Program

Dec 01, 2016

OVERVIEW

The School for Educational Evolution and Development in Washington, DC (SEED DC) is an urban, public, college preparatory boarding school for low-income students from underserved communities that aims to prepare students for college enrollment and success.  In addition to the academic program, SEED DC aims to develop students’ behavioral, social, and life skills to promote positive behaviors such as rigorous study habits and self-control, while discouraging negative behaviors such as alcohol use and unprotected sex.  A 2011 evaluation found that students randomized to the SEED treatment group had significant gains in standardized test scores in the seventh and eighth grades, compared with students who did not participate in SEED. A later evaluation found that random assignment to SEED DC resulted in significant positive impacts on students’ standardized test scores and proficiency levels — particularly in math — in years one and two, compared with students who did not win admission to SEED, but in the third and fourth year there were no significant differences between the groups. Winning the lottery to enter SEED DC did not significantly increase the proportion of students who graduated from high school in four years. Participating in SEED DC had significant positive impacts on courses taken, hours spent on homework, activities at school, school atmosphere, and reduction in tobacco use.  Participating in SEED also had significant undesired impacts on students’ frequency of risky behavior and students’ scores on measures of ‘grit’ or perseverance.

DESCRIPTION OF PROGRAM

Target population: Low-income middle and high school students from underserved communities in Washington, DC

The School for Educational Evolution and Development (SEED) Foundation was founded in 1997 and opened its first school in Washington, DC in the fall of 1998. The primary mission of SEED DC is to prepare students from low-income and underserved communities for college enrollment and success.  SEED DC is the nation’s first urban, public, college-prep boarding school that provides students with an intensive, integrated academic and boarding school program that includes scheduled study time, constant access to positive role models, and life skills training. SEED’s model is based on the presumption that, for certain disadvantaged students who face overwhelming barriers to success at home and in the community, piecemeal reform efforts will not be sufficient.  The student body expanded from 40 seventh graders in 1998 to 320 students in sixth through twelfth grade in 2011.

SEED DC and its Baltimore, MD counterpart combine a “No Excuses” charter school model with a five-day-a-week boarding program. SEED schools rely heavily on data to drive instruction, have a paternalistic culture with high expectations, and have an extended school day and extensive after-school tutoring for students who need support. The middle school curriculum focuses on developing basic skills in reading and math, and the high school uses an intensive college preparatory curriculum. Both the middle and high schools have an extended school day, from 8 a.m. until 4 p.m., and provide students with extensive after-school tutoring as needed.  All students take interim assessments in English and math four times per year. After each assessment, academic and Student Life staff members meet to identify priority issues, and teachers develop lesson plans to “reteach” skills that students have been unable to master.  High school classes are co-ed, students’ time after school is less structured than in the middle school, and students begin more rigorous college-preparatory activities.

From Sunday evening through Friday afternoon, students live on campus, in double bedrooms in same-sex dorms. Students are organized into “houses” of 12-14 students within the dormitories.  The residential staff is separate from the school faculty, although the two groups interact often to discuss student progress. SEED holds community dinners and gives parents the opportunity to serve as tutors during study hall, assist during extracurricular activities, and participate in book clubs.  The Student Life Department aims to develop students’ behavioral, social, and life skills while reinforcing what they are learning in the classroom.

The goal of the middle school curriculum is for all students to enter the high school program performing at or above grade level.  Students are provided with tutoring outside of the classroom and extra periods of instruction and students who need more time to master grade-level skills can take a “growth year.” A readers and writers workshop model is used for language arts instruction, and all students take algebra by the eighth grade.

The high school curriculum is a college-preparatory program; to graduate, students must complete four years of English, four years of mathematics (through at least Algebra II), three years of social studies, three years of science, three years of a foreign language, one and a half years of physical education and health, one year of arts, one half year each of U.S. government and politics, Washington, DC history, and technology, as well as five and a half years’ of elective courses. Students must also take the SAT or ACT college admissions test, apply to at least five colleges or universities, and complete sixty hours of community service. The school offers Advanced Placement courses in English Literature, English Language, U.S. History, Government, and Biology.

SEED aims to promote positive behaviors such as rigorous study habits and self-control, while discouraging “risky” behaviors such as alcohol use and unprotected sex.  The SEED-created Habits for Achieving Life-Long Success (HALLS) program teaches students social and basic life skills, such as decision making and communication strategies, and the importance of taking responsibility for oneself and others. HALLS activities focus on a variety of topics, such as bullying, dating relationships, and appropriate dining etiquette.  Middle school programming is intended to develop and refine social skills that are connected to meeting behavior expectations and routines, such as following instructions the first time they are given, adhering to the school dress code, and learning how to disagree appropriately. Ninth-grade activities aim to develop and reinforce skills and habits necessary for success in high school, such as planning ahead, using anger control strategies, and building strong self-esteem.

