What does the standard error of measurement (SEM) tell us in test interpretation, and how should it be used?

• A. SEM is a measure of test validity.
• B. SEM provides a fixed, exact score for all administrations.
• C. SEM is not related to measurement error.
• D. SEM indicates the amount of error around a test score and helps form confidence intervals.

Answer: (D)

The main idea here is that the standard error of measurement tells us how much an observed test score can be expected to differ from a person’s true ability due to measurement imperfections. It is not about validity, nor does it give a fixed exact score for every administration. Instead, SEM is linked to reliability: higher reliability means a smaller SEM, so scores are more stable estimates of the true score.

You use SEM to form a confidence interval around an observed score, which helps you interpret what the score likely reflects in terms of the true score. For example, if a student earns 85 and the SEM is 3, a 95% confidence interval around the true score is roughly 85 ± 1.96×3, or about 79 to 91. This interval conveys the range in which the true score probably lies, given the measurement error.

This concept also guides decisions about change over time or cutoffs. If a change in score is smaller than the SEM, it may reflect measurement noise rather than real improvement. The SEM is often estimated from the test’s reliability and observed score variability (for instance, SEM = SD × sqrt(1 − reliability)).

In short, SEM quantifies the amount of error around a test score and provides a practical way to gauge the precision of interpretation through confidence intervals.