Sailer Is Right: Measure School Achievement Relative To IQ!

I was drawn to Steve Sailer`s latest article (Sailer`s Four-Point Plan For Improving Schools), because it reinforces my own thoughts. Steve gave four ideas for improving schools. The one that gave me the biggest kick was “#4: we need to measure school achievement relative to the IQ of each student.” Amen!

We are constantly, shrilly condemning failing schools” when we should be condemning “failing students.” But no, that`s not quite right either. We should not condemn the students since they are in most cases doing their best with the intellectual talent that they were born with. No, condemnation is not justified, either of a school or its students, when both are giving all they have to give. And as I have seen, that is generally the case.

Four years ago, I gave a short presentation in Montgomery, Alabama to a conservative group formed under the aegis of Grover Norquist.  My subject was “Notes on School Ability and Achievement”.  I sounded one of the same subversive ideas that Steve puts forth – namely that a student`s ability largely determines his achievement despite any innovations his school may invoke, and that the aggregate ability of a school`s students all but dictates that school`s performance (at least as measured by standardized test scores).

A colleague and I downloaded from the Alabama State Department of Education test scores (actually test rankings expressed in percentiles relative to a normalized national sample) for the 25 elementary schools in Huntsville, Alabama. Then we performed an analysis of how each school fared compared to the ability of its students. [See Table—the table also includes the percentage of black students in each school.]

Alabama gives both the Stanford Achievement Test (SAT), which measures student achievement, and the Otis-Lennon School Ability Test (OLSAT), which measures a student`s aptitude – in other words, his IQ.

We developed a measure we called “Teaching Effectiveness”, which is the quotient of achievement divided by ability (multiplied by 100 to give it a convenient scale).

If a school has a Teaching Effectiveness of 100, then that school can be said to be performing “at par,” since the ranking of its students` achievement on the percentile scale matches that of their ability. Any school with a TE below 100 is performing below par. And, of course a TE above 100 means that that school is performing above par. That school is exceeding what would be expected, given the ability of the students enrolled there.

The chart shown here made the key point in my presentation.  


At a glance we can see how achievement and ability are close to coincident. And especially interesting is this:

At the left end of the chart we see that five of the six schools with the highest achievement scores are in fact performing below par. These schools are the ones who would get the accolades from the school administrators, the media, and the public. But they actually are doing nothing remarkable.

Now look to the right end of the chart and the schools with the low achievement scores. These schools are the strugglers, the ones over which many hands are wrung and many teeth gnashed, all to the acute discomfort of the principals in charge. These principals will have no flowers delivered to their offices.

Yet four of the six schools with the lowest achievement scores are teaching above par!

This is just one set of data from the elementary schools in one city system, but we find the same general profile more or less universally.

If Steve needs vindication for his point 4, here it is.

Hugh McInnish [email him] is a consulting engineer and publisher of