PSYC 3250 Chapter Notes - Chapter 7: Emor, Confidence Interval, Standard Error
Using and interpreting information about test reliability: how reliability can be used in evaluating, interpreting, and improving psychological tests, tests are used both to. To predict their standing on important criterion variables (success in school, job performance) the reliability of test scores affects the accuracy of both of these uses. 2 most important uses: estimates how much scores may fluctuate due to measurement of error, estimates effects of unreliability on correlation between tests and criteria of interest. Does not provide accuracy of scores in absolute terms. 100, but varied by 10 points due to error: test scores can vary because of measurement error (iq score of 110 could really have been. Reliability coefficient doesn"t reveal how much variability can be expected based on measurement error. Standard error of measurement (sem) is a function of 2 factors: reliability of test (rxx, variability of test scores ( x, sem = x 1 - rxx.