Questionnaire Designing for health surveys.pptx

Questionnaire designing for health surveys

1. Designing a questionnaire Dr Swati Mahajan

2. Golden Rule Don’t try to create your own questionnaire---unless you REALLY have to…..

3. What is a Questionnaire? A tool for collecting information to: Describe  - The magnitude or severity of a disease  or symptoms    Compare  - Exposure among cases/controls      Explain- Knowledge Attitude and Practices

4. Does a questionnaire give you: Subjective information Objective information

5. When should we use questionnaires? No objective measures exist to assess the variable of interest (e.g. diet, exercise, smoking history, asthma control etc) Not feasible to use objective measures (time frame of study, cost of data collection, completeness of objective data)

6. What information would we need about exposure to biomass? Type of fuel (e.g. LPG, electricity, wood, straw, dung, rubbish) Hours of exposure/day How many years in your life—when? Ventilation Type of stove Possible confounders e.g. kerosene lamps

7. General Format Self-administered (mailed or personal contact)   Interviewer administered (face to face interview)   Telephone interviews Web-based, mobile technology e.g. cell phone, Survey Monkey  

8. Questions Format Open ended questions -more information but difficult to code, enter and analyze What do you know about your COPD?   

9. Questions Format  Close ended questions less information but easy to code, enter and analyze Have you ever smoked cigarettes? Yes__ No__

10. Advantages of Questionnaires Can relatively easily and economically reach a large number of people Provide quantifiable answers Relatively easy to analyse May be less expensive

11. Common problems with questionnaires Language too advanced Questions may be leading ie suggest the answer you want Too many questions Questions may have 2 parts Poor response rate

12. Potential pitfalls of Questionnaires & Solutions Variability in administration within and between technicians: training & monitoring Do participants understand? Simple language Do participants recall the information? Does recall vary by outcome? Time frame of recall Did participant complete the entire questionnaire (missing data)? Fatigue? short Did participant deliberately under-report or over-report? Make participant feel at ease

13. Golden Rule Don’t try to reinvent the wheel!!!

14. Golden Rules about Questionnaires Try to find an existing standardized questionnaire that has been used in good papers (literature search) Find out if it’s available for you to use (do you have to get permission, pay) Get advice about translation If you can’t find one, get good help in developing your own…NOT easy!

15. Language and translation It matters NOT a word-for-word exchange NOT an exact science NOT something anyone who speaks a second language can do Best if bilingual and bicultural Back-translation (and compare with original text to check on fidelity)

16. Why you should try and use an existing questionnaire Using a previously validated and published questionnaire will save you time and money You will be able to compare your own findings with those from other studies You need only give outline details of the instrument when you write up your work You may find it easier to get published

17. What are General Considerations when Selecting Existing Measures Do a comprehensive literature search Contact researchers working in your area of interest directing to ask for information about measures (current or suggested) Evaluate measures from the standpoint of use with your population & with your selected data collection method When using existing measures be careful about translation and adaptation

18. What to look for in existing instruments Does the instrument meet your specific study goals and practical limitations? Does the instrument measure the variable you want to assess? What age group is targeted by the instrument? Are there instructions available for how to administer the instrument? How difficult is it to obtain a copy of it? How expensive is it?

19. Important Issues Who should administer a questionnaire? Can the person who administers the questionnaire change the question? Can the person who administers the questionnaire answer questions the respondent may ask? Is it important to train the interviewer?

20. Bias in Data Collection Information Bias Selection bias

21. Bias in Data Collection Information Bias = systematic differences in the measurement of a response Response bias – e.g. cases being more likely to remember than controls Observer bias – e.g. interviewer probes or prompts cases to recall events more than controls “Are you absolutely sure your parents didn’t smoke when you were a baby? Think back”

22. Bias in Data Collection Selection bias Those who respond are different from those who do not (e.g. more motivated, worried, or non-responders too sick too respond)

23. Strategies to minimize bias Observer bias Use a standardized method of measuring or collecting data Avoid administration by people who know “too much” Be non-judgmental Selection bias Use a representative sample Take steps to increase response rate

24. Practical Ways to increase response rates …… Be positive and respectful when contacting participants Explain why the study is important and how participation can contribute Give a tangible reward Make the questionnaire interesting Do not inconvenience the participant Make questions appear short and easy Avoid asking highly personal information (if it must be asked, use ‘soft’ wording and inform about confidentiality) Say thank you!!

25. Piloting is essential to all studies!!!!! Pilot with a similar group of people to your intended subjects Piloting questionnaires will allow you to: Identify problems with content Identify problems with administration Identify problems with coding Turn back before its too late! Always analyze pilot data-you may want to revise your study questions or hypotheses

26. Keep in Mind …. Are you asking only the necessary & relevant questions? Will you really use all those questions? Constraints in overall length by survey mode: In person: 15 minutes Telephone: 15 minutes Self-administered: 2 pages Scholes and Greene

27. Is the Questionnaire Feasible? Is it the right length? Does it require special training to administer? Can you afford it?

28. And don’t forget to… Include a brief covering letter or verbal introduction stating: Who you are Who you work for Why you are investigating/researching Where you obtained the respondent’s name How and where you can be contacted Absolute guarantee of confidentiality Self-addressed envelope for mailed surveys!!

29. THANK YOU