Study Committee,
Presented to the State Board of Education
http://www.sbe.wa.gov/reports/CoM%20Final%20Report.htm
POSITION 3
Nick Brossoit,
Christie Perkins, Patty Raichle,
Laura Jo
Severson, Andy Wheeler, Ron Woldeit
While certain inferences from a test may be validly drawn for some groups
of students but not for others, any test that is being used for high stakes
decisions for all students must be proven to make valid inferences for all
students.
There are different kinds of validity, and adequacy must be demonstrated
separately for each kind. Testing experts agree that tests must be valid for
each purpose for which they are used. If a test is being used to determine
mastery of math, can we validly assume that a student who passes the test knows
math, knows the kinds of math expected to be mastered, and has mastered that
math at the expected level of performance? If the test is being used to
determine whether a student should graduate from high school, can we validly
assume that a student who passes the test knows the right information at the
right level of performance for success after high school?
Content validity research conducted by OSPI has determined that the 10th
grade WASL generally matches the 10th grade EALRs. However, this
determination is challenged by outside research. The Fall 2002 report from the
Stanford Research Institute raises questions about the match of the Math test
items to the 10th grade math EALRs, test coverage of all 10th
grade Math EALRs, and appropriate levels of difficulty. These issues point out
serious concerns about the content validity of the 10th grade Math
WASL. Corrections and follow-up
research need to occur before the State Board can determine the content
validity of the Math WASL. Obviously, similar research needs to be conducted on
the other three WASL-tested areas.