Table 1.
Participant Demographics. Students were marked as Hispanic regardless of race chosen and were not marked in any other category to prevent double counting. Some groups may not sum to 100% due to rounding.
Table 2.
Careers presented on the survey. These careers were presented without any designation of STEM or non-STEM. Careers were classified into these categories during data processing.
Table 3.
STEM programs listed in the survey.
Table 4.
Overview of logistic regression variables.
Table 5.
Percent of students that attended each type of STEM program or no program, broken up by demographic groups. Rows may not sum to 100% due to rounding.
Table 6.
Logistic regression results from multiply imputed data predicting STEM career interest at the end of high school. Model 1 includes control variables only. Model 2 adds STEM program attendance as the primary predictor variable. Shaded cells are statistically significant at the 0.05 level or lower.
Fig 1.
Average adjusted predicted probability of STEM career interest at the end of high school from model 2.
Error bars represent one standard error of the predicted mean. Gen. pop. program: general population STEM program not specifically designed for underrepresented minorities. URM Program: STEM program designed specifically for underrepresented minorities. Green points are the predicted probability of STEM career interest for students who did not attend a STEM program. Blue points are the predicted probability of STEM career interest for students who attended a general population STEM program. Red points are the predicted probability of STEM career interest for students who attended a STEM program designed for underrepresented minorities.