How to Measure Survey Reliability and Validity by Mark S. Litwin

Author:Mark S. Litwin [Litwin, Mark S.]
Language: eng
Format: epub
Tags: Social Science, Research, Reference
ISBN: 9781506319810
Google: qN0sCgAAQBAJ
Publisher: SAGE Publications
Published: 1995-08-03T05:18:31+00:00

Predictive validity is one of the most important ways to measure a test’s accuracy in practical applications; however, it is seldom used in longitudinal medical experiments that rely on surveys. Because the time frames are often several years long in such trials, secondary interventions may be implemented during the trial to alter the course of a disease or medical condition. If the final outcomes were compared with a test score from the start of the study, their correlation may be diminished. This would falsely decrease the measured predictive validity of the test and perhaps call into question the statistical qualities of an otherwise valid survey instrument.

Example 3.6 about SAT scores demonstrates that predictive validity (or any other psychometric statistic) may be used in various ways to support different hypotheses. You must be careful with the conclusions you draw from any of these measured psychometric properties. A good exercise is to ask peers in your area who are unfamiliar with your hypothesis to look at a summary of your data and draw conclusions. If enough people draw the same conclusion, you may be somewhat reassured that your inferences are correct. You may also ask peers to take your data and statistics and try to support a point that is opposite to your conclusions. This may open up your mind to different interpretations. It may unhinge your argument, or it may guide your approach to collecting more irrefutable evidence.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.