Understanding Test Scores
What can I learn from test scores?
Test scores provide an estimate of
your child’s performance in the area measured by the test. So, if the test
purports to measure spelling, the test score will give you an estimate of your
child’s performance (on that test) in spelling. The accuracy of your child’s
score on that test depends on many factors including
Why administer special tests to
measure factors the teacher already understands about my child?
One source of information is never
sufficient for understanding a child’s performance. Test scores provide
additional information to support what the classroom teacher is finding in
school. The more information available the more likely it is that relevant
services and supports can be implemented to facilitate your child's academic
success.
Test scores provide helpful information for problem solving and
decision-making.
What do all the different scores mean?
Raw Scores:Because of differences in tests (number of total items,
types of scoring procedures, frequency of use, etc.), it is difficult to
compare raw scores from one test to another.
Let's say your child took two math tests. He scored
Each of these tests measures a
different skill and requires the performance of an entirely different set of
tasks. Comparing the scores10 and 62 tells us very little unless we have
additional information (like the total number of items or the types of problems
on each test), especially if the content on each test was very different.
For example, perhaps the score on the timed test was for all the basic
multiplication facts in the 7s, 8s, and 9s, and the score of 10 was on basic
addition facts in the 1s and 2s!
Standard Scores:
Standard scores are scores that have been standardized - scores that have
been converted to a common scale.
Because the scores are on a common
scale, they can be compared to scores from other tests with the same common
scale. There are several types of standard scores. Generally, however,
standard scores from 90 to 110 represent an average range.
Scaled Scores:
Scaled scores are another type of standard score, generally used for
reporting subtest performance.
Generally, in special education
assessment, many tests are administered. An entire group of tests over many
subject areas is called a test battery while one specific test measuring
a snapshot of a larger content area is called a subtest.
For example, a series of tests may be administered to determine a child's
performance in reading.
Individual tests within the battery
measure specific aspects of reading like
Scaled scores can be compared to
other scaled scores with the same battery, but not to scaled scores from
another battery (standard scores can be compared to
different tests).
Using the raw score example
above, if your child's test scores were part of a battery and were converted to
scaled scores, you would be able to meaningfully compare the two. If his
performance on the multiplication test was 9 and the addition test was 10, you
would know that he performed in an average range on both tests! You would also
know that he compared favorably with others of his same age and grade level.
Percentile Ranks:
Percentile Ranks indicate the rank order of
the scores for a given test.
A percentile rank of 60 indicates
that the test taker scored as well or better than 60% of the others who took the
test, or that there were 40% who scored less well.
Remember, percentile ranks are not
the same as percentages! A percentage
indicates how many of the items were answered correctly when the raw score is
divided by the total number of items on the test.
A percentile rank is the position of
your child's score when the scores of all test takers are ranked from highest
to lowest.
Percentile ranks for the common
special education assessments are based on the standard and scaled scores of
those in the norm sample. The average percentile rank is 50 and an
average rank is between the 25th and 75th percentiles.
Norm Sample:
The norm sample consists
of all
of the test takers who took the test for the publisher during test
development.
The scores of those individuals become
the sample to which your child's performance is compared.
Most reputable tests have very large norm samples that include test takers from
The degree to which individuals like your child were
included in the norm sample is an indication of the likelihood that the
comparison of your child's score to the scores of others in that norm sample is
valid and reliable.
Test publishers are often careful to include a wide variety of demographic
characteristics in their norm samples. They attentively account for ethnicity,
exceptionality, culture, school type and location and other characteristics in
their selection of test takers to include in the norm sample.
Use the following table to look up
your child's score and see where it falls in common standardized assessments
used for special education.
| Scaled Score |
Standard Score |
Percentile | Descriptor |
| 17 to 20 |
131 and above |
98 and above |
Very Superior |
| 15 to 16 |
121 to 130 |
92 to 98 |
Superior |
| 13 to 14 |
111 to 120 |
77 to 91 |
High Average |
| 8 to 12 |
90 to 110 |
25 to 75 |
Average |
| 6 to 7 |
80 to 89 |
9 to 23 |
Low Average |
| 4 to 5 |
70 to 79 |
2 to 8 |
Below Average |
| 1 to 3 |
69 and below |
2 and below |
Well Below Average |