Project Description: Success for All (SFA) developed and evaluated a program designed to use the best available information on effective programs for disadvantaged students in a comprehensive reform of the urban elementary school. Rather than focusing on remedial services after students fall behind, the approach is to provide intensive services in the early grades (pre-K through 3rd grade) and improved classroom programs in the later elementary grades for all but the most academically handicapped students. The major components of the program include one-on-one tutoring by certified teachers, regrouping and reduction in class size to accommodate whole-class instruction for reading, eight-week curriculum assessments to see that all children are making adequate progress, the implementation of half-day preschool and full-day kindergarten programs that emphasize building of language and pre-literacy skills, a family support team to enlist parental support in the education and health of their children, a full-time project facilitator to coordinate all activities, and a building advisory committee to recommend changes in the program.
The project includes four activities:
(1) a longitudinal evaluation of SFA (Activity 20001);
(2) an evaluation of SFA with language minority students ( Activity 20002);
(3) an evaluation of the beginning reading component of SFA ( Activity 20003); and
(4) an investigation of the Building Advisory Committee in SFA (Activity 20004).
Project Director: Slavin, Robert E.
Activities:
Activity ID: 7111-20001
Name of Activity: Longitudinal Evaluation of Success for All
Description of Activity: Success for All developed and evaluated a program designed to use the best available information on effective programs for disadvantaged students in a comprehensive reform of the urban elementary school. Rather than focusing on remedial services after students fall behind, the approach is to provide intensive services in the early grades (pre-K through 3rd grade) and improved classroom programs in the later elementary grades for all but the most academically handicapped students. The major components of the program include one-on-one tutoring by certified teachers, regrouping and reduction in class size to accommodate whole-class instruction for reading, eight-week curriculum assessments to see that all children are making adequate progress, the implementation of half-day preschool and full-day kindergarten programs that emphasize building of language and pre-literacy skills, a family support team to enlist parental support in the education and health of their children, a full-time project facilitator to coordinate all activities, and a building advisory committee to recommend changes in the program.
This activity is a longitudinal evaluation of Success for all (SFA) as implemented in five schools. The purpose of the evaluation is to assess the long-term impact of the program on student achievement, referrals to special education, and retentions. In two of the schools, a full version of SFA is in its fourth or fifth year of implementation; in the three remaining schools, a less costly version of the program with fewer elements is in its fourth year of implementation.
Program implementation began in the Baltimore City Schools under the auspices of Chapter 1 and other federal funds prior to the initiation of the Center. Center activities were designed to assess the impact of the program on student achievement and on such variables as referrals to special education, retentions, and self-esteem through a controlled experiment. In Year 1, measures were taken on all students in grades pre-K to 3. In year 2, the measures were extended to grade 5 and in each subsequent year a grade level was added. The program's use has expanded into approximately 200 schools in 37 districts in 19 states. The project is continuing to analyze 6th and 7th year effects.
In the course of this activity, evaluations of Success for All were extended to cover 15 schools in seven states, including the original five sites in Baltimore.
Study Design: Each Success for All (SFA) school is matched with a comparison school that does not have the program, and individual children in SFA schools are matched with children in non-SFA schools. Outcome measures are examined for all students in grades pre-kindergarten to six. Analyses of variance are used to compare matched experimental and control students. Additional analyses are performed to determine the effects of the program on high, average, and low achievers.
Unit of Analysis: The unit of analysis is the individual student.
Generalizability: The findings are based on a purposive sample of SFA and matched control schools; thus results are only generalizable to the particular high poverty schools involved in the study.
Sample Description: The five SFA schools are located in Baltimore City, a poor urban school district with a high percentage of black students. One of the SFA schools in the longitudinal evaluation is the poorest school in Baltimore City and almost entirely black. The four other SFA schools in the longitudinal evaluation are among the 30 poorest schools in Baltimore City, with at least 75% of the students receiving free lunch. The five comparison schools, also located in Baltimore City, are matched with SFA schools on the basis of historical achievement, percent of students receiving free lunch, racial makeup, and other variables. Within schools, individual children are matched on the basis of standardized test scores and age.
