The research approach is a combination of case study methodology and benchmarking of "best practices." Used in businesses worldwide as they seek to continuously improve their operations, benchmarking has more recently been applied to education. Benchmarking is a structured, efficient process that targets: key operations and identifies promising practices in relationship to traditional practice, previous practice at the selected sites (lessons learned), and local outcome data. The methodology is further explained in a background document,¹⁹ which lays out the justification for identifying promising practices based on four sources of rigor in the approach:

• Theory and research base;

• Expert review;

• Site evidence of effectiveness; and

• Systematic field research and cross-site analysis.

The steps of the research process were: defining a study scope, seeking input from experts to refine the scope and inform site selection criteria, screening potential sites, selecting sites to study, conducting site visits, collecting and analyzing data to write case reports, and writing a user-friendly guide.

Site Selection Process

For this guide, over 400 charter secondary schools from 25 states and Washington D.C. (out of 40 states that have charter school laws) were considered. These states had the largest numbers of charter schools and charter school legislation old enough to have schools beyond the start-up phase. Based on recommendations from the advisory group, information from state department staff and state association leaders, and review of achievement data, the initial list of 400 schools was narrowed to 70 charter high schools that served students through grade 12, had already graduated a class of seniors so that graduation and college-going data were available, and had met AYP for at least the past two years.

The list of 70 schools was narrowed to 26 by selecting schools that either demonstrated high academic achievement on state testing, or outperformed comparable local schools, or demonstrated increasing achievement with a predominantly low-income or minority population of students. Information about program features and additional outcome data were gathered on these 26 schools, using phone interviews to fill gaps in information. From this group of 26 schools, eight schools were selected as case study sites based on the compiled information and criteria ratings in a screening matrix. (See next section for more on criteria.) Demographic variation, a range of promising practices, and geographic location, along with achievement data, were all considered in the final site selection.

Site Selection Criteria

A cross section of schools were selected to highlight secondary charter schools successfully meeting the needs of traditionally underserved populations of students (e.g., low income, special education, African-American and Latino students), with strong high school programs serving a range of grade configurations (e.g., 9–12, 5–12, 7–12) and in a range of geographic locations (i.e., seven states including rural, urban and suburban schools), all making academic achievement gains. Schools were selected based on the following criteria, prioritized by the advisory group as key issues for consideration.

Demographic Criteria

Many schools demonstrated that they were working hard to educate students who have been largely underserved in traditional public schools. The schools that were selected had two or more of: 40 percent or more free and reduced-price lunch, 40 percent or more African-American and Hispanic students, 20 percent or more special education students, or rural location.

Achievement Criteria

Schools selected met AYP targets for at least two consecutive years, including the most recent year for which data were available. Researchers looked for schools that scored at least a baseline of the 50th percentile in math or reading on state standardized tests with demonstrated evidence of continued improvement over several years, or schools that were consistently high achieving in the 90th percentile range annually. Data from state Web sites and SchoolMatters.com provided achievement information.

Achievement Gap Criteria

Researchers looked for additional evidence that schools were making progress eliminating achievement gaps. A school was considered to be narrowing the achievement gaps if, internally, gaps among students of different races and socioeconomic backgrounds were closing over time or if the school was demonstrating higher achievement for low-income, minority, or special education students in comparison with a similar population of students in its local district public schools.

Study Framework and Data Collection

A conceptual framework was developed to guide the study of the selected sites. While many things happen at a busy school site, each case study needed to focus on those practices most likely to contribute to a school's success. The framework used in this study was an adaptation of the framework used in the previous guide on charter schools in this Innovations in Education series,²⁰ which was derived from the research literature on charter schools and on organizational effectiveness. The major categories in the framework were mission-driven school, school operations and educational program, external partnerships, and governing for accountability. The framework for this study of charter high schools additionally highlighted issues of particular salience in secondary schools, for example, issues related to transitions both into the school from the lower grade levels and out of the school to college or work. These key secondary issues were determined through a review of the high school reform literature and the input of the researchers on the project's advisory group. Sites also offered their own views of the key factors in their success, which constituted additional input to the analysis.

A site visit was conducted at each school to gather the information for this guide. Each site visit lasted for one or two days and included informal observations throughout the school, attendance at events, and interviews. The primary source of data was interviews with a variety of role groups, including students, parents, teachers, board members, administrators, and school partners. An interview protocol was developed based on the study framework and adapted to each role group. That is, separate but overlapping sets of questions were developed for teachers, administrators, parents, etc. Most interviews were tape-recorded with key interviews later transcribed for more detailed analysis.

Documents from each school served as an additional source of information. Collected during the site visit, these documents included such items as school schedules, sample assessments, lesson plan forms, teacher planning protocols, newsletters, application forms, brochures, charter plans, and report cards. Principals and executive directors also completed a standard form to facilitate consistent compiling of school demographic and outcome information.

Analysis and Reporting

A case report was written about each site, and reviewed by site administrators for accuracy. From these case reports, artifacts, and transcripts of interviews, the project team identified common themes that contributed to success across the sites. This cross-site analysis built on both the research literature as reflected in the study scope and also emerging patterns in the data.

This descriptive research process suggests promising practices—ways to do things that other educators have found helpful, lessons they have learned—and practical "how-to" guidance. This is not the kind of experimental research that can yield valid causal claims about what works. Readers should judge for themselves the merits of these practices, based on their understanding of why they should work, how they fit the local context, and what happens when they actually try them. Also, readers should understand that these descriptions do not constitute an endorsement of specific practices or products.

Using the Guide

Ultimately, readers of this guide will need to select, adapt, and implement practices that meet their individual needs and contexts. Schools coming together in learning communities may continue the study, using the ideas and practices from these sites as a springboard for their own action research. In this way, a pool of promising practices will grow, and schools can support each other in implementation and learning.