Bishop S,
Faulk D, Patterson J, Santy
P.
Aviat
Space Environ Med.
1993 May; 64: 453.
University
of Texas Medical Branch,
Galveston 77555.
INTRODUCTION. The screening
and evaluation of astronaut
candidates often includes
various measures of
cognitive and intellectual
capacity. There are
indications that use of such
measures is of equivocal
value in identifying
functionally significant
differences in a population
of highly accomplished and
highly screened individuals.
METHODS. Analyses of group
differences in IQ as
measured by the
Multidimensional Aptitude
Battery (MAB) were conducted
across Shuttle astronaut
candidates exploring
differences between and
within sex, selection and
job category. Factor
analysis of the subscales
was undertaken to validate
the compilation of IQ scores
for the MAB which differs
from many other IQ scales in
grouping Arithmetic with
Verbal components. RESULTS.
Arithmetic subscale
clustered with Performance
subscales and not with
Verbal subscales as
originally designed. A
reformulated Verbal and
Performance IQ score was
computed on prorated
subscale scores regrouping
Arithmetic with Performance.
Subsequent analyses of the
original and revised Verbal
and subgroup Performance IQ
scores found substantial
subgroup differences in
findings between these
versions. CONCLUSION. The
reversal of findings between
the two versions of Verbal
and Performance IQ indicate
that inclusion of the
Arithmetic subscale
significantly affected the
importance of the global
scale to which it was added.
This suggests extreme
caution in applying measures
normally valid in most
populations to samples which
do not represent the norms
on which these instruments
were developed.
Source:
http://gateway.nlm.nih.gov/MeetingAbstracts/102212611.html