Dependent Variables: The specific outcome measures are: (1) California Achievement Tests -- Reading and Mathematics (Grades 1-3); (2) individually administered measures of reading scales from the Woodcock and Durrell informal reading inventories (Grades K-3); (3) writing samples (Grades K-3); (4) the Test of Language Development (TOLD) (Grades Pre-K and Kindergarten); and (5) information from school records on attendance, retention/promotion, special education referrals, and disciplinary referrals.
Independent Variables: The independent variable is the treatment (SFA program) versus control (matched non-treatment comparison) condition.
External Variables Controlled: The school and individual matching procedures in this study control for student achievement, as measured by standardized test scores and student age, and for school effects, as measured by historical achievement and poverty level.
Statement of Finding(s): The students in Success for All, at all grade levels and in each succeeding year of the study, outperformed control school students in reading by significant amounts.
Description of Finding(s): Evaluations of 15 schools in seven states find that the Success for All program significantly increases reading performance across the sites. Schools using Success for All are significantly improving the reading performance of disadvantaged students in Baltimore, Philadelphia, Charleston, Memphis, Montgomery, Fort Wayne, and Caldwell, Idaho.
Sites and Schools
Baltimore, Maryland. As of Spring 1992, the original Success for All school had implemented the program for five years and four additional schools had done so for four years. Students in the Success for All schools are reading significantly better than students in control groups at all grade levels. The effects are particularly large for students who are in the lowest 25 percent of their classes. Effects are also found on standardized tests administered by the district, on attendance, and on retention in grade.
More recent data show that the positive effects of Success for All continue through the fifth grade. In 1993, fifth-graders in the program scored a full grade level ahead of controls on individually administered tests, 75 percent of a grade equivalent ahead on standardized CTBS tests, and substantially ahead on the Maryland School Performance Assessment Program in reading, writing, and social studies.
Philadelphia, Pennsylvania. One elementary school began the program in 1988; two others began in 1991. The first school serves a population that is majority Southeast Asian; the other two schools serve African-American students. In the first school, Asian students' reading skills exceed those of comparable students in a control school by more than a full grade equivalent in first and second grade; third and fourth graders exceed their controls by six and five months, respectively. Non-Asian students at the school also outperform their control counterparts.
In the two schools that began Success for All in 1991, first graders are reading above grade level; the lowest achieving 25% of Success for All students are about at grade level on average.
Charleston, South Carolina. One school began Success for All in the 1990-91 school year. This is the only Success for All school being evaluated that is not a Chapter 1 school; thus its funding is lower than most Success for All schools but its population is relatively less needy. This school has the highest mean reading level of all the Success for All schools and is four months ahead of its control school.
The evaluations in Baltimore, Philadelphia, and Charleston are being conducted by CDS researchers. The following evaluations are being conducted by an independent evaluation team at Memphis State University led by researchers Steven Ross and Lana Smith.
Memphis, Tennessee. One school, nearly 100% free lunch and serving an African-American student body, began the program in 1990. The 1993 results show that first and second graders are reading well ahead of their controls, with second grade low achievers doing particularly well, exceeding the progress of control low achievers by almost nine months. A 1992 evaluation also found positive effects for Success for All first and second graders on the Tennessee Comprehensive Assessment Program Reading Comprehension Scale.
Fort Wayne, Indiana. Two predominantly African-American schools, left out of a magnet school plan created as part of the district's desegregation efforts, were given additional funds to improve student achievement. They combined these funds with Chapter 1 money to begin Success for All. Students in the two schools are reading above grade level, almost three months ahead of control students; much of the gain is attributable to gains by the lowest achieving 25% of the students.
Caldwell, Idaho. One rural school, the first to be evaluated (although several other rural districts are implementing the program), began the program in 1991. The implementation was considered exemplary. The 1992 evaluation found first graders reading at an average grade equivalent of 2.22 -- but these high levels represented no difference between Success for All and controls, not even for the lowest achieving students. The evaluators note that the control school was an "extraordinary place with an exceptionally able principal and staff in a new facility."
Success for All students were then contrasted with control students in a more comparable school that used Reading Recovery, a first-grade tutoring program. The first grade students scored similarly in both schools, but non-tutored and special education students in Success for All scored significantly better than their matched comparison students.
