Problems Associated with the Use of
GRADE EQUIVALENCY SCORES

The practice of using grade equivalency scores to identify learning disabled children in educational reports and IEP’s is wide spread and misleading. Grade equivalents are not only confusing, they are also statistically invalid and, for the most part, meaningless.

Grade equivalency scores do not represent equal units. For example, the difference between second and third grade equivalency scores may not be the same as the difference between eleventh and twelfth grade equivalency scores.

Grade equivalents are obtained by means of interpolation and extrapolation and consequently do not reflect the scores actually obtained by children.

When grade equivalency scores of two or more kinds are compared, misleading conclusions may result. For example, it is not necessarily the case that a child who obtains a fifth grade equivalent on a mathematics test measuring addition and subtraction can also do fifth grade mathematics requiring knowledge of multiplication and division.

Grade placement scores resulting from tests produced by different publishers are likely to give conflicting results. Grade placement scores on one test may extend up to 4.5 grade equivalents as compared with only 2.5 grade equivalents on another test in the same coordinated achievement battery.

Grade placement scores are so confusing that a lower score on one test may indicate a relatively higher performance than on another test. Because of the difference in the size of standard deviations, this could easily happen:

Example: a grade placement score of 8.5 in reading may be equal to a percentile rank of 60, but a grade placement score of 8.2 in arithmetic may be equal to a percentile rank of 98.

Especially for higher elementary grades and beyond, grade placement scores cannot meaningfully be compared from test to test, even within the same battery.

The normative data for most tests are usually collected at one point every year. How, then, are grade equivalents obtained for every month? They are extrapolated at the upper and lower ends of the growth curve. This estimation produces scores that are systematically too low in the Fall and too high in the Spring. Problems associated with this practice are:

• A high probability of over-identifying learning disabled children (increased false positives) exists if screening is conducted in the Fall. 
  
  

• A high probability of under-identifying learning disabled children (increased false negatives) occurs if screening is conducted in the Spring.

The "curve fitting procedures" used to generate grade equivalency scores introduces errors large enough to invalidate evaluations.

• Due to the relative stability of the average reading level in the population after junior high school, and to the extrapolation of grade equivalents at the upper grade levels from the earlier grades, the growth curve between age and achievement flattens out in the upper grades.
  
  

• Grade equivalents at higher grade levels become ludicrous with large fluctuations resulting from a raw score difference of only two or three points on a 100 item test.

Grade equivalents vary markedly from test to test and from subtest to subtest within the same battery. Therefore, there is no technically sound reason to justify their use in the identification, diagnosis and remediation of learning disabilities.

Suggestion for further reading:

Berk, R. A. (1981) What’s Wrong with Using Grade Equivalent Scores to Identify LD Children. Academic Therapy, 17(2) 133-140.

Sattler, J. M. (2001) Assessment of Children: Cognitive Applications (Fourth Edition)

This site was designed by Margaret J. Kay, Ed.D.  Its contents are presented for informational and educational purposes only and are not to be construed as professional advice on medical, legal, technical or therapeutic matters.  By using and accessing the information on this site, you agree to waive any rights to hold the site developer, or any individual and/or group associated with this site, liable for any damage that may result from the use of the information presented here.

© Copyright 2007-2008 Margaret J. Kay. All rights reserved.

The copyright of design, text and images on this web site is owned by Margaret J. Kay or the individual copyright owners as noted elsewhere on this site.  You may download and reprint articles from this web site for non-commercial, private, educational purposes only.  You may not in any way modify, or publicly distribute, any information contained within this site without specific permission form the copyright owner.

Send mail to MJK@MargaretKay.com with questions or comments about this web site.

Last modified: September 22, 2008