The cost of SEED DC was about $39,275 per pupil per year in 2010, compared with $20,523 per student in District of Columbia public schools.

EVALUATIONS OF PROGRAM

Curto, V. E., & Fryer Jr., R. G. (2011). Estimating the Returns to Urban Boarding Schools: Evidence from SEED (Working Paper No. 16746). Cambridge, MA: National Bureau of Economic Research. 

Evaluated population: A total of 221 students from the 2007 and 2008 cohorts participated in this study, with 129 students assigned to participate in SEED DC and 92 students assigned to the control group.

All of the students who applied to SEED were African American, 75 percent were eligible for free lunch, and 12 percent were in special education. Lottery winners had slightly higher baseline reading and math scores, compared with lottery losers, but the differences were not significant. Free lunch status and special education status were balanced among lottery winners and losers. Those in the treatment group were marginally significantly less likely to be English Language Learners than those in the control group; seven percent of SEED group were English Language Learners, as compared with 11 percent of the non-SEED group.  Those in the treatment group were also marginally significantly less likely to live with other legal guardians than those in the control group; four percent of SEED group lived with other legal guardians, as compared with 13 percent of the non-SEED group.

Approach: Assignment to the SEED DC treatment group or the control group was determined by lottery. Because far more females than males applied, separate lotteries were held for males and females, with the result that almost all male applicants were accepted. The treatment group was comprised of students who won the lottery and participated in SEED DC for at least one day.  The control group was comprised of students who entered the lottery but did not win and therefore did not attend SEED DC.  The two sources for data were files at the SEED school and DCPS administrative data on student demographics and outcomes. The data from SEED consisted of applications and demographic data and were matched to the administrative data from DCPS, which included student race, gender, free and reduced-price lunch eligibility, attendance, and math and reading achievement scores in grades three through eight and grade ten.  All SEED DC students took interim assessments in English and math four times per year.

Results: The study found that, after controlling for previous scores and demographic variables, participating in SEED DC led to significant gains in standardized test scores in the seventh and eighth grades, as compared with those in the control group; participating in SEED DC resulted in an effect size of 0.198 standard deviations for reading and 0.230 standard deviations for math.  The study also found that female participants scored significantly higher than male participants in reading; females had an effect size of 0.382 standard deviations, compared with an effect size of -0.138 for males. The study found no significant differences in scores between special education and non-special education students, or between free lunch and non-free lunch participants.  The study found a significantly greater improvement in reading scores for SEED participants who scored below the median score at baseline, as compared with those who scored above baseline; those who scored below the median score at baseline showed an effect size of 0.314 standard deviations, as compared with an effect size of -0.048 standard deviations for those who had scored above the median at baseline. The estimated return on investment in SEED is 4.59 percent, compared with the return on investment in a typical “No Excuses” charter school — without a boarding option — of 18.5 percent.

Untermann, R., Bloom, D., Byndloss, D. C., & Terwelp, E. (2016). Going away to school: An evaluation of SEED DC. New York, NY: MDRC. 

Evaluated Population: A total of 766 students who applied to SEED DC as fifth- or sixth-graders between 2006 and 2011 participated in this study.  At the beginning of the study period, students entered SEED as seventh-graders, but two years later SEED began to enroll sixth-graders and in 2010-2011 students were admitted only as sixth-graders. The students who applied to SEED were 99 percent African-American, roughly three-quarters were economically disadvantaged, and approximately 15 percent were in special education. In the year they applied, just under 50 percent of the SEED group scored at or above ‘proficient’ on the district-wide reading and math exams.  At baseline, the SEED group students were significantly less likely to have been eligible for free or reduced-price lunch than those in the control group; 79 percent of SEED group were eligible for free or reduced price lunch, as compared with 87 percent of the non-SEED group.

About 20 percent of applicants who won the SEED lottery (the SEED group) did not enroll in SEED; a little more than half of these participants enrolled in other charter schools, and the remainder enrolled in public middle schools. Between the second and third follow-up years, roughly 20 percent of the SEED lottery winners in cohorts 1 and 2 who were enrolled in SEED chose to leave and attend a different school.

Approach: The evaluation examined both the implementation of SEED DC and student outcomes, including standardized test scores, high school graduation, attitudes (college aspirations), and behaviors (homework completion, alcohol and drug use).  Analyses drew on separate samples for sixth-grade and seventh-grade entrants to estimate the short-term effects of SEED and drew on the sample of students from the first two years of the study (cohorts 1 and 2 of the seventh-graders) to estimate the longer-term impacts. Because the study followed the students only through the 2013-2014 academic year, too few graduated from high school or enrolled in college during the study period to allow assessment of post-secondary impacts. Short-term academic impacts were measured by standardized test scores in the first two years of follow-up. In the fall of 2013, when most of the respondents were high school seniors or had recently graduated, the SEED DC students and the students in the control group in cohorts 1 and 2 were surveyed about their school experiences.

