Learn Your Special Education Laws, Special Education Rights, and Share IEP Goal Ideas

Aug 29
Profile photo of Jess

by Jess

There is an old adage: “Statistics never lie, but statisticians do.” In other words the results are not just the data, but rather how one analyzes and interprets the data. When reviewing assessment results, be sure to look at the actual assessment scores reporting grade level and percentile performance. I have had school psychologists claim that a child scoring in the first or second percentile is in the “low average range.” When they say this, my question to them is, “How low does the student have to score to be in the low range?” Also review the individual sub-test scores because the Broad Scores are averages that don’t always identify the needs of the student. 

On most assessments, results in the 1st to 7th percentile identify a student in need of special education help. Unfortunately, this varies from one assessment to another. The courts have ruled that a score more than 1.5 standard deviations below the mean qualifies the student for help. This decision makes interpretation more difficult. You can always ask the school psychologist, “What score is 1.5 standard deviations below the mean for this assessment?” Just knowing what to ask will put them on guard.

Another question I like to ask is, “For each subtest, how many questions did the student actually answer correctly?” Most assessments change the beginning questions for the student based on their age. This is called the basal and it assumes the student would give correct answers for all the questions below the basil. For older students this can result in them getting the first 5, 10, even 20 questions assumed correct though they may not answer many of the questions after the basal. Knowing how many questions they answered right is very important. Just because the results are reported in percentiles does not means there are 100 possible outcomes. When the student only answers 10 or less questions, the score would be very different had the student missed one more questions or had one more question correct. Statistically, this result makes the “precision” of the assessment very low. I see this very often in the Woodcock Johnson Academic Achievement and Cognitive Ability Assessments.

Based on the assessment results, the school will be basing their recommendations for qualifying for special education, and then on what services to provide,. This is why it is very important to look at the assessment results in detail. This is one of the services I provide our clients. I often perform a complete review of the assessment results and the school psychologist’s interpretation of the results. I also perform this service for other advocates and parents that represent themselves. Normally, I review the protocols which are the documents completed by the student when taking the assessment. I do this to check the actual scoring of the results. You would be surprised how often I find simple addition errors in the scoring.

One of the other areas of great concern is the selection of which assessments to administer. The school psychologist has many to choose from, but I find most settle into repeatedly using the assessments with which they are most familiar. This is fine unless the particular assessment was used in the recent past, or if the assessment is not inappropriate given the suspected handicaps of the student.

The selection of which assessments to use is a science. For example, non-verbal tests need to be used for many students. Sometimes key subtests are not administered. This can result in student problem areas not being identified.

The bottom line is that Assessing a student is a real challenge from selecting which assessment to use and which subtests to administer, to scoring correctly and interpreting the results properly.

Tim Runner is the founder of Advocates for Kids. He represents students and their parents all over the United States. He gives answers to the 100 most asked special education questions on his website: www.special-education-answers.com and he volunteers his time and knowledge on www.allexperts.com where he has answered over 1,000 questions and has high ratings from people asking questions.

Note: Tim Runner is uniquely qualified to perform assessment reviews because he has a degree in mathematics from the University of California with minors in psychology and physics. He took all the undergraduate and graduate statistics classes offered in the Psychology and Mathematics Departments.


VN:F [1.9.22_1171]
Rating: 0.0/5 (0 votes cast)
Be Sociable, Share!
Create Your FREE Profile

One Response to “What do Assessment Results Really Say?”

  1. I actually take issue with a few points in your article. First off, it is BASAL rates, not BASIL. The latter is an herb. Please have someone edit your work for simple but ridiculous errors prior to publishing.

    Second, the WJIII, the assessment you attack as being unreliable, has been come to be known as the gold standard of cognitive testing NOT because it is willy-nilly in its assessment, but rather because it has been proven, again and again, to accurately assess cognitive functioning. Yes, one or two answers one way or another will change a score drastically, but that is what it is supposed to show. This is NOT the SAT’s where they ask the same kind of question a dozen times to have you prove you know what you are doing.

    Thirdly, the 1.5 standard deviation below the norm IS, by its name, a STANDARD measure. If your IQ is 100, and the standard deviation is 15 points, then 15+7.5 gives you your 1.5 standard deviations BELOW the NORM. In other words, a score of 77.5.

    However, those numbers DO depend on that the actual IQ score was, and it slides in scale depending. A higher IQ will often mean you need a greater discrepancy, and a lower IQ means you don’t need as big a gap between ability and achievement to qualify as Learning Disabled.

    As for finding errors in adding up a student’s score, most of the assessment scores are computer generated today, so hopefully you and your calculator can catch a break soon.

    VA:F [1.9.22_1171]
    Rating: -1 (from 1 vote)