Computer Science Trends and Trade-Offs in California's High Schools

• In Event: Moscone Center West Roundtable Session 4
  In Roundtable Session: Progress and Challenges in K–12 Computer Science Education: Evidence From Four States

Abstract

Purpose
This study seeks to understand the extent to which recent calls to increase access to computer science (CS) at the secondary level have been successful in terms of access and participation. We also explore potential staffing, curricular, and accountability challenges and trade-offs faced by school administrators considering CS expansions.

Perspective and Background
Despite years of advocacy for more CS curriculum in schools, we know little about how, if at all, CS courses have expanded in practice. Even more poorly understood are the ways in which schools expanding CS have managed associated trade-offs. For example, we know little about which courses schools replace with CS. It is also not clear what the qualifications of CS teachers are or whether teachers are demographically representative of the students they serve. These considerations must be weighed by school administrators and are therefore likely influence CS adoption. This is particularly true because CS course expansions may not be aligned with measures by which administrators are held accountable (e.g., standardized test scores).

Data and Methods
We use public course-level panel data on all high schools in California between the 2003-2004 and 2017-2018 school years to answer three broad questions about these issues. First, to what extent has secondary CS course taking expanded in California, both overall and for students of different ethnic backgrounds and gender identities? Second, which other courses, if any, see enrollment declines as CS courses expand, and does this have implications for student achievement on statewide standardized tests? Third, do students have equitable access to teachers as measured by certifications, experience, or the extent to which teachers are demographically representative of the students they teach? We combine basic descriptive techniques with generalized difference-in-difference models, using school and year fixed effects as well as time-varying school controls to explore how course enrollments, teacher characteristics, and test scores change within schools over time.

Results
We find that today two-thirds of high school students in California, including a majority of students from historically underrepresented ethnic groups, are enrolled in schools that offer CS, up from one-third in 2003. CS courses now represent 0.5% of all course enrollments, but significantly smaller shares for female, black, and Hispanic students. CS teachers remain predominantly white, even where the enrolled student population is not, and are primarily authorized to teach math or business, both allowable under California regulations. However, CS teachers are less likely than other teachers to be fully certified. CS enrollment growth has come at the expense of social science and English/language arts (ELA) courses and, to a lesser extent, business and language courses. Perhaps relatedly, CS enrollment is associated with lower achievement on state exams in ELA, but not in math.

Significance
Over and above needing to ensure equitable access to CS courses for students, policymakers need to attend carefully to related curricular trade-offs and staffing considerations. These issues have equity implications of their own and may determine whether, and to what extent, students and schools benefit from expanded CS course offerings.

Authors

• Paul Bruno, University of Illinois at Urbana-Champaign

• Colleen M. Lewis, Claremont University - Harvey Mudd College

• Anisha Kaul, Scripps College