An Investigation of the Dependability and Standard Error of Measurement of Words Read Correctly Per Minute Using Curriculum-Based Measurement

Generalizability (G) theory was used in two studies to assess the variability in words correct per minute (wcpm) scores caused by student skill and passage variability. Study one was a small “n” study with a sample of 14 third-grade students and study 2 had a sample of 37 third-grade students. Reliability-like coefficients and the SEM based on a specific number of assessments using different combinations of passages demonstrated how manipulating probe variability could reduce measurement error. Results of the two studies showed that 69% and 81% of the variance was due to student skill, 20% and 10% of the variance was due to passage or probe variability, and 10% and 9% of the variance was due to unaccounted sources of error. Reliability-like coefficients ranged from .68 to .99 and SEMs ranged from 18 to 4 wcpm depending on the number of probes given. When passage variability was controlled, SEMs were decreased and ranged from 12 to 4 wcpm. Results indicated that wcpm scores yield high reliability-like coefficients, but also have a large SEM that can be reduced by administering multiple alternate passages. Discussion focuses on conducting research designed to identify more equivalent passages in order to reduce erroneous relative and absolute decisions