Archive for the ‘magnet schools’ Category

Principals and Charter Competition

Wednesday, April 6th, 2011

Many advocates for school choice argue that the competition created by giving parents multiple options of schools for their children will improve education in traditional public schools.  Does the research support this claim?  The latest research brief, How Do Principals Respond to Charter School Competition?, released by the Choice Center examines the factors that contribute to principals’ (including traditional public school, private school, and magnet school leaders) perceptions of competition from charter schools  and the effect this perception has on promoting changes in leadership behavior. 

Areas studied include proximity of charters and the relation to perception of competition, principals’ allocation of time and financial resources in response to perceived and actual competition, and their ability to recruit teachers and students.  The brief also includes a clear and concise description of the key study variables mentioned above as well as policy and research implications.  For more on the findings, check out the full brief!

This brief is based on a paper by Marisa Cannata which is included in our book, School Choice and School Improvement, released in March by Harvard Education Press.  For more information about the book, check out the table of contents and selected appendices or the Harvard Education Press webpage for the book!

School Choice Week – Student Composition and Peers

Friday, January 28th, 2011

As National School Choice Week draws to a close, the final topic of the week is student composition and peer effects.  One of the points often made by opponents of school choice is that the student body of schools of choice has a different make-up than traditional public schools in the area.  Peer effects, the impact of other students at a school on a student’s achievement, should also be considered when examining the effects of school choice on student composition. So, what does the research tell us about these topics?

One concern raised about charter schools has been that they will take the best students and cause traditional public schools to retain the lower-performing students making it harder for them to improve.  This perception is examined in Do Charter Schools “Cream Skim” Students and Increase Racial-Ethnic Segregation? by following students as they moved to charter schools in seven sites around the country.  Similar concerns about vouchers are explored in Do Vouchers Lead to Sorting Under Random Private School Selection? Evidence from the Milwaukee Voucher Program as well as how a voucher program could be designed to limit sorting by ability but students may still be sorted by parental self-selection.  Racial and ability sorting are not the only types of sorting to cause concern.  One that is often overlooked is sorting by gender.  The Gender Gap in  Charter School Attendance finds that charter schools enroll a significantly higher fraction of girls than boys.  The paper goes on to explore possible causes for this.  Sorting of any type may have consequences.  In Magnet Schools and Peers: Effects on Mathematics Achievement, Ballou uses the lottery outcomes for enrollment to measure the effects of peers and finds that peers race and parental income have a substantial impact on achievement.  Sorting  by race, socio-economic status or gender could have a real impact on the peer effects of a school and in turn student achievement.

Do not let the end of National School Choice Week be the end of your research.  Continue to use available research to inform your opinions, and potentially policies and advocacy, on school choice.  Check back often for updates on what is happening in the world of school choice and how it relates to the research done here by the National Center on School Choice.

For more information on racial/ethnic and ability sorting, check out these articles:

White Parents, Diversity and School Choice Policies: Where Good Intentions, Anxiety and Privilege Collide

Parental Choice in the Netherlands: Growing Concerns about Segregation

Charter Schools in North Carolina

School Choice Week – Student Achievement

Monday, January 24th, 2011

January 23rd – 28th is School Choice Week.  During this focus on school choice, take some time and consider what research shows about the impacts of school choice.  Each day this week, this blog will highlight a different aspect of school choice research.

One of the biggest questions to consider when analyzing school choice is: What is the impact on student achievement?  This question spans various types of school choice including magnet schools, charters, and vouchers.  In New York City Charter Schools: Who attends them and how well are they teaching their students?, Hoxby and Murarka report that charter schools have a positive impact on the academic growth of their students.  In a study of charter schools in North Carolina, Bifulco and Ladd report that, on average, charter schools have had a negative impact on student achievement overall.  Another study of charters, this time in Idaho, found mixed results of the effects of charter schools, varying dependent on the type of analysis.  Witte, Cowen, Fleming, and Wolf examined the impact of the Milwaukee Parental Choice (Voucher) Program and found that although the voucher student panel achievement tended to be higher, it was not statistically significant.  In reviewing the literature on magnet schools, Ballou found that the results from studies of the effects on student achievement are also mixed.  This is just a sampling of the research papers available on our website that address student achievement and school choice.

All of these studies and more can be found be visiting the research page of our website!  Take some time this week to become more informed about the research findings concerning school choice!

Black Box Project Featured at IES Conference

Thursday, July 1st, 2010


A team of NCSC scholars presented findings from the Opening the Black Box project Wednesday at the Institute of Education Sciences’ Annual Research Conference in Washington. The project is conducting ongoing investigations into which practices and conditions in schools are most associated with positive student achievement gains by linking responses to principal and teacher surveys about their schools to test scores of their students. The study includes thousands of surveys from hundreds of schools, allowing researchers to identify differences between schools of choice and traditional public schools in areas ranging from levels of innovation to teaching practices to characteristics of principals and teachers, and measure how these factors affect student achievement.

