A study by University of Windsor professors Ken Cramer and Stewart Page has found systemic problems in the statistical methodology Maclean’s uses for its annual university rankings.
The study, published in Academic Matters, a quarterly magazine covering postsecondary education, found that “the data seldom allow students or others to use the indices as logical or reliable indicators.”
Maclean’s ranking groups universities into three categories: Medical/Doctoral, Comprehensive, and Undergraduate, and uses 14, 13, and 13 indices respectively to rank universities in their category. In each category the authors found that less than half of the indices used were statistically significant, and that indices correlated unpredictably with each other.
The study also criticized Maclean’s “view of higher education as a type of financial issue and investment”, and encouraged taking a more holistic approach to university rankings.
Maclean’s has published its rankings since 1990. In 2006 over twenty universities jointly refused to participate in the rankings, including the University of Toronto.
With files from Academic Matters and Maclean’s
