The project uses a longitudinal and comparative research design and will generate evidence that is descriptive [case study and observational], associative/correlational [quasi-experimental] and causal [quasi-experimental]. Original data are being collected on high school students using school records [student end of course scores, state assessment tests, attendance], assessments of learning, personal observation, and survey research [self-completion questionnaire, structured interviewer-administered questionnaire, semi-structured or informal interviews, and focus groups]. Instruments or measures being use include the YouthTruth survey administered by the Center for Effective Philanthropy.
Intensive STEM-focused model-based curricula will be compared with traditional curricula using 10 selected STEM-focused high schools and 10 matched comparison schools. The impact questions follow the same students over time as they progress from 9th through 12th grade. In all analyses, students will be nested within schools, and their outcomes will depend, in part, on the individual school environment. Given this nested design, analyses will incorporate hierarchical linear modeling (HLM) to determine the impact of attending a STEM school on the student-level outcomes. To estimate changes in student performance over time, we will rely on the analysis of piecewise linear growth curves within a hierarchical linear model.