Standard Error of Measurement (SEM)
A statistic estimating how much an individual's observed test score is likely to vary from their "true" score due to measurement error. Expressed in the same units as the test score.
Standard Error of Measurement (SEM) gives a concrete number to test imprecision. If a Big Five test reports SEM = 5 percentile points, then a person's observed score of 60 corresponds to a 68% confidence interval of 55-65 and a 95% confidence interval of roughly 50-70.
SEM is mathematically linked to reliability: SEM = SD × √(1 − reliability). A test with reliability 0.90 has SEM about one-third of its standard deviation; a test with reliability 0.70 has SEM about 55% of its standard deviation.
Clinical and educational testing standards (APA 2014; ITC 2013) require SEM disclosure for any test used in consequential decisions. It is why score ranges or confidence bands — not single point estimates — should accompany high-stakes reporting.
Source: APA Standards for Educational and Psychological Testing (2014).