This disquieting judgment is based on research into the predictive validity of intelligence tests like the WPPSI, tests of learned reasoning ability such as the CAT, and tests that address both intelligence and learning aptitude such as the Otis-Lennon. Bronson and Merryman report that researchers find the correlation between scores on these types of tests given to children before they start school and actual achievement two or three years out is a scant 40%. According to Dr. Donald Rock, senior research scientist with the Educational Testing Service, it isn’t until second or third grade that IQ measures become meaningful. IQ test authors put the mark even later, identifying score reliability as occurring somewhere around age 11 or 12.
Hmmm . . . what’s a school admission office to do? Well, Bronson and Merryman demonstrate that the answer does not lie in other existing tests such as those that rate attention skills or those that measure social skills and emotional intelligence to provide behavior ratings. These instruments, NurtureShock tells us, are even less predictive of future academic success. Their respective correlations to subsequent achievement are 8% and 20%.
Bronson and Merryman sum up the dilemma, “The issue isn’t which test is used, or what the test tests. The problem is that young kids’ brains just aren’t done yet.” Scientists studying the thickening cortex and shifting neural processes of the developing brain may ultimately help us understand why two-thirds of children experience a change, upward or downward, of 15 IQ points between ages 3 and 10. Until we have more information and better tests, a measure of caution seems in order for those of us in admission and for teachers and for parents, as well, as we consider test results for young children, lest we miscategorize, underestimate late developers, or make inappropriate demands on those with early development.
The good news at St. Patrick’s is that, unlike many other schools, we do not require standardized testing of our Nursery or PK applicants and, accordingly, create classes that include children with early and with later development patterns. We do test, of course, at Kindergarten and above but, like most good schools, see testing as just one piece of a much larger picture as we make admission decisions aimed at creating balanced classes.
Some other interesting points presented in Chapter 5 of NurtureShock.
- On an IQ test, fully one-third of the highest-scoring incoming Grade 3 students would have scored below average in testing administered before entering Kindergarten.
- In some cities, preschools proudly advertise that they test children as young as 27 months of age prior to admission.
- It is common for gifted children to make uneven progress.
- Teams at Cornell, Stanford, and Kings College, London have all found that children’s cognitive networks differ from those of adults.
- MRI scans of children and adults taking a simple verbal test showed that both used about 40 clusters of brain matter to accomplish the task but that only 50% of the brain clusters were the same.
- Every single scholar that Bronson and Merryman spoke to warned of classifying young children on the basis of a single entry test result.
_text_