Montgomery, Alabama. Two schools began the program in 1990-91. Kindergartners who entered in the 1991-92 school year were assessed at the end of first grade in spring of 1993. These children in both schools, compared to controls, registered large effect sizes, averaging +1.32 (five months ahead of controls). Among the lowest 25%, Success for All students averaged a grade equivalent of 1.46, a huge effect size of +2.86 compared to the reading of control students.
Other Success for All schools, although not undergoing rigorous evaluations like those reported above, contribute further evidence to show that an effective program can be successfully replicated.
In Charleston, West Virginia, the elementary school with the highest African-American population and poverty rate in the district began the program in 1990. No other school in Charleston has a similar population; thus no control group was used in the evaluation. The school has shown steady increases in standardized test scores after the introduction of Success for All.
In Wichita Falls, Texas, a school that was historically the lowest achieving school in the district began the program in 1991. In fall, 1992, the percentage of third graders meeting minimum expectations in reading increased from 48% to 70% while the district percent declined by three points. The school's students increased from 8% to 53% in the percentage meeting minimum expectations in writing while the district as a whole gained only three percentage points.
Reference: Slavin, Robert E., Nancy A. Madden, Lawrence J. Dolan and Barbara A. Wasik. "Success for All: Evaluations of National Replications." CDS Report No. 43, July 1993
Are data from the study available? Yes
Name of Activity: An Evaluation of Success for All with Language Minority Children
Description of Activity: The activity evaluated the effects of this program in a Philadelphia elementary school serving large numbers of Southeast Asian children who have limited English proficiency. This activity is an evaluation of a form of the Success for All (SFA) program adapted to meet the needs of language minority students. The study is conducted in a school that has a substantial number of students whose first language is not English.
Study Design: This evaluation compares two schools: (1) the Success for All (SFA)school and (2) a matched comparison school. Outcome measures include a measure of English language proficiency in addition to measures of student achievement, referrals to special education, and retention.
Unit of Analysis: The unit of analysis is the individual student.
Generalizability: The findings are based on a purposive sample of an SFA school and a matched control school; these results are only generalizable to the particular high poverty schools involved in the study.
Sample Description: SFA is implemented in an urban disadvantaged school in Philadelphia. Approximately 50% of the students in this school are from Cambodian, Laotian, or other Asian backgrounds, and these students begin school with little or no English proficiency. The comparison school also has a large number of Southeast Asian students and a similar historical level of achievement and poverty. Individual students are matched on standardized test scores and age.
Dependent Variables: The measure of English language proficiency is unspecified. Other outcome measures include: (1) California Achievement Tests -- Reading and Mathematics (Grades 1-3); (2) individually-administered measures of reading scales from the Woodcock and Durrell informal reading inventories (Grades K-3); (3) writing samples (Grades K-3); (4) the Test of Language Development (TOLD) (Grades Pre-K and Kindergarten); and (5) information from school records on attendance, retention/promotion, special education referrals, and disciplinary referrals.
Independent Variables: The independent variable is the treatment (SFA program) versus control (matched non-treatment comparison) condition.
External Variables Controlled: The school and individual matching procedures in this study control for student achievement as measured by standardized test scores and student age, and for school effects, as measured by historical achievement and poverty level.
Statement of Finding(s): Students in the Success for All School significantly outperformed students in the control school in reading.
Description of Finding(s): Asian students' reading skills exceed those of comparable students in a control school by more than a full grade equivalent in first and second grade; third and fourth graders exceed their controls by six and five months, respectively.
Are data from the study available? Yes
Name of Activity: Evaluation of the Beginning Reading Component of Success for All
Description of Activity: Success for All developed and evaluated a program designed to use the best available information on effective programs for disadvantaged students in a comprehensive reform of the urban elementary school. Rather than focusing on remedial services after students fall behind, the approach is to provide intensive services in the early grades (pre-K through 3rd grade) and improved classroom programs in the later elementary grades for all but the most academically handicapped students. The major components of the program include one-on-one tutoring by certified teachers, regrouping and reduction in class size to accommodate whole-class instruction for reading, eight-week curriculum assessments to see that all children are making adequate progress, the implementation of half-day preschool and full-day kindergarten programs that emphasize building of language and pre-literacy skills, a family support team to enlist parental support in the education and health of their children, a full-time project facilitator to coordinate all activities, and a building advisory committee to recommend changes in the program. The Success for All (SFA) program uses an innovative beginning reading program that emphasizes: (1) building of language skills in kindergarten using Peabody Language Development kits and Story Telling and Retelling (STaR); (2) instruction in auditory discrimination, sound blending, and phonetic strategies; (3) partner oral reading in phonetically regular mini-books; (4) multi-sensory practice in letter sounds; and (5) direct instruction in story structure and reading and listening comprehension skills. Reading is taught to the entire class in both kindergarten and first grade; there are no reading groups. The purpose of this activity is to determine the unique effects of the beginning reading component of the SFA program.
