What is test-retest reliability and how is it measured?
What will be an ideal response?
You estimate test-retest reliability when you administer the same test to the same (or a similar) sample on two different occasions. This is a classic way to check on the stability of a measure. This approach assumes that there is no substantial change in the construct being measured between the two occasions. In estimating test-retest reliability, the focus is on analyzing the data collection instrument as a potential source of error. The amount of time allowed between measures is critical. You know that if you measure the same thing twice, the correlation between the two observations will depend in part on how much time elapses between the two measurement occasions. The shorter the time gap, the higher the correlation; the longer the time gap, the lower the correlation, because the two observations are related over time; the closer in time you get, the more similar the factors that contribute to error. Since this correlation is the test-retest estimate of reliability, you can obtain considerably different estimates depending on the time interval. Ideally, the interval between the two observations should be long enough so that values obtained the second time around are not affected by the previous measurement (for example, the respondent may just simply remember his or her response if the time interval is too short) but not so distant that knowledge of new things over time alters the way the study participants responds to the question.
You might also like to view...
Every year, ______ people who have already had at least one heart attack have a second or subsequent attack.
A. 50,000 B. 100,000 C. 200,000 D. 500,000
Failure of a "theory of mind" may be due to decreased activation of the __________.
A. superior temporal sulcus B. visual association cortex C. extrastriate cortex D. frontal cortex
You memorize the definitions in this module and learn them completely after one hour of study
Two years later, you relearn them in 45 minutes. This difference of 15 minutes (or 25 percent retention) would be considered your __________ score. Fill in the blank with correct word
Advanced topic: One advantage of using the structural model to understand the analysis of variance (as opposed to the method which involves averaging the variances of each group) is that
A) it provides more accurate results. B) it makes analyzing the results easier when group sizes are different. C) it is more structured, and therefore more rigid. D) it is more robust to violations of assumptions of normal population distributions.