This article presents data on four schools in the Minneapolis Public School District (MPS) implementing the Title I schoolwide program. The study from which the data were drawn aims to show how Title I students—and the resources available to service them—are distributed in MPS.

Since the enactment of the Elementary and Secondary Education Act in 1965, federal Title I has provided additional resources to schools with high numbers of students from low-income families. Although the federal government appropriates nearly $6 billion each year in compensatory education through Title I, these federal resources have produced mixed results in inner-city schools. In 1988, in an effort to address these concerns, the U.S. Congress encouraged the implementation of schoolwide projects in schools in which at least 75% of the students are from low-income backgrounds. The aim of this implementation program was to reduce the "categorical" character of Title I, thereby enhancing programmatic coordination in order to benefit the entire student population.

The Title I schoolwide project in MPS was adopted during the 1990-91 school year to accommodate a growing population of low-income, educationally disadvantaged students in grades K-3. Prior to 1989, no schools in the district met the 75% poverty requirement for schoolwide projects. However, Minneapolis is currently experiencing trends of rising poverty, increasing ethnic diversity, and population shifts to the suburbs. These socioeconomic changes have been accompanied by increases in the percentage of children with special needs in MPS. The district estimates that there are approximately 1,000 newly enrolled students at the K-3 level each year, and that 95% of these students are eligible to receive Title I services.

Project Findings

The following results are based on an analysis of normal curve equivalent (NCE) gains in the vocabulary, reading, and mathematics scores of students in each of the four project schools (AL, AT, BI, and CH) as compared to the gains of Title I students in MPS schools without schoolwide projects. Gains were measured according to the scores of 44,464 students on the California Achievement Test, which were collected during school years 1989-90 to 1992-93. Of the two cohorts of students used in the study, each was subdivided according to type of school, race, poverty, or lack of poverty. Analysis is also based on site visits conducted during the 1993-94 and 1994-95 school years.

School AL: In school AL, the general tendency in vocabulary and reading was positive and showed incremental growth. The most significant gains were evinced by white students from low-income backgrounds in reading and vocabulary . Although low-income students did not perform as well as nonpoor students, they did show significant score gains. Overall, the school exhibits social-and race-equalizing effects, i.e., gains made by low-income African-American students tend to equal those made by nonpoor white students.

The success of the AL program may be attributed to several factors. First, the program is enhanced by two fully equipped computer labs funded by the Title I program. The computer-assisted program not only fo sters collaboration between teachers and students, but also facilitates individualized instruction, accommodates different ability levels, easily identifies students' strengths and weaknesses, and emphasizes skill building. In addition, the focus of the computer program is on basic skills learning, which may be directly applicable to the tests used to measure achievement in the early grades. Finally, the school promotes a racial balance in teacher assignments by actively recruiting teachers who reflect the racial makeup of the student body.

School AT: In school AT, gains were made in reading and vocabulary achievement scores. There were significant reading gains among low-income and nonpoor white students, as well as nonpoor African-America n students. Math scores, on the other hand, were stable or showed some decline, depending on student characteristics. There were differences in achievement scores between African-American and white students, but not between low-income and nonpoor students.

The gains in reading at school AT are consistent with the program's emphasis in that area. Several program components may contribute to this success. First, the school has developed a home reading program in which students are encouraged to read books at home, and are given both individual and classroom rewards for meeting certain reading goals. Second, the use of instructional staff to deliver Title I services and the exceptional expertise of one outstanding educational assistant may also be important factors in the success of the program. Finally, the school has an "integration music program" that teaches academics through music and includes skills work and reading through songs. The music program provides additional basic skills instruction and reinforces the reading and vocabulary curriculum of the regular classroom program.

School BI: There were few consistent trends in gain scores at school BI. There were gains in vocabulary for both low-income and nonpoor white students, but significant losses for nonpoor African-American students in school BI. Achievement scores show increasing poverty and racial effects; that is, the gap between the performance of low-income and nonpoor students, and between achievement scores of African-American and white students, have both increased.

The reason for the lack of significant and consistent gains may be twofold. First, most program changes in school BI occurred before the advent of the schoolwide program, and thus these effects may have been extant before project implementation. Second, the importance that the school places on monitoring and tracking student progress—e.g., counting words read per minute and graphing them on a reading progress chart—may emphasize fluency at the expense of comprehension. Although systematically charting school progress is important, the usefulness of these techniques, especially for African-American students, is questionable, as may be evinced by the African-American nonpoor students who lost substantial ground in vocabulary achievement from one year to the next.

School CH: Gain scores in reading at school CH remained stable over time, with no major fluctuations. Math scores, however, showed some significant losses over time, particularly for the low-income students. For example, the white nonpoor students did not make progress in math during the first year of the program, and the math scores among low-income African-American students continued to remain low. In other words, students' math skills appeared to wane over time decreasing from previous years.

These results are consistent with a program that was not modified with the introduction of the Title I schoolwide project. Since its implementation, the emphasis of the program at school CH has been to maintain current staffing levels, and little attention has been directed to curriculum changes, assessment, instructional practices, or staffing patterns other than the use of Title I teachers and educational assistants. The size and organization of the school may also contribute to these results.

CH was the largest school in the study sample, and the largest in MPS, with close to 1,000 students in grades K through 3. In addition, organization within the school may be problematic; the school was constructed without walls and, although dividers have been set up to designate classrooms, there are no self-contained classrooms. Furthermore, the practice of grouping two classes together may have a negative impact on student achievement, as the wider ranges of abilities housed in single classrooms may decrease instructional time.

Conclusions and Implications

• It is noteworthy that, in the Title I schoolwide project schools, low-income African-American students are not al ways the lowest-achieving students. This finding stands in contrast to the district as a whole, where low-income African-American students consistently achieve at the bottom compared to nonpoor African-American, low-income white, and nonpoor white students. This finding suggests that the schoolwide Title I programs are contributing significantly to the achievement of low-income African American students.

• Low-income students in project schools make significant gains when compared to low-income students in nonproject schools. Furthermore, the size of the gain intensifies over time, i.e., the longer a student is in a schoolwide project school, the greater the performance gain.

• Two explanations may account for these race and poverty effects: (a) the social and economic differences in Title I schoolwide project schools may be mitigated by the project itself; and (b) the potential inclusion of all students in the schoolwide program may decrease the stigma attached to Title I students.

• These results suggest that schoolwide projects offer advantages to low-income and minority students when compared to students in non-schoolwide project schools. This study also indicates that three strategies seem to work in improving the performance of Title I students in schoolwide projects: (a) direct teacher knowledge of student skills; (b) the use of assessments to inform student progress leading to the development of strategies to deal with particular student problems; and (c) an emphasis on reading instruction and a program that provides a basic educational foundation that is reinforced throughout the curriculum.