Question and Answer

One-on-One With Today’s Educational Thought Leaders

The Problem with Grade-Level Equivalents
Eleanor E. Sanford-Moore, Ph.D., a MetaMetrics senior vice president, oversees the company’s research and development initiatives, including working with test publishers to link the Lexile and Quantile Frameworks with standardized reading and mathematics assessments and programs. Dr. Sanford-Moore is a former senior testing consultant for the North Carolina Department of Public Instruction, where she was responsible for developing the state’s End-of-Grade Tests (reading and mathematics) and End-of-Course Tests for 10 high school subjects. She has written test reviews for Buros Institute’s “Mental Measurements Yearbooks,” and has served on various national committees related to large-scale assessment. We sat down with Dr. Sanford-Moore to get her perspectives on the limitations of using grade equivalents to guide student learning.

1. What are grade equivalents?
   Grade equivalents are scores based on the performance of students in a test’s norming group. The grade equivalent represents the grade level and month of the typical, or median, score for students. For example, a fifth-grade student who earns a 5.9 on a norm-referenced test has earned a score similar to the students in the test’s norming group who were in their ninth month of fifth grade. Normative data are often collected at one point in the year from students in two or more grades. To obtain scores for all months and grades outside of the norming group, scores are extrapolated from the actual student scores.



2. Why doesn’t MetaMetrics use grade equivalents?
   Although grade equivalents are used widely in education today, they are a deceptively simple way to characterize students’ test scores. The misconceptions about these types of metrics have been well documented (AERA/APA/NCME, 1985; Airasian, 1994; Miller, Linn and Gronlund, 2009; Stiggins, 1009). In fact, in 1991, the International Reading Association (IRA) crafted a resolution about the misuse of grade equivalents. In it, the IRA “strongly advocates that those who administer standardized reading tests abandon the practice of using grade equivalents to report performance of either individuals or groups of test takers.” MetaMetrics elected not to report student ability levels as grade equivalents; instead the company built developmental scales for reading and mathematics: the Lexile and Quantile scales.



3. What is the difference between a Lexile or Quantile measure and a grade equivalent?
   There are many fundamental differences between Lexile and Quantile measures and grade equivalents. For example, a Lexile measure represents a student’s level of reading ability on the developmental Lexile scale. Similarly, a Quantile measure represents a student’s level of mathematics ability on the developmental Quantile scale. In contrast, grade equivalents represent a student’s ability level in comparison to students who were in the specific test’s norming group. Lexile and Quantile measures can stand alone in their interpretation. They do not depend on who was in the norming sample, when the norming test administrations occurred, or which testing instrument was used.



4. Why is a grade equivalent often misinterpreted as being a grade-level standard?
   The structure of the grade equivalent (grade.month) makes this a common and unfortunate misinterpretation. The grade equivalent does not represent a grade-level curriculum standard. A grade equivalent of 5.9, for example, does not represent the desired level of achievement for all fifth-grade students. It simply represents the norming group’s median score, or projected score, for fifth-grade students in their ninth month of schooling. Achieving the same score as the average student in the norming group may not be an appropriate goal for your students. Lexile and Quantile measures, on the other hand, are not generated from grade-level norms and do not presume a specific grade-level interpretation. Struggling students are not stigmatized with a grade equivalent that labels them as “below grade.” Rather, students have an independent Lexile measure that enables them to select appropriately difficult books and other reading materials within their Lexile range. MetaMetrics has studied typical Lexile and Quantile ranges for students in specific grades. Educators who are interested in this type of normative comparison can find this information on the Lexile and Quantile Web sites (www.lexile.com/FAQ and www.Quantiles.com/FAQ).



5. Why should a grade equivalent not be used to determine the appropriate grade placement for a student or the level of the material he or she should be studying?
   Grade equivalents never should be interpreted literally, but, rather, as rough estimates of grade-level performance. Imagine a student scores a 6.9 on a fourth-grade mathematics test. It should not be assumed that he or she has mastered the sixth-grade mathematics content. In fact, it may be unknown how sixth-grade students would perform on the fourth-grade test. Additionally, it can not be assumed that the student has the prerequisites for seventh-grade mathematics. All that is known for sure is that the student scored well above the average fourth-grade student in the norming group in mathematics. Because Lexile and Quantile measures do not suggest grade-level placement, they eliminate this type of misinterpretation.



6. Why should grade equivalent units not be used in mathematical calculations, such as determining the mean?
   The grade equivalent scale is like a ruler that has inches of different lengths. It is not an equal-interval scale. The grade equivalent units do not represent equal amounts of ability at different points along the scale. A student who moves the same number of grade equivalents at one level on the scale, such as from 2.5 to 2.9, has not necessarily “grown” in ability the same amount as a student who moves the same number of grade equivalents at a different level on the scale, such as 8.5 to 8.9. The amount of growth in ability needed to move from 2.5 to 2.9 is much greater than that required to move from 8.5 to 8.9. Because grade equivalents are not equal-interval units, they should not be used in mathematical calculations, such as averaging. The Lexile and Quantile scales, in contrast, are equal-interval scales. Regardless of the point on the scale, the amount of growth in ability required to move between two points is the same. In other words, moving from 240L to 340L on the Lexile scale represents the same increase in ability as moving from 840L to 940L. Lexile and Quantile measures can be used in mathematical calculations.



7. Are there instances when using a grade equivalent is appropriate?
   There are a few appropriate uses for using grade equivalents. Grade equivalents can be used to compare a student’s performance with that of the test’s norming sample, which for most tests is a nationally representative sample of students. For example, a sixth-grade student was tested in reading in May. His grade equivalent was 6.9. It can be concluded that this student is performing similarly to average students in the national standardization sample. Grade equivalents also can be used to interpret the performance of a group of students. Once the mean scale score for a group of students is calculated, it can be converted to a grade equivalent for the group as a whole. To illustrate: the mean scale score for students in Mrs. Johnson’s fourth-grade class tested during the last month of the school year was 693. When converted to a grade equivalent score (4.9), it can be concluded that the students in Mrs. Johnson’s class are reading at a level consistent with the students in the norming sample at the end of the school year.

LearningLink is published bi-monthly by MetaMetrics, Inc., an educational measurement company based in Durham, N.C. The company develops scientifically based measures of student achievement that link assessment with instruction, foster better educational practices, and improve learning by matching students with materials that meet and challenge their abilities. MetaMetrics developed the widely adopted Lexile Framework for Reading; El Sistema Lexile para Leer, the Spanish-language version of the Lexile Framework; The Quantile Framework for Mathematics; and The Lexile Framework for Writing. Contact the editor.

MetaMetrics®, the MetaMetrics logo and tagline, Lexile®, Lexile Framework®, Lexile Analyzer®, the Lexile logo, Quantile®, Quantile Framework® and the Quantile logo are trademarks of MetaMetrics, Inc., and are registered in the United States and abroad. The trademarks and names of other companies and products mentioned herein are the property of their respective owners. Copyright © 2009 MetaMetrics, Inc. All rights reserved.