Expectancy-value profiles in math and science: A person-centered approach to cross-domain motivation with academic and STEM-related outcomes

The need to enhance the STEM workforce and, in turn, the STEM educational pipeline is a prevailing issue in the U.S. One critical component in this pipeline is students' interest in STEM majors and their persistence in such majors, theorized to be a function of both students' perceived value and expectancy beliefs in the subject matter. Using an expectancy-value lens, we examined cross-domain patterns of high school students' expectancy beliefs and values in both mathematics and science using a person-centered or profile approach. With data from the High School Longitudinal Study, latent profile analysis revealed five profiles characterized as Low Math/Low Science (i.e., endorsing low levels of expectancy and value beliefs in math and science), Moderate Math/Moderate Science, High Math/High Science, Low Math/High Science, and High Math/Low Science. Taking into account aspects of students' background and school context, we found that motivational profile membership predicted math and science high school achievement, college persistence, and both STEM major intentions and major choices. Moreover, there were a number of gender and racial/ethnic differences and contextual variation in profile memberships as well. Implications for theory and educational practice are discussed in relation to study findings.