Preliminary data from a groundbreaking study of Chapter 1 indicate that the federal compensatory-education program has had little success in improving the achievement of the educationally deprived children it is intended to serve.
Data from two testing cycles, collected for the “Prospects’’ longitudinal study, show no decrease in the performance gap between Chapter 1 participants and their more advantaged peers.
They also show no significant difference in achievement between Chapter 1 students and similar students who did not participate in the program. The findings were cited in less detail in the National Assessment of Chapter 1 released earlier this year. (See Education Week, Feb. 24, 1993.)
The preliminary Prospects report also paints a detailed statistical picture of students in high-poverty and low-poverty schools.
Expected To Be Influential
As the first major longitudinal study of Chapter 1 since it was created in 1965, the five-year effort, which is being conducted by Abt Associates Inc. of Bethesda, Md., is expected to have a strong influence on the shape of the program.
A final report is not due until 1997. But lawmakers asked for an interim report in time for this year’s rewrite of the Elementary and Secondary Education Act, of which the $6.3 billion Chapter 1 program is the cornerstone.
The 396-page interim report, which was transmitted to Congress in July and recently released to the public, focuses on the characteristics of students who receive Chapter 1 services and their schools.
The study surveyed the achievement and characteristics of more than 28,000 1st, 3rd, and 7th graders in the spring of 1991 and tested them again in the spring of 1992. The researchers plan to track the students through the course of the study.
Measuring Relative Progress
The best preliminary results posted by Chapter 1 students were those of children tested in reading in the 7th and 8th grades, who showed a growth of 1.9 “normal curve equivalents’’ between 1991 and 1992.
The N.C.E. scale is designed to show students’ progress relative to other students’. A student whose N.C.E. score does not change has learned during the year, but has not gained any ground on his peers.
The reading scores of Chapter 1 students tested in the 3rd and 4th grades dropped from 35.8 to 34.4 during the period.
In comparison, the N.C.E. scores of students participating in other compensatory reading programs increased from 37.5 to 38.6, while the scores of students not participating in any such program increased slightly, from 57.3 to 57.5.
In mathematics, the test scores of almost all the groups declined between 1991 and 1992.
Scores dropped from 36.7 to 34.8 for Chapter 1 3rd graders, from 37.2 to 34.9 among other compensatory-education students, and from 56.6 to 55.6 for 3rd graders not receiving services.
For 7th graders, math scores dropped from 34.2 to 32.9 among Chapter 1 students and from 53.4 to 52.2 among students who did not participate in compensatory programs, but increased from 26.2 to 28.0 among 7th-grade students receiving compensatory math instruction in a program other than Chapter 1.
The study’s authors also say that the significant difference in overall N.C.E. levels between Chapter 1 students and those not receiving services implies that “the two groups are probably making gains in qualitatively different skills,’' as N.C.E. gains at higher levels are more likely to involve higher-order skills.
In an effort to control for socioeconomic factors that may affect the scores of Chapter 1 students and nonparticipants, the authors manipulated their data to provide a more even demographic match--a technique that will be used extensively over the course of the study.
But the results “did not fundamentally alter the conclusions derived from the raw gain scores that there is little or no evidence to suggest any differential gain in achievement between students receiving compensatory services and those not receiving such support,’' according to the report.
The study also provides ample support for the contention that educational performance is dramatically affected by the concentration of poor students in a school.
It found that students in low-poverty schools score 50 percent to 75 percent higher on reading and math tests than students from high-poverty schools do, and that the average achievement among all students in high-poverty schools is about the same as that of Chapter 1 students in low-poverty schools.
“The relative annual gains made by students in low- and high-poverty schools are approximately the same, leaving the achievement gap between these students unchanged,’' the authors conclude.
Bilingual-education advocates say that many language-minority students are excluded from Chapter 1, but the report notes that such students “constitute a significant portion of the children being served by Chapter 1,’' and are overrepresented in the program relative to their proportion of the population.
The study found that 92 percent of limited-English-proficient 1st graders, 89 percent of such 3rd graders, and 84 percent of such 7th graders receive some language help.
The study also reports that:
- Children in high-poverty schools are more likely than those in low-poverty schools to live with a single or unemployed parent, or a parent without a high school education. They are less likely to read at home or use the library. They get lower grades, miss more days of school, and are more likely to have developmental problems.
- The most prevalent method of Chapter 1 instruction remains pull-out programs, although students in high-poverty schools are more likely than those in low-poverty schools to receive Chapter 1 services in their regular classrooms.
- Chapter 1 teachers are more likely than other teachers to have an advanced degree, and typically have more experience.
Copies of the report can be obtained for free by calling the Education Department’s publications office at (202) 401-3132, or Abt Associates at (301) 913-0500.
A version of this article appeared in the November 24, 1993 edition of Education Week as Chapter 1 Fails To Spur Gains, Data Indicate