Media Releases >
Media Releases Archive
> Researchers question Fraser Institute school rankings
Researchers question Fraser Institute school rankings
Document Tools
Contact
Ruby Ng, faculty of education, 604.268.7152 (w), 604.828.9516 (cell)
rubyng@sfu.ca
Marianne Meadahl, Media & PR 604.291.4323
Ruby Ng, faculty of education, 604.268.7152 (w), 604.828.9516 (cell)
rubyng@sfu.ca
Marianne Meadahl, Media & PR 604.291.4323
June 5, 2006
Three education researchers at Simon Fraser University are advising the public to be cautious about how they interpret the Fraser Institute’s BC school rankings to rate public schools.
Using well-known statistical and psychometric methods, professors Phil Winne and John Nesbit, and research associate Carmen Gress illustrate how composite ratings such as these may not be precise enough for practical decisions such as choosing a school. Their report Cautions About Rating BC's Schools highlights some of the issues affecting these kinds of ratings.
The trio points out that such school ratings inevitably have some error built into them. “As a result, any particular school’s rating is somewhat unreliable,” says Winne. He and his colleagues used basic psychometric methods to demonstrate that “the odds are 19 to 1 that any particular elementary school’s rating could be about 1.1 points lower or higher than what was published in the Fraser Institute’s report.”
Using straightforward statistics, Winne and his team also show that a student with an average score on Grade 12 provincial exams might receive a grade from their teacher as high as A or as low as F in English 12. The Fraser Institute’s ratings of secondary schools are largely based on provincial exam scores. “Teachers can observe a student’s work over a much broader range of the curriculum than a one-shot test. It’s important not to over-interpret what a single test score can reveal,” suggests Winne.
“We aren’t claiming ratings like these don’t have uses,” Winne observes. “What we urge is not to overlook the particular limitations of any kind of school rating.”
The SFU researchers offer several practical reminders to parents: “Don’t rely on a single indicator when judging a school — a variety of factors contribute to a successful learning experience. And, bear in mind that every measure has some drawbacks.”
The data used in the analysis was obtained through a research agreement with the BC Ministry of Education.
-30-
Using well-known statistical and psychometric methods, professors Phil Winne and John Nesbit, and research associate Carmen Gress illustrate how composite ratings such as these may not be precise enough for practical decisions such as choosing a school. Their report Cautions About Rating BC's Schools highlights some of the issues affecting these kinds of ratings.
The trio points out that such school ratings inevitably have some error built into them. “As a result, any particular school’s rating is somewhat unreliable,” says Winne. He and his colleagues used basic psychometric methods to demonstrate that “the odds are 19 to 1 that any particular elementary school’s rating could be about 1.1 points lower or higher than what was published in the Fraser Institute’s report.”
Using straightforward statistics, Winne and his team also show that a student with an average score on Grade 12 provincial exams might receive a grade from their teacher as high as A or as low as F in English 12. The Fraser Institute’s ratings of secondary schools are largely based on provincial exam scores. “Teachers can observe a student’s work over a much broader range of the curriculum than a one-shot test. It’s important not to over-interpret what a single test score can reveal,” suggests Winne.
“We aren’t claiming ratings like these don’t have uses,” Winne observes. “What we urge is not to overlook the particular limitations of any kind of school rating.”
The SFU researchers offer several practical reminders to parents: “Don’t rely on a single indicator when judging a school — a variety of factors contribute to a successful learning experience. And, bear in mind that every measure has some drawbacks.”
The data used in the analysis was obtained through a research agreement with the BC Ministry of Education.
-30-