Dear Mr. Boekelheide,
We are writing about the MAP for Primary Grades Initial and Preliminary Norms. In October we were given hand written notes for our daughter�s RIT scores and associated percentiles:
Fall 1st Grade: Reading RIT 195 (81%) Math RIT 187 (77%)
These percentiles did not make sense to us, so we started asking questions and digging deeper. Eventually we were told the scores had been renormed using the MAP for Primary Grades Preliminary Norms. We compared the data from the Preliminary Norms with the RIT Scale Norms for Early Primary Grades (RSN for EPG) dated November 29, 2007.
Specifically, the standard deviations (SDs) for the Preliminary Norms appear to be 33 points for math and 31 points for reading. The SDs for the RIT Scale Norms are 12 points for math and 10 points for reading. To put it in perspective, the difference between the beginning of year median for 1st grade and 3rd grade is 32 points in reading and 28 points in math.
After pointing this out to our school officials, they contacted NWEA and found out the Preliminary Norms should not have been used. We are concerned about how the Initial and Preliminary Norms were developed, how they were distributed to school personnel, and how NWEA has communicated with parents and schools about the validity of this data.
Testing gives objective information that is necessary for teachers and parents to make informed decisions about instruction and curriculum in the classroom, school, district, and home. Accurate assessments are critical to the credibility of these tests. Inaccurate data presents a whole host of problems. For example, the Individuals with Disabilities Act requires Individual Education Plans to include measurable annual goals and a description of how the child's progress toward meeting the annual goals will be measured. It is critical to include accurate, objective measurement in IEPs. Changes in test scores over time provide the means to assess educational benefit or regression.
In addition, a number of programs for gifted children require a qualifying test score. For example, the Johns Hopkins CTY Talent Search requires achievement at the 95th percentile or higher on one or more subtests of a nationally-normed standardized test and MAP is one of the tests CTY accepts. When parents are given erroneous data showing their child is at the 81st percentile instead of the 99th percentile, children can lose out on opportunities to participate in these types of education programs.
Educational and psychological tests are designed to present normal bell curve distributions with predictable patterns of scores. When school officials are using data that shows a child is 1 standard deviation from the mean instead of 3 standard deviations from the mean, there should be some concern about whether children at that school are getting appropriate services designed to meet their unique needs. For above the mean, this is a difference between the 84th percentile vs. the 99th percentile. For below the mean, this is a difference between the 16th percentile vs. the 1st percentile. The need for intervention may be obscured if the wrong percentiles are considered. These are some of the impacts of having erroneous data.
We have attempted to address this at the local level and will continue to work with our school district in getting the correct data to teachers and parents. Unfortunately we are concerned that other districts have made similar errors and may also be using the erroneous data. For example, we found the MAP for Primary Grades Initial and Preliminary Norms posted on different districts� web sites:
http://www.bismarck.k12.nd.us/uploads%5Cresources%5C2530%5Cmap_pri_fall_percentiles.pdf
http://www.bismarck.k12.nd.us/uploads%5Cresources%5C2531%5Cmap_pri_spring_percentiles.pdf
http://teachers.greenville.k12.sc.u...20Training%20Reference%20Links091207.docWe feel the ramifications of inaccurate data necessitate NWEA taking additional steps to ensure districts, teachers and parents have the correct data instead of the erroneous data. Please let us know what NWEA is willing to do to resolve this issue.
Sincerely,
