Reliability of School Readiness Test Score
By Inderbir Kaur Sandhu, Ph.D
My son just turned 5 in March, and is going to 1st grade in September. I
had him tested for school readiness on the advice of the school, since
he is a bit young to be moving on, but is too advanced to stay in
kindergarten. We live in Portugal, and he has been in a Portuguese
school for 2 years, and has therefore only been speaking Portuguese for
2 years. His mother tongue is English, and since he was born in Korea,
and his dad is Korean, he also speaks quite a lot of Korean.
At the moment, he is able to do addition and subtraction, often into the
double digits, and he can multiply by 2, 5 and 10. He is also reading
independently, with very little help, on an advanced level for his age
(DK Readers level 2, Houghton Mifflin Reading Series nos 1.3, 1.4
comfortably, 1.5 with a little more help).
As part of the school readiness testing he had to do an IQ test. The
test that was administered was the Wechsler test, and it was
administered in Portuguese, over 2 sessions. He had a head cold for the
2nd session, and had had some medication for that as well. He ended up
scoring 118 on the test (I'm still waiting for the full report to
arrive, so this is based on a telephone conversation with the
psychologist, and I therefore have no idea what the verbal an non-verbal
scores are as yet).
I have to admit that this result has left me a bit flummoxed, since I
have an IQ of over 145, and his father has an IQ of over 135.
What I would like to know is whether the fact that the test was
administered in his 2nd strongest language and that he had a cold could
have affected the outcome of the test very significantly? I must admit
that I'd expected a score of at least 130, given that there is not an
immediate family member on either my side or his dad's side with a score
that is lower than that.
I do realise that this score is not indicative of his actual potential,
creativity, personality, worth etc, and I take the results with a pinch
of salt. The psychologist has also recommended that he does go to the
first grade, even though he's a bit young, since he's a very focused,
sociable child. So the IQ score doesn't trouble me too much. However, I
don't want his potential to be underestimated or for him to be
under-challenged because of an error-prone score either. My feeling as a
parent is that the score is inaccurate, but I'm also concerned that my
expectations of him are a bit high because I'm his parent.
I can understand your concern and I guess most parents would be as well.
He does indicate traits of advanced learners and as you stated both
yours and your spouses's IQs, it is not surprising to expect scores in
the gifted range for your son. Based on his age, the test is probably
the WPPSI III (Wechsler Preschool and Primary Scale of Intelligence -
Third Edition). This is a test for preschoolers which is from the
Wechsler's group of tests that can be tested for children from 2.6 to
7.3 years of age. It is said to accurately measure intellectual
abilities in young children. This newer version features shorter, more
game-like activities that hold the attention of children as young as 2.6
The WPPSI-III contains the 14 subtests which are:
Block Design, Information, Matrix Reasoning, Vocabulary, Picture
Concepts, Symbol Search, Word Reasoning, Coding, Comprehension, Picture
Completion, Similarities, Receptive Vocabulary, Object Assembly, and
The subtests can also be combined to measure verbal IQ, performance
(fluid) IQ, processing speed quotient, general language composite and a
full scale IQ. The verbal IQ, performance IQ and full scale IQ are taken
from the core subtests. The other scores involve optional or
supplemental subtests and are not required.
The strengths of the test for preschoolers are that of being colourful,
current (being the latest), and interesting for children. Subtests rules
offer a variety samples and often second chances to assure the child
performs at the best of his or her abilities. The layered scoring allows
partial credit, for intermediate knowledge and rewards detailed or
extensive knowledge on subject matter which makes up for most abilities.
However, it is not without drawbacks. One weakness of this test is when
testing the very young, the test scales expect IQ to increase in very
sudden increments every two to three months. Therefore, is advisable for
test users to schedule testing to make it either most likely, or least
likely that a child would qualify for supplemental services. For retest,
that it may take up to two years. Due to this, a child who has taken the
WPPSI-III and needs to be re-evaluated in a short period of time should
be given another test.
As for the comparison with the WISC-IV, the WPPSI-III is found to be
rather similar to the WISC-IV on several subtests. Therefore, it is
important to select the suitability of the test between these two who
qualify in the age ranges of both tests. This is because the
similarities could create “practice effects” and may influence the
scores. I would recommend that for giftedness, the WISC-IV should be
given. For general academic concerns, the WPPSI-III would serve as a
better indicator. For reading, another test that complements the WPPSI-III
would be the WIAT-II which you may want to consider since he is already
reading quite early.
In addition, some studies show that intelligence tests such as the WPPSI-III,
especially for pre-K level, are unreliable and their results vary widely
with various factors such as retesting, practice (familiarisation), test
administrator, time and place. There are claims that some commercially
available materials improve results simply by eliminating negative
factors through familiarisation which in turn puts children at a
comfortable frame of mind.
Therefore, as for the scores, it is hard to tell without looking at the
detailed scores. He is in the higher range of ability but not in the
gifted range. The detailed score may give a better indication of whether
there may be in areas that needs more attention. The scores can vary up
to 10 points due to physical conditions, but a very significant
difference may not be possible - unless the tester is incompetent.
Having said that, I would not rule anything out and if after looking at
the detailed scores, you find some discrepancies, a retest can be
possible if the case and evidence of “error in score” is strong enough.
My best wishes to you.