8 Face Validity

Conducting Survey Research
Version 2.0 March 31, 1999

ADDITIONAL COPIES & COPYING PERMISSION Additional copies of this resource are available free of charge in Ontario only. Please direct requests to [phone] (phone) or [e-mail]. This workbook is also available on our website at http://www.thcu.ca. This workbook was developed as a resource guide to accompany a one-day seminar. The content is structured around the logical steps of implementing a focus group. The material is relevant and practical but not comprehensive. We encourage users of this workbook to supplement the information contained in this workbook with the more comprehensive information available from the books and articles provided in the resource list. The
The extent to which your questionnaire is measuring what it appears to be measuring 8 Content Validity. The extent to which items on the questionnaire are representative of the domain under study. 8 Construct Validity. The extent to which an instrument measures the construct or trait under study. Regardless of the trait under study you can identify some theoretical constructs about that trait which your questionnaire should be able to measure and you can test by administering your questionnaire in situations where you know those constructs to be true. 8 Criterion Validity. The extent to which the questionnaire is measuring similar to a ‘gold’ standard, another measure that has been used and accepted in the field. There are two types concurrent and predictive validity. 4 Internal Reliability. Questions measuring the same construct are correlated to each other and not to other constructs. 4 Test-retest Reliability. If you were to do the survey exactly the same way, under the same conditions you would get the same results. 4 Responsiveness. The questions can detect
Testing your questionnaire for validity and reliability can be cost and time prohibited. It is important to test your questionnaire as much as possible. At the very least you should be able to test your questionnaire for face and content validity. 4 If your questionnaire is going to be used repeatedly it would be worthwhile to spend the resources to test your questionnaire thoroughly. 4 For more comprehensive and detailed information ‘Health Measurement Scales: A Practical Guide to their development and use’ 2nd Edition David L. Streiner and G.R. Norman. (1995) is one of the best books available for questionnaire development and testing. OTHER TIPS AND GUIDELINES 4 Use language of target group. 4 Only ask “need to knows.” 4 Involve target group/stakeholder in design. 4 Avoid double barrel questions. 4 Keep it simple. 4 Consulting the literature related to the topic or other specialists in the field can be helpful in constructing the