Harvard University will undertake a retrospective cohort study to analyze college students? education background and previous IT-related experiences for predictors of performance when students study Computer Science in college, while controlling for demographic differences. The study will reveal how decisions concerning the use of technology made by high school teachers in the teaching of science and mathematics and the growing number of high school computer science courses relate to the performance of 12,000 college computer science students nationwide, who will be recruited from randomly selected 2-year and 4-year colleges and universities. This study has the capacity to both reveal and evaluate the use of the most promising educational practices and interventions, including both in-school computing instruction and after-school programs, competitions, and clubs, and to assess the effects of a match between experiences during the high school years with the evolving landscape of introductory college computer science courses. Such knowledge can only help to ameliorate the high failure rates that prematurely end the pursuit of a potential computer science degree, especially for females and under-represented minorities who historically opt out of this field earlier and in greater numbers than other classmates. Both college and high school computer science instructors can also benefit from understanding the factors contributing to effective preparation for students? first college computer science course.
