.
What are the three formats that closed-ended questions come in?
1) Date and time variables
2) Numeric variables
3) Categorical variables
What are the four formats of categorical questions?
1) Dichotomous (yes/no)
2) Dozens of possible answers
3) Ordinal (ranked)
4) Nominal (unordered)
What does anonymity do?
Protects participants and allows them to provide honest answers to sensitive questions.
Questions must be framed in a way that
Protects the respondents identity.
For numeric responses the question should state:
How specific the answers should be
For categorical questions responses:
The response categories should be listed and should include an “other“ category.