Abstract
BACKGROUND: Much of science, technology, engineering, mathematics, and medical (STEMM) education policy and research centers around developing the upper levels of the STEMM workforce sector. However, there are many positions in this workforce, "middle-skill careers," that are largely responsible for executing the innovations and are largely ignored in STEMM education research. RESULTS: Using data from the National Educational Longitudinal Study of 1988, we found differences in what predicts STEMM-related vs. non-STEMM careers across skill-level. For instance, underrepresented minorities and those exhibiting school transgressions are more likely to be working in middle-skill STEMM fields than in middle-skill non-STEMM fields as adults; the same is not true of the high-skill workforce. CONCLUSIONS: One-size-fits-all policies for broadening participation in the STEMM workforce across skill-level are unlikely to be successful. Further, programs that are designed to generate wonder and fascination with STEMM content may be successful in attracting more girls. However, to promote greater participation of individuals from traditionally underrepresented ethnic minority groups in STEMM, programs that support choices toward higher educational attainment, specifically four-year college degree attainment, are more likely to be successful.