Evidence about the role that “soft factors” like student engagement and school environment play in influencing whether high school students go on to enroll in college is hard to come by. Over the past two years, the Center for Labor Market Studies (CLMS) of Northeastern University, with support from the Nellie Mae Education Foundation and the Rhode Island Board of Governors of Higher Education, has explored the impact of these and other factors on the college-going rates of recent high school graduates from Rhode Island public high schools.
This study is based on a unique database that CLMS painstakingly constructed linking institutional level data about high school students, teachers, parents and school-level characteristics with information on college-enrollment behavior of high school graduates from each public school in the state. The unit of observation in this study is the high school. The measure of student high school experiences in this study is defined here as encompassing not only the academic and social attitudes, practices and outcomes of students, but also the academic and social organization and climate of their high schools, and the involvement and experiences of their teachers and parents. The sets of measures developed for this study provide new insights into the role that these soft factors play in explaining differences in college enrollment rates across high schools in the state.
Summary of Key Findings
Student Academic Performance. Consistent with many other studies of the determinants of college enrollment, we found moderate to strong connections between alternative measures of academic ability and college enrollment. Measures of academic performance across high schools including the mean mathematics and verbal SAT scores, percentage of students scoring at the proficient level in the 11th grade standardized test in English language arts and mathematics, and the graduation rate of the high school, all had a statistically significant association with a higher college enrollment rate.
Student Demographic Characteristics. Again consistent with many other studies, we found moderate to strong connections between socioeconomic status and college enrollment rates across high schools. The college enrollment rate is low among schools with higher shares of students with disabilities, Hispanic or black students, students from low-income families and students with poor English-language proficiencies. Schools with higher shares of students who access primary health care through community clinics or emergency rooms have lower college attendance as well. Conversely, schools with larger shares of children who access primary health care through a physician’s office have higher college attendance rates. Access to health care through community clinics or emergency rooms is more likely to occur among children from lower-income families, while children from higher-income families are more likely to use a physician’s office to access primary health care.
Student Behaviors. Schools with high dropout rates and high rates of turnover in the student body have considerably lower college attendance rates among those who graduate. However, consistent with our research in other states, we found that incidents of student suspension are not related to college attendance in Rhode Island. Student health risk behaviors such as illegal drug use, the use of tobacco products and alcohol have very modest connections to postsecondary enrollment, whereas student nutritional practices including consumption of fruits and vegetables and eating breakfast regularly have a fairly strong relationship to college enrollment in Rhode Island; we speculate this may also be closely related to socioeconomic status. Excessive television viewing (two hours or more per day) is fairly strongly related to lower rates of college attendance, whereas excessive use of Internet-based services (email, text messaging and chat rooms) is not related to college attendance in the state.
Student Engagement and Connectedness. Measures of student engagement with their teachers and connectedness and a sense of belonging in the school are found to be unrelated to college attendance rates of high schools in Rhode Island. Out of six variables measuring student engagement and connectedness, only one variable—how students get along with their peers—is modestly negatively correlated with the college attendance rate at the high school. Schools with a larger share of students who have trouble getting along with each other tend to have lower college-enrollment rates.
School Environment and Expenditures. The study uses a number of measures to gauge the school environment including the college-going climate, career-preparation activities, student safety and the availability of social support from teachers and staff members. Among the variables measuring the college-going climate in the high school, we found that attending a high school with many college-going peers and peers who engage in college-related activities like taking advanced placement exams and SAT tests is closely related to the college attendance rate. Access to teachers to discuss college does not appear to have a strong connection to college going rates. We find no association between college-going rates and career-preparation activities including career-exploration activities, field trips and career training in the classroom and through activities in the community. Social support from teachers related to family and personal problems does not correlate to college attendance, while academic support from teachers does. Student reports on the safety of the school environment are not related to college-going rates at Rhode Island’s high schools. The study found no relationship between per-pupil expenditures and college-going rates of high schools.
Teachers: This study examined the relationships between college-going rates at high schools and teacher quality, curriculum and instructional practices, teacher supports, and teacher engagement and connectedness. Our analysis found that only a few teacher-related measures are related to students’ college going rates. Teacher quality—measured by shares of teachers who are hired with emergency certification—has a weak correlation with the college-going rate of high schools in Rhode Island. Higher shares of classes not taught by qualified teachers is not related to the college-going rate of Rhode Island high school graduates. Only one factor out of the list of curriculum and instructional practices—regular instruction in writing skills—is found to have a moderate to weak negative relationship to college enrollment. Other teaching practices including examining student work to guide instruction, relating instructional material to student interests, as well as teaching students problem-solving and decision-making techniques are unconnected with college-going rates.
Teacher supports in this study are measured by the self-reported level of support in the form of professional development, curriculum development, support from colleagues and exchange of information with them, availability of sufficient preparation time and information about school improvement, decision-making opportunities and authority regarding management of student discipline and behavioral problems. Although teacher views of the level of supports that they receive vary sharply, these reported levels of support are not connected with the college enrollment rate of students. Teacher connectedness and engagement measured by teacher perceptions of their connectedness toward other teachers, students and their parents are largely unrelated to the college-going rates at the high school.
Parents. Three measures of parent engagement and connectedness with the school are found to be related with the high school’s college going rate. Higher college going rates were found at schools where large shares of parents believe that the community supports the school, the school promptly responds to their concerns and questions, and the school asks them to volunteer.