The NCSC team shared the 90-minute session with two research groups that recently released high-profile school choice studies: The final report on impacts of the Washington, D.C., voucher program, which was led by researchers from the University of Arkansas, and a multi-state randomized controlled trial of charter schools, led by analysts at Mathematica Policy Research. All papers had produced interesting findings and the presentations generated lively discussion. For the NCSC group, it also facilitated a discussion around how to deal with the methodological challenges surrounding school choice research.

The two other studies in the session included only schools or programs that used lotteries, enabling the researchers to create experimental conditions with students randomly assigned to the treatment or control group based on whether they won a voucher or charter school seat in the lottery. Such conditions are highly desirable because they remove any question of whether the groups differ in some way related to their academic performance – e.g. one group doesn’t perform better because it was chosen by more motivated students, for example. But random assignment is rare in education settings and difficult to do in school choice research unless schools are oversubscribed, forcing researchers to find other ways to remove bias from their analyses. The NCSC team dealt with the problem by matching schools of choice with traditional public schools on specific sets of criteria, including geographic proximity, demographics, and grade level configurations. The team concentrated on finding the best matches possible for each set of criteria to provide comparison groups for an array of research questions.

Feedback at the conference focused on whether defining a “well-matched” pair of schools can be restricted so much that we may miss out on part of what it means to be in a school of choice. For example, only matching charter schools to traditional public schools that are of the same size may be inappropriate given that part of what makes a charter school unique is that they tend to be smaller. This was highlighted by one finding of the DC voucher study that indicated that the effect of winning the voucher lottery on student’s likelihood of graduating from high school was mediated by the fact that the voucher gave students access to smaller schools. Further, the school characteristics that define a well-matched pair may depend on the specific research question or perspective being sought. Examining the choice set available to parents about where to send their child or to teachers when looking for a job may require matching on different criteria than an analysis designed to help policymakers revise school choice laws. A lesson for the project going forward is that while as researchers we will still pursue close matches, the criteria used to create those matches could differ for each study. Watch this blog for more postings about productivity from the Black Box study.

Read papers included in the NCSC presentation:



Learn more about the groundbreaking NCSC study Opening the Black Box in Choice and Regular Public Schools: A Study of What Makes Schools Work

More Parents are Sending Children to Schools of Choice

Thursday, April 29th, 2010
New Report Covers Public and Private Schools

School choice is growing in popularity. The percentage of parents in a national survey reporting that their children attend public schools of choice increased from 11 to 16 percent between 1993 and 2007, according to a new report by the U.S Department of Education. The popularity of private schools rose, too, although not as sharply. The percentage of parents sending their children to private schools, both religious and nonsectarian, during that period increased from 10 to 12 percent of the total.

The survey also measured parent satisfaction with schools. Between 2003 and 2007, the percentage of students attending their parents’ first-choice public school rose from 83 percent to 88 percent. Slightly more than a quarter of public school students had parents who reported moving to a neighborhood for the school, regardless of whether it was an assigned or chosen school. A majority of parents reported satisfaction with their child’s school, regardless of the type, though the satisfaction tended to be highest for public schools of choice and private schools.

The findings are based on responses in the National Household Education Survey, which asked thousands of parents about their children’s education in 1993, 1996, 1999, 2003 and 2007. For the first time in 2007 the report breaks out information on enrollment in charter schools and characteristics of homeschooled students. That year, 2 percent of students were enrolled in charter schools, a majority of them in cities. That increase is due in some part to rapid growth in the sector. Between 2002-03 and 2006-07, the number of charter schools increased from nearly 2,600 to more than 4,100 and the number of jurisdictions allowing them grew from 36 to 41. Homeschooled students, who made up nearly 3 percent of total enrollment, according to the 2007 survey, were more likely to live in rural areas or suburbs than in cities.

In 2007, black and Hispanic students were more likely than whites to enroll in public schools of choice (24, 17 and 13 percent, respectively), while low-income students were more likely than wealthier peers to choose those schools. Students at chosen public schools also were more likely to live in cities than suburbs or rural areas. For private schools, parents’ wealth and education made a difference. Higher percentages of parents with more income and education chose religious and nonsectarian private schools than did parents with lower levels of income and education. The report contains many more breakdowns of demographic characteristics for each year of the survey.

The telephone survey asked parents whether their children attended public or private schools. If the answer was public, they were asked if they chose the school or it was assigned. If they said private, the interviewer asked if it was religious or nonsectarian.

The report notes that school choice offerings have expanded greatly over the last several decades and that parents in some communities can now choose from an array of public school options beyond their assigned neighborhood school, including charter schools, intra-district and inter-district choice plans and magnet schools, as well as options for sending their children to private schools, aided in some areas by publicly financed vouchers. They also can choose to school their children at home.

Read the full report “Trends in the Use of School Choice: 1993 to 2007.”


You Might Find These NCSC Research Projects Particularly Interesting


The NCSC is funded by a 5 year, $13.3 million grant from the U.S. Department of Education's Institute of Education Sciences. Its lead institution is Vanderbilt University in Nashville, Tennessee. The center is housed on the campus of Peabody College, one of the nation's top graduate schools of education.