Educability-and-Group-Differences-1973-by-Arthur-Robert-Jensen

Culture-biased Tests 319
3. The exact value of h2 is not a crucial feature in this formulation; I
have used 0-80 here because it is the average value of h2 in studies
of the heritability of intelligence in Caucasian populations. Other
values would serve as well, although if they differed appreciably for
groups A and B, the regression lines would be non-parallel and
therefore, of course, would cross, creating more complex hypothetical
outcomes. It should be noted that h2 is the slope of the regression
line of phenotypes on genotypes and of genotypes on phenotypes;
while h (i.e., the square root of h2) is the correlation between
genotypes and phenotypes.
4. Analyses recently completed in our laboratory show that the rank
order of the percent passing (p) each item of the PPVT (as well as
of Raven’s Progressive Matrices) is virtually the same for very large
and representative California school samples of Negroes and whites.
The correlations between p values for these tests are above 0-95,
and it is interesting that in this respect the two racial groups are
even more alike than are boys and girls within each race. In other
words, the cultural biases in the test are more apparent with respect
to sex differences than with respect to race differences. (The sexes
do not differ appreciably in mean score, however, while the racial
groups differ about one standard deviation, or 15 IQ points, on the
average.) Moreover, the increments or decrements in the p values
of adjacent test items (i.e., pi~p2, Pi~Pb etc-) correlate above 0-90
between the racial groups. In this measure, too, the races differ
less than the sexes within each group.
Also it was found that the matrix of item intercorrelations and the
factor structure of these tests is not significantly different for white
and Negro samples when these are roughly matched for mental age
or total score. These properties of the data, for example, do not in the
least distinguish between fourth grade white children and sixth
grade Negro children. Yet they distinguish between fifth grade and
sixth grade Negro children and between fifth grade and sixth grade
white children. A culture-bias hypothesis would predict greater
Negro-white differences than adjacent grade differences in item
intercorrelations. The findings, on the other hand, are more consistent
with a development lag hypothesis.
In multiple-choice tests (as the PPVT and Raven are), there is
no systematic or significant racial difference in the choice of distractors
on those items that are answered ‘wrong’. A special scoring
key was made up so as to score as correct whatever response was
given by the largest number of children in the Negro sample. When
the tests were scored by this key, the Negro sample still averaged
lower than the white sample.
Scales based on subgroups of items which discriminate either