But these raw differences narrow substantially
when differences in test scores are accounted for.
Not exact matches
When we clump children together into a single diagnostic category based on
test scores, we not only fail to address what might be causing the dyslexia, but we also ignore variability
in performance that limits our ability to identify individual
differences.
Specifically, I examine whether the results change
when I adjust my results to account for
differences in student characteristics, including prior (age 7)
test scores; gender; eligibility for free lunch; special education needs; month of birth; whether first language is English; ethnic background; and census information on the home neighborhood deprivation index.
The first paper, released
in July 2009 by Roland Fryer and Steven Levitt, found that while there are no mean
differences between boys and girls
in math
when they start school, girls gradually lose ground, so that the gap between boys and girls after six years of schooling is half as large as the black - white
test score gap.
The study found that Spanish - speaking students earn lower grades
in advanced courses than non — Spanish - speaking students, but these
differences disappear
when students have the same GPA and
test scores in the prior year and attend the same school.
When we administer these exams, we attempt to keep the
testing conditions consistent for all students so that we can be sure that
differences in scores are due to
differences in knowledge and abilities rather than other unrelated factors.
When you have statisticians trying to explain
differences in test scores as being equivalent to N days
in school, one imagines that a 40 day
difference in test date would have some impact.
First, we know from earlier studies that student attainment levels - high school graduation or enrollment
in post-secondary education - may be higher among voucher users even
when test score differences between them and their public school counterparts are nonexistent.
The
difference in test scores produced by the incentive system was about the same as that detected
in earlier studies that measured
differences in student performance
when kids were taught by great teachers rather than average teachers.
However,
when tested with the three group severity variable a group
difference in PSI
scores did emerge F (2, 942) = 7.71, p <.001.