Journal of Educational Measurement · 1969 · 16 citations · 0 references
EducationEducational AssessmentEducational MeasurementPsychometricsScale DevelopmentStandardized TestingAssessment ValidityInternal ConsistencyMeasurement PrecisionRating Scale DevelopmentScoring RubricsStandards-based AssessmentGrading RubricsNumeric Grade ScalesPsychometric ReliabilityPsychometric ScalingRating Scale AnalysisRating Scale ConstructionScale PrecisionScale Validity
The use of broad categories in grading is likely to increase the relative amount of error present in the measures on which the grades are based. This is true regardless of the degree of inaccuracy of those measures. An example to illustrate this generalization and an explanation of its basis in psychometric theory are presented.