Standard Error of Measurement and Alternate-Forms Reliability Stanford Achievement Test Series
To review the standard error of measurement, see the Guideline for Exercise 21. Additional information on its meaning and interpretation are given below in the excerpt. Standard errors of measurement are based, in part, on the reliability estimates for tests. In the excerpt shown below, alternate-forms reliability was used. This type of reliability can be obtained only when there are two alternate forms of a test (that is, the same skills are tested with two forms that contain somewhat different content). To get the reliability coefficient, one group of students takes both forms (usually at different times) and then the two sets of scores are correlated. Note that when a correlation coefficient is used for this purpose, it is called an alternate-forms reliability coefficient. See Appendix A to review the correlation coefficient before attempting this exercise.