House, Ernest R. (2007). Regression to the Mean: A Novel of Evaluation Politics. Reviewed by William L. Brown, Michigan Department of Education

House, Ernest R. (2007). Regression to the Mean: A Novel of Evaluation Politics. Charlotte, NC: Information Age Publishing Pp. 231     $40     ISBN 978-1-59311-849-5 Reviewed by William L. Brown Michigan Department of Education March 22, 2008 Ernest House is Emeritus Professor in the School of Education at the University of Colorado at Boulder and served for ten years as a member of the Editorial Board of the Education Policy Analysis Archives. He has been affiliated with the Center for Instructional Research and Curriculum Evaluation (CIRCE) at the University of Illinois, Urbana-Champaign, and has been a visiting scholar at UCLA, Harvard, and New Mexico, as well as in England, Australia, Spain, Sweden, Austria, and Chile. As a well-known educational evaluator, House knows well the cultural milieu that surrounds this sensitive professional activity. His evaluation projects have included a critique of the national Follow Through Evaluation for the Ford Foundation (1977) and an audit of the Promotional Gates Program evaluation for the Mayor's Office in New York City (1981). In Regression to the Mean, he follows the activities of a university professor named Reeder who is called upon to critique program evaluations for grants sponsored by the National Science Foundation, as well as to assess the quality of an evaluation of a “Second Chance” promoted by the Mayor of New York City for failing students in the city’s public schools. Given the obvious parallels between the programs House evaluated in the late 1970s and early 1980s, and those being evaluated by Professor Reeder, one is bound to assume that there is a good dose of reality in this fictional account. The book is very well written, and sustains the reader’s attention from start to finish. It gives a realistic glimpse into university faculty politics, the stresses of politically charged evaluations, and the conflicts experienced by a divorced man looking for love. The book’s title, Regression to the Mean, suggests that the story will relate to the statistical phenomenon that goes by that name. Indeed, the Second Chance program (which Reeder says would be better named Last Chance) is based on a testing program in which students are tested at the end of the school year to see if they are ready to advance to the next grade. Lower-scoring students are placed into a summer program where they receive remedial instruction followed by the same test to see if they can move ahead or if they must repeat a grade. Because of the nature of measurement error and regression to the mean, a significant number of children will make “gains” and qualify for promotion, whether they have actually learned anything or not. Worse still, the use of a single test lends itself to the possibility that security will be compromised and students might actually be drilled on “live” test items rather than learning the underlying knowledge and skills that are considered to be prerequisites for success in the next grade. As with any good modern novel, the hero uses his evaluation travel requirements as an opportunity to sustain romantic trysts with women in both Washington, DC and New York City. One of Reeder’s New York City liaisons intrudes into the love life of an important figure in the Second Chance program, leading to some difficult political and legal implications for this presumably objective evaluator. Without going into details that might spoil the plot for those who would like to read this very interesting and brilliantly told book, suffice it to say that many professional evaluators and a large number of evaluator wannabees will find reading this volume to be a good use of their time. As such, the book has instructional value as well as being good entertainment. After reading Regression to the Mean, I am led to wonder if House’s evaluation reports were as entertaining as his first step into fiction writing. The author, who was the 1990 recipient of the Paul F. Lazarsfeld Award for Evaluation Theory (presented by the American Evaluation Association), has demonstrated that it is possible to take a dry subject such as educational evaluation and turn it into a very enjoyable read. I heartily recommend that anyone who is considering life as an evaluator consider reading this book. One barrier is likely to be its price, which comes out to about twenty cents per page. This might discourage many prospective readers from buying it. As in some of his earlier writing, House examines how evaluation impacts modern societies and how this in turn impacts evaluators. Those who may have shied away from some of House’s previous, weightier tomes will learn much if they read Regression to the Mean. About the Reviewer William L. Brown is Coordinator of Test Development for the Michigan Department of Education’s Office of Educational Assessment and Accountability. He is responsible for developing K-12 assessments in the four basic content areas as required by the No Child Left Behind Act, and for monitoring indicators of student success for the Power of We Consortium in the Lansing Tri-County Area. He received his doctorate from Michigan State University in 2003. Email: brownb6@michigan.gov