Reliability: Introduction
Reliability Session Definitions & Basic Concepts of Reliability Theoretical Approaches Empirical Assessments of Reliability Interpreting Coefficients Test Refinement and Reliability
Conceptions of Reliability “He’s often late!” “My car won’t start!” S.E.M.
Components of Measurement Measured Value = True Value + Systematic Error (Bias) + Random Error The usefulness of a measure depends on the ratio of the true value to any error variance that it produces
Classical Test Theory Random error vs. systematic, or bias Classical theory assumptions: –Error independent of score –Mean of errors = 0 –Observed score = true score + error –Random errors tend to cancel out if sufficient observations made
Internal Consistency Logic: problem of change over time Alternate forms Split-half Kuder-Richardson & alpha Internal consistency Item-total correlations Number of items & reduction in error term Spearman-Brown formula # items reliability
Sources of Variance: Which to Include in Estimating Reliability? Error Patients Observers Time Measurement instrument
Reliability Subject Variability Subject variability + Measurement Error Reliability = Subject Variability Subject Var. + Observer Variability + Meas’t Error or,
Generalizability Theory Separates sources of variability: –Observer inconsistency –Between observers –Subject change over time Quantifies these Helps to show how to optimize design (and administration) of test given these performance characteristics.
Reliability versus Sensitivity of a Measurement Metaphor of the combs
Statistics to use: ICC vs. Pearson r ICC = 1.0; r = 1.0 r = 1.0; ICC < 1.0
What is the Reliability when: Every student is rated “above average” Physician A rates every BP as 5 mm Hg higher than physician B The measure is applied to a different population The observers change The patients do, in reality, improve over time