School Rankings: Descriptive Statistics or Incentives to Manipulate?
17 Pages Posted: 24 Dec 2013
Date Written: December 24, 2013
Abstract
Recently, President Obama has proposed that the Department of Education start ranking colleges based on “value.” School rankings have, for some time, formed an important component of students’ decisions on which institutions to attend. However, as we illustrate in this paper, such rankings can be easily manipulated. We discuss some of the metrics used in outcomes assessment, as well as many of the statistics and rankings used by potential students in selecting a school. We discuss how sampling bias can make many of these statistics suspect, and how this creates an incentive for schools to spend resources on inflating them. This paper, one in a series designed to provide examples of how statistics can be used to mislead, is of particular interest to students, most of whom have made use of some ranking in deciding which school to attend. Many of the “tricks” described to improve rankings also have serious ethical implications. This leads to a discussion of ethical issues in the use of statistics, which would be particularly useful to those looking to introduce a discussion of ethics into a statistics course.
Keywords: Misuse of statistics, PISA scores, rankings of schools, ranking of law schools, higher education, and best value lists
JEL Classification: C00, C10, C40, I20, L15, L21, L82, L86, M31, M37, O3
Suggested Citation: Suggested Citation