Formal programs that provide research experiences for teachers (RET) have been in existence for more than 20 years. Currently there are more than 70 formalized Scientific Work Experiences Programs for Teachers (SWEPTs) nationwide. The underlying assumption of most RETs is that these intensive summer work immersion experiences, coupled with appropriate follow-up activities during the school year, expand teachers’ professional skills and networks, and thereby improve the performance of their students. Many SWEPTs have collected anecdotal evidence indicating their program’s positive impact on teachers. Missing from all SWEPT evaluations is quantitative evidence that teacher participation in these programs affects student interest and performance in the subject taught by the SWEPT teacher. As professional evaluators attest, it is difficult to differentiate the roles of teachers and teaching practices in changing student academic interest and performance from other factors (e.g., curriculum, school administration, non-random assignment of students, etc.). This study controls for many of these factors by comparing interest and achievement of students in classes of SWEPT teachers with students in classes of comparison teachers in the same school and teaching the same subject. The study’s longitudinal design is commensurate with the philosophy and practices of the participating SWEPTs.

Research experience programs engage teachers in the hands-on practice of science. Program advocates assert that program participation enhances teachers' skills in communicating science to students. We measured the impact of New York City public high-school science teachers' participation in Columbia University's Summer Research Program on their students' academic performance in science. In the year before program entry, students of participating and nonparticipating teachers passed a New York State Regents science examination at the same rate. In years three and four after program entry, participating teachers' students passed Regents science exams at a rate that was 10.1% higher (P = 0.049) than that of nonparticipating teachers' students. Other program benefits include decreased teacher attrition from classroom teaching and school cost savings of U.S. $1.14 per $1 invested in the program.