At a first glance, exam school stu­dents do well on most mea­sures of achieve­ment. This may be why many par­ents dream of send­ing their chil­dren to top exam schools. However, while exam school stu­dents clearly excel in school, the ques­tion remains: Does receiv­ing an exam school edu­ca­tion add value rel­a­tive to receiv­ing a stan­dard pub­lic edu­ca­tion for already high achiev­ing stu­dents? To answer this ques­tion, SEII researchers employed a research design which com­pares test scores of the mar­ginal appli­cants – those who fall just above and just below admis­sion cut­offs. Using this design, the study esti­mates the causal effect of exam school atten­dance on stu­dent achieve­ment as mea­sured by stan­dard­ized state test scores. The esti­mates show lit­tle effect of exam school offers on most stu­dents’ achieve­ment in most grades. Students who win these prized seats might have ben­e­fit­ted from expo­sure to a higher-achieving group of peers and a more demand­ing cur­ric­ula, but the results indi­cate that in spite of these fac­tors, the stu­dents just above the admis­sions cut­off do not reap these ben­e­fits in the form of improved test scores. While there are a hand­ful of small gains, the intense com­pe­ti­tion for exam school seats does not appear to be jus­ti­fied by improved achieve­ment for most stu­dents.

http://seii.mit.edu/research/study/do-elite-exam-schools-add-value/