Results: The study found that for the seventh-grade entrant sample, being offered the opportunity to attend SEED significantly increased students’ math scores, as compared with those in the control group; in the first year, SEED group scores were higher than control group scores by 0.24 standard deviations, roughly equivalent to a 76 percent improvement on top of the typical annual gains. In the second year, the SEED impact on math scores was 0.33 standard deviations, equivalent to one and a half years of typical growth.  The study also found that participating in SEED had a significant impact on students being at or above proficiency level in math. In the first year, 66.3 percent of SEED students were at or above the proficiency level in math, compared with 48.4 percent of those in the non-SEED group. In the second year, 67.4 percent of SEED students were at or above the proficiency level in math, compared with 46.2 percent of those in the non-SEED group

Students in the SEED group were not found to score significantly better than students in the control group in their first year of follow-up on the standard reading exam, but in the second year, the SEED group’s test scores exceeded those of the control group by 0.23 standard deviations, equivalent to one year of typical growth in reading. The study also found that participating in SEED had a significant impact on students being at or above proficiency level in reading in the second year; 55.6 percent of SEED students were at or above the proficiency level in reading, compared with 37.9 percent of those in the non-SEED group.

The study found that for the sixth-grade entrant sample in the first year, being offered the opportunity to attend SEED significantly increased students’ likelihood of being at or above proficiency in math, compared with those in the control group; 49.8 percent of SEED students were at or above the proficiency level in math, compared with 30.2 percent of those in the non-SEED group.  In the second year, SEED group math scores were higher than control group scores by 0.39 standard deviations, and SEED group reading scores were higher than control groups cores by 0.21 standard deviations, equivalent to one and a half years of typical growth.  In the third and fourth years, the study found no significant differences in math or reading test scores or proficiency levels between students in the SEED group and those in the non-SEED group.  The study also found no significant difference in four-year high school graduation between the two groups.

The study found that compared with students in the non-SEED group, students in the SEED group took significantly more years of foreign language courses; 2.35 years for SEED students, compared with 2.01 years for non-SEED group students. The study also found that more SEED group students took physics courses than non-SEED group students; 72.2 percent of the SEED group took a physics course, compared with 55.5 percent of the non-SEED group.

The study found that significantly more students in the SEED group went on field trips and participated in student leadership groups or student government, and received tutoring or extra help from a non-teacher adult, compared with students in the non-SEED group. The study also found that significantly more of the SEED group students had an internship, compared with those in the non-SEE group.  The study also found that SEED group students reported having more orderly classrooms and more academically motivated peers than those in the non-SEED group. The study found no significant differences in measures of weekday home living environments between the two groups.

The study found that students in the SEED group spent significantly more hours per week doing homework than students in the control group; SEED group students spent, on average, 9.14 hours per week doing homework, compared with 5.12 hours for non-SEED group students. Students in the SEED group reported marginally significant lower tobacco use in the past 30 days than control group students. However, the study also found that students in the SEED group reported significantly more frequent recent risky behavior (e.g., skipping school, arguing with parents, or hitting someone), compared with control group students. SEED group students scored significantly lower than control group students on scales designed to measure “grit” or perseverance.

The study found no significant impacts on student reports of development of rigorous study habits or organizational skills.

The study found that being accepted to SEED DC did not have a significant impact on students’ probability of four-year high school graduation.

SOURCES FOR MORE INFORMATION

Web Site: http://www.seedschooldc.org/

Contact Information: 4300 C Street, SE, Washington, DC 20019  Tel. (202) 248-7773

References

Curto, V. E., & Fryer Jr., R. G. (2011). Vilsa E. Curto and Roland G. Fryer Jr., Estimating the Returns to Urban Boarding Schools: Evidence from SEED (Working Paper No. 16746). Cambridge, MA: National Bureau of Economic Research Working Paper No. 16746 (2011).

Untermann, R., Bloom, D., Byndloss, D. C.,  & Terwelp, E. (June, 2016). Going away to school: An evaluation of SEED DC.  MDRC, New York, NY: MDRC.

KEYWORDS: Adolescents (12-17), Youth (16+), Young Adults (18-24), Middle School, High School, Males and Females (Co-ed), Urban, School-based, Attendance, Reading/Literacy, Mathematics, Academic Achievement/Grades, Academic Motivation/Self-Concept/Expectations/Engagement, College Enrollment/Preparation, High School Completion/Dropout, Social Skills/Life Skills, Self-Esteem/Self-Concept, Dating Violence, Bullying

Program information last updated on 10/27/2016.

Subscribe to Child Trends

Short weekly updates of recent research on children and youth.