Gi/ted Classes
439
Simple Depth Model wilhout Intelligence Test
As
was the case with the breadth models, two simple depth models also were con-
sidered that required less extensive testing resources. The Simple Depth Model without
Intelligence Test assessed each child on the same three areas used
in
the Simple Breadth
Model without Intelligence Test-namely, school grades
in
major subjects, creativity test
scores, and behavior traits. Children who scored extremely
or
superlatively well (i.e.,
in
the top
2%
or
3%
of fourth graders
in
that town)
in
at least one
of
these three areas were
then admitted to gifted classes.
Simple Depth Model with Intelligence Test
In
the Simple Depth Model with Intelligence Test, moderate emphasis was given to
intelligence tests. Children were selected for gifted classes by this model
if
they scored ex-
tremely
or
superlatively well relative to the other fourth graders
in
the town on either
in-
telligence tests
or
school grades
in
major subjects.
Intelligence Test Model
In
the Intelligence Test Model, children were admitted to gifted classes
if
they
scored above a certain cutoff point on an intelligence test. This model differed from the
breadth and depth models described above, since it is a “single criterion” model that
assesses each child on only
a
single area, rather than on multiple areas.
METHOD
Subjects
The subjects were 159 white fourth graders (96 boys, 63 girls) from a small middle-
class to lower-middle-class town
in
southeast Michigan. Subjects were approximately
10
years of age. The 159 children included all the fourth graders
in
the town, and were
divided among six untracked classrooms (four classrooms
in
one school, two
in
another
school). Seven of the 159 children were receiving special education services that made
them ineligible for gifted classes.
School personnel had previously observed that there seemed to be an unusually large
number of bright fourth graders that year. After that observation, the Wechsler
Intelligence Scale for Children-Revised (WISC-R) was administered to the most promis-
ing fourth graders,2 some of whom were tested by Dirks, Wessels, Quarfoth, and Quenon
(1980). Testing revealed that 16 of the fourth graders were known to have Full Scale IQs
of
I30
or
higher
on
the WISC-R, including
12
children with a Full Scale WISC-R IQ of
132
or
higher.
Assess men
t
A
reas
Whenever possible, data were included that had previously been gathered by school
officials.
As
is
the case
in
many “real life” data samples, these data are characterized by
some
missing scores and some tied scores, which are noted below.
The promising children
(N=90)
who were given full
WISC-Rs
included any child who met any of these
criteria: scored in the top fifth of the sample on any assessment instrument
used
in the study: scored in the top
fifth of national norms on a group-administered intelligence
test:
was a transfer student with good recommen-
dations from previous teachers: was described by the child’s current teacher as being intelligent
or
as being
capable of doing better in class than s/he was currently doing; had scored well on the Mechanical Aptitude sub-
test
of the Differential Abilities Test (given
to
girls in conjunction with another research project); or had fre-
quently played a particular game that resembled one of the
WISC-R
subtests.