Download the Study

The Elite Illusion:  Achievement Effects at Boston and New York Exam Schools

The Elite Illusion: Achievement Effects at Boston and New York Exam Schools

School Reform, January 2014

A study con­ducted by researchers Atila Abdulkadiroglu, Joshua Angrist, and Parag Pathak reveals that atten­dance at Boston and New York’s highly selec­tive exam schools have lit­tle impact on stu­dent achievement.

Parents gauge school qual­ity in part by the level of stu­dent achieve­ment and a school’s racial and socioe­co­nomic mix. The impor­tance of school char­ac­ter­is­tics in the hous­ing mar­ket can be seen in the jump in house prices at school dis­trict bound­aries where peer char­ac­ter­is­tics change. The ques­tion of whether schools with more attrac­tive peers are really bet­ter in a value-added sense remains open, how­ever. This paper uses a fuzzy regression-discontinuity design to eval­u­ate the causal effects of peer char­ac­ter­is­tics. Our design exploits admis­sions cut­offs at Boston and New York City’s heav­ily over-subscribed exam schools. Successful appli­cants near admis­sions cut­offs for the least selec­tive of these schools move from schools with scores near the bot­tom of the state SAT score dis­tri­b­u­tion to schools with scores near the median. Successful appli­cants near admis­sions cut­offs for the most selec­tive of these schools move from above-average schools to schools with stu­dents whose scores fall in the extreme upper tail. Exam school stu­dents can also expect to study with fewer non­white class­mates than unsuc­cess­ful appli­cants. Our esti­mates sug­gest that the marked changes in peer char­ac­ter­is­tics at exam school admis­sions cut­offs have lit­tle causal effect on test scores or col­lege quality.