Standard Error of Measurement
The standard error of measurement (SEM) is a statistical phenomenon and is unrelated to the accuracy of scoring. All test results, including scores on tests and quizzes designed by classroom teachers, are subject to the standard error of measurement.
If a student were to take the same test repeatedly, with no change in his level of knowledge and preparation, it is possible that some of the resulting scores would be slightly higher or slightly lower than the score that precisely reflects the student's actual level of knowledge and ability. The difference between a student's actual score and his highest or lowest hypothetical score is known as the standard error of measurement.
In the example below, a student who correctly answered 30 of the 60 questions on a grade-8 science test had a scale score of 403. The standard error of measurement at this achievement level was 15 scale score points.
|Grade 8 Science||Core 1|
|Raw Score||Scale Score||SEM||Scale Score
Because of the standard error of measurement, the potential exists that a small percentage of students may score lower than anticipated on a test, given their level of knowledge and preparation. Testing experts refer to this phenomenon as a "false negative."
Conversely, the possibility exists that a small percentage of students may score higher than otherwise would have been expected. Testing experts refer to this phenomenon as a "false positive."
Virginia accounts for the standard error of measurement on Standards of Learning (SOL) tests by allowing students retakes of tests needed for credit toward graduation. Students who score within 25 points of passing an SOL test qualify for an expedited retake. Students who score within 25 points of passing SOL tests in history/social studies and science also may receive a locally-awarded verified unit of credit.