Are Hybrid Schoolers Reluctant to Participate in Private School Choice Programs? Experimental Evidence

• In Event: Homeschooling in America: The Changing Regulation, Motivation, and Population

Abstract

Objective and Rationale
Enrollments in hybrid schools – schools in which students attend physical classes for fewer than five days per week and are homeschooled in some form for the rest of the week – are trending up and many parents express a diverse range of reasons for enrolling their children in hybrid schools. While prior research invariably finds that parents prioritize academic quality when choosing a school for their child, academic quality is not always the highest priority and this preference is subject to various tradeoffs including other program characteristics. Though enrollments have risen, little is known about the pedagogical goals pursued by hybrid schools. We aim to help close this gap in the literature with a stated preferences experiment of hybrid school leaders’ perceptions of program success.
Methods and Data
We use a stated preferences experiment known as a conjoint experiment in which respondents are presented with a set of hypothetical programs, each with randomly assigned student outcomes across six components – standardized testing, higher education matriculation, civic engagement, family life, religious observance, and labor market outcomes. As outcomes are randomly assigned, we are able to estimate causal average marginal component effects (AMCEs) for each attribute on the likelihood a school leader would consider a program successful. Sixty-three school leaders participated in a survey experiment in which we randomly assigned attributes to hypothetical programs and asked school leaders to identify the most successful program. We find that hybrid school leaders consider a broad range of student outcomes when evaluating program success, including labor market outcomes, civic outcomes, and family life.
Results and Significance
Students’ religious observance produced the largest effect sizes (54 percentage points for students who were “actively involved in a religious community” and 32 points for students who were “involved in a religious community”), a reasonable finding considering that roughly two-thirds of the schools represented in our sample have some religious affiliation. Labor market outcomes also produced large effects, particularly when students were “gainfully employed” (31 points), engaged in entrepreneurial business ventures (27 points), or employed in a high-paying job (17 points). School leaders also favored programs whose graduates were civically engaged through voting (29 points), volunteering (20 points), or giving to charitable causes (15 points). Finally, family life outcomes mattered, but only for graduates who were married with children (16 points). We do not find evidence that test score outcomes and higher education matriculation contribute meaningfully to perceived success. Furthermore, we find some evidence of effect heterogeneity, particularly for school leaders in rural settings, who placed a higher premium on students matriculating to a “prestigious” college or university (32 points) and on various labor market outcomes relative to school leaders in other settings.

Authors

• Matthew H. Lee, Association of Christian Schools International

• Eric Wearne, Kennesaw State University

• John Thompson, Kennesaw State University