Frisbie, D. A. (2005). Measurement 101: Some fundamentals revisited. Educational Measurement: Issues and Practice 23(2), 21-28. Retrieved January 15, 2007 from http://www.blackwell-synergy.com/doi/pdf/10.1111/j.1745-3992.2005.00016.x
Frisbie argues that many measurement concepts are misunderstood and misused in the field of education. During his experience as a university professor and director of the Statewide Testing Program in Iowa, Frisbie has found that there is much confusion about the terms validity, reliability, and criterion- vs. norm-referenced tests, and he present s potential consequences of these misunderstandings. More specifically, the author maintains that inconsistencies in the use and meaning of “validity”, leads to miscommunication within and outside the field, weak validation, and incomplete directions for test administration. He also finds that confusion about the term “reliability” leads to challenges in instrument selection and in establishing technical adequacy. Frisbie discusses the differences between criterion-referenced versus norm-referenced tests and argues that using these labels to describe the test rather than the kind of score interpretation leads to problems with test selection, development, scoring, and interpretation. He also addresses additional points of confusion including test quality of a criterion-referenced context versus a norm-referenced context, standardized testing, the meaning of “on grade level”, and percentiles versus percentile ranks. Frisbie concludes with suggestions for teachers, journal editors, assessment writers, and professional development providers to help develop a greater awareness of the need to use a common language in the field.
Data Quality Assurance, Multiple Measures, Quality of Assessment Data