![](https://csdnimg.cn/release/download_crawler_static/19894739/bg1.jpg)
THE EFFECTIVENESS
OF
FOUR READINESS TESTS
AS PREDICTORS OF FIRST GRADE ACADEMIC ACHIEVEMENT’
GABRIEL A.
TELEGDY
windsor
Separate
School
Board
The progress that young children make in their first year of formal education
is largely dependent upon their “readiness skills.” These include
:
language usage,
motor coordination, visual and auditory perception, number and letter knowledge,
attention to details and following directions. Several readiness tests have been
developed to measure the extent of
a
child’s mastery of the skills necessary for first
grade success. The tests usually consist
of
several subtests each proposing to
measure
a
distinct ability, which in combination give an indication
of
the indi-
vidual’s readiness for the demands
of
formal education. The implicit assumption
of
these
tests
is that the scores obtained by
a
child are predictive of his future
academic achievement in grade one. Frequently, however, the predictive validity
of readiness tests
is
low (i.e., they measure a variety
of
skills that are not predictive
of actual school performance). While the content and construct validity of these
tests appear adequate, factor analysis often reveals overlaps among subtests which
result in lengthy tests measuring only one or two distinct traits (e.g. Telegdy,
1974).
The present study was designed to examine the effectiveness of selected readi-
ness tests to predict first grade success. Four commonly used tests were adminis-
tered to children of both sexes
at
the end of their kindergarten year. One year
later all children were reassessed with two standardized achievement tests. Cor-
relation coefficients were computed to reveal the predictive validity of each readiness
test employed, i.e. which of the four tests best predicted actual first grade achieve-
ment. Multiple correlation coefficients were then calculated to reveal whether the
combination of the tests into
a
battery would improve their predictive power
sufficiently to warrant such an endeavour. And finally, a combination of the
maximum R2 improvement and stepwise multiple regression procedures was em-
ployed to arrive at the best subset of predictor variables.
METHOD
Test
Materials
When a test user wishes to predict an individual’s future performance, the
common procedure
is
to correlate the scores obtained on the predictive test with
scores on a criterion variable. The data are presented in the form
of
correlation
coefficients and the size of the coefficients defines the predictive validity of the
test, The chief problem in using correlational techniques to establish validity
is
the identification of an acceptable criterion that provides
a
direct measure
of
the
predicted variable. The establishment of
a
criterion for school achievement
is
further complicated by
a
growing suspicion among professionals about the validity
of
standardized achievement tests. For the purposes of the present study, it was
‘The present study
is
art
of
a
larger investigation within the Windsor Separate School System
into the assessment
of
reainess
among
kindergarten children. The author
is
indebt,ed
to
Lynne
E.
Shuliak and Pauline
Y.
Lee
for
their invaluable assistance.