A+ LearningLink Assessment and NWEA MAP: A Comparison Study
Nicholas B. McDonald, Ph.D.
Thomas S. Trautman, Ed.D.
|Study Type: Quasi-Experimental|
|Grade Level: 3,4, and 5 Grades|
|Special Populations: Rural|
|Special Programs: N/A|
The importance of having assessments that dependably estimate student reading ability affects a wide range of educational practices. For example, it is critically necessary when matching individual students to suitable reading material. Different students are able to comprehend books of varying complexity and one purpose that reading assessments serve is to match appropriate level texts to students. With a suitable match between reading ability and text difficulty, the student learning process for any subject matter is enhanced.
The Lexile Framework for Reading® has been adopted by many states and incorporated into a number of published assessments as a measure of reading difficulty (Lennon & Burdick, 2004). The American Education Corporation (A+ LearningLink Assessment, 2009) recently incorporated the Lexile scale into its own reading assessment software. Because other independent assessments exist which generate Lexile scores, opportunities arise to compare Lexile scores generated by the American Education Corporation software, the A+ LearningLink Assessment (A+LL), to those independent tests. One such assessment is the Northwest Evaluation Association’s (NWEA) Measures of Academic Progress (MAP).
In early August 2009, an elementary school in a small Kansas district assessed its elementary student population’s reading ability with the widely-used MAP (Measures of Academic Progress, 2009). These same students, third, fourth, and fifth graders, were assessed with A+LL in November 2009. With these two assessments both producing a Lexile score for the same students (although the testing times were separated by about 3 months of instruction), it is possible to assess the amount of agreement or correlation between the two tests. Initial evaluation of the raw results suggested that the A+LL assessment produced consistently higher Lexile scores than the MAP assessment and concerns about a potential over-estimation of student reading abilities by the A+LL assessment were raised by that district.
To address this district’s concern about an over-estimation of student language abilities as reported by the A+LL assessment, two different statistical approaches were taken. First, statistical examinations (a Pearson correlation between the two test scores and a mean score comparison) were employed to evaluate if the observed difference in the Lexile scores generated by the MAP and the A+LL assessments was real or not. The second analysis examined how closely the A+LL results matched the MAP results in assigning the students to the classifications of Lexile’s Typical Reader and Text Measure categories (MetaMetrics, Inc., 2009).