developing a questionnaire. goals discuss asking the right questions in the right way as part of an...

24
Developing a Questionnaire

Post on 21-Dec-2015

215 views

Category:

Documents


2 download

TRANSCRIPT

Developing a Questionnaire

Goals Discuss asking the right

questions in the right way as part of an epidemiologic study.

Review the steps for creating a questionnaire and the categories of information typically collected.

Review types of questions and how to properly organize them into a questionnaire.

Steps in creating a questionnaire

1. Identify the leading hypotheses about the source of the problem

2. Identify the information needed to test the hypotheses

3. Identify the information needed for logistics of the study and to examine confounding

Steps in creating a questionnaire

4. Write the questions to collect this information

5. Organize the questions into questionnaire format

6. Test the questionnaire7. Revise the questionnaire8. Train interviewers to administer

the questionnaire

Categories of information for an epidemiologic questionnaire

Identifying information Demographic information Clinical information Exposure or risk factor information Source of information

Identifying information Important for the logistics of the study Includes

Respondent's name or other identifiers Contact information: address and telephone

number Allows

Subject identification Questionnaire updates as more information

becomes available Linkage of questionnaire to other records Prevention of duplicate entry of records

Demographic information Includes items such as age, sex,

education level, and location Is used to characterize the population at

risk Is important in the search of potential

confounders Needs to be evaluated to determine if it

affects the relationship between exposure and disease

Clinical information Includes

Signs and symptoms of disease Date of onset of illness Results of laboratory testing

Allows To characterize the illness To decide who has the outcome of

interest To chart the time course of the problem

Exposure or risk factor information

Is used to test the hypotheses under investigation

Is probably the major focus of the questionnaire

Should be specific to the problem under investigation

Often includes: The respondent’s exposure to the factor of

interest The route, amount, and timing of exposure

and other details of exposure (e.g., brand, distributor)

Source of the information Identifies the individual supplying

the information Helps to assess validity

Is the interviewee the study subject or a surrogate?

Identifies the interviewer Helps to correct problems

Unanswered questions, illegible handwriting, nonsensical responses, inadequately trained interviewers

Types of questions

Open-ended Fill-in-the-blank Closed-ended

Categorical responses Ordinal responses

Types of questions

Open-ended Do not provide response choices,

therefore possible responses are limitless

Are useful in characterizing attitudes, beliefs, and behaviors

Are usually limited to hypothesis-generating activities

Types of questions

Fill-in-the-blank Do not provide response choices Require short responses (one or two

words) Are used to measure a simple

respondent attribute, collect a date, or quantification

Require categorization and coding of responses

Types of questions

Closed-ended Response choices are provided

Categorical responses: categories have no particular order or inherent numerical value with respect to one another

Ordinal responses: responses describe a range of choices and have a quantitative values with respect to each other.

Types of questions

Closed-ended Investigator must

Anticipate likely responses Present the responses as a list of

mutually exclusive choices State responses clearly and concisely

Selecting the type of question

Depends on: The kind of information you need Your expectations about that

information During early exploration, when you know

little about a problem, you will be more likely to use open-ended questions

As you learn more about the problem and can anticipate responses, you will be more likely to use closed-ended questions

Writing questions

Wording depends on: Knowledge of the problem Hypothesis being tested Information being collected Affected population

Wording of questions Use language respondents can

understand Test translated questionnaires with

native speakers and “backtranslate” to test the translation

Limit each question to a single idea Word each question as precisely as

possible

Wording of questions Do not phrase questions in a way

that suggests a response Avoid double negatives Always include a “Don’t know” or

“Refused” category option In closed-ended questions, be sure

that categories cover all potential responses and are mutually exclusive

Questionnaire design

Introduction Identify the sponsoring organization Explain the purpose of the study State how long the interview is likely

to take and reassure the participant that their answers are confidential

Questionnaire design Length

As short and focused on the hypothesis being tested as possible

Try to gather additional information of interest to the investigators without compromising parsimony

Logic Organize questions in a way that promotes

rapport between respondent and interviewer Do not skip from topic to topic; the

questionnaire should appear logically organized

Questionnaire design Layout

Clearly state instructions Number questions and pages Include an identifying code for the

respondent on each page Separate responses from questions Include skip patterns

Ending statements Thank the respondent Provide contact information

Conclusions The first step in questionnaire

development is to frame and refine the hypotheses under study and decide what information is needed to test the hypotheses. Then design the questionnaire paying careful attention to the type of question used, wording and organization.

ReferenceUniversity of North Carolina at Chapel Hill School of Public Health,Department of Maternal and Child Health (2003).  Data Skills Online:  AMaternal and Child Health Toolbox.  "Designing Questionnaires."http://www.sph.unc.edu/toolbox/