Program implementation began in the Baltimore City Schools under the auspices of Chapter 1 and other federal funds prior to the initiation of the Center. Center activities were designed to assess the impact of the program on student achievement and on such variables as referrals to special education, retentions, and self-esteem through a controlled experiment. In Year 1, measures were taken on all students in grades pre-K to 3. In year 2, the measures were extended to grade 5 and in each subsequent year a grade level was added. The program's use has expanded into approximately 200 schools nationwide. The project is continuing to analyze 6th and 7th year effects.
Study Design: The beginning reading component of the Success for All (SFA) program is being introduced in kindergarten and first grade classes in schools that do not have the other components of the program. Teachers who volunteer to participate in the study are randomly assigned to treatment and control conditions. (Control teachers receive training and materials to implement the program after the study is completed.) Outcome measures include scores from individually-administered informal reading inventories and standardized tests.
Unit of Analysis: The unit of analysis is the individual student.
Generalizability: Generalizability of findings is limited to the participating grades and schools.
Sample Description: The sample upon which this study is based is not defined except insofar as the sample schools do not have other components of SFA.
Dependent Variables: The dependent variables are scores from the Woodcock and Durrell informal reading inventories and routinely administered standardized tests.
Independent Variables: The independent variable is the treatment condition, experimental or control.
External Variables Controlled: Extraneous teacher variables are controlled through random assignment of teachers to conditions.
Statement of Finding(s): The Success for All Beginning Reading Curriculum had significant effects on student reading performance compared to control students.
Description of Finding(s): In the two schools evaluating this component of Success for All, the Beginning Reading Program in itself increased the reading achievement of first-graders compared to control students by effect sizes of +.23 on individually administered measures and +.29 on the California Achievement Test.
Are data from the study available? Yes
Name of Activity: An Investigation of the Building Advisory Committee in Success for All
Description of Activity: Success for All developed and evaluated a program designed to use the best available information on effective programs for disadvantaged students in a comprehensive reform of the urban elementary school. Rather than focusing on remedial services after students fall behind, the approach is to provide intensive services in the early grades (pre-K through 3rd grade) and improved classroom programs in the later elementary grades for all but the most academically handicapped students. The major components of the program include one-on-one tutoring by certified teachers, regrouping and reduction in class size to accommodate whole-class instruction for reading, eight-week curriculum assessments to see that all children are making adequate progress, the implementation of half-day preschool and full-day kindergarten programs that emphasize building of language and pre-literacy skills, a family support team to enlist parental support in the education and health of their children, a full-time project facilitator to coordinate all activities, and a building advisory committee to recommend changes in the program.
One important element of Success for All (SFA), a program for disadvantaged urban elementary students, is the involvement of teachers in directing the school. As part of their evaluation of SFA, Center researchers conduct this investigation of the functioning of the Building Advisory Committee and other program elements directed toward eliciting staff participation in planning and implementing school change.
Program implementation began in the Baltimore City Schools under the auspices of Chapter 1 and other federal funds prior to the initiation of the Center. Center activities were designed to assess the impact of the program on student achievement and on such variables as referrals to special education, retentions, and self-esteem through a controlled experiment. In Year 1, measures were taken on all students in grades pre-K to 3. In year 2, the measures were extended to grade 5 and in each subsequent year a grade level was added. The program's use has expanded into approximately 200 schools nationwide. The project is continuing to analyze 6th and 7th year effects.
Study Design: This study is a qualitative examination of the Family Support activities that are part of the Success for All Program activities in Baltimore schools.
Unit of Analysis: Family Support Teams in Five Schools
Generalizability: Each Family Support Team and its work is unique to the individual school.
Sample Description: See previous description of schools in Baltimore City participating in Success for All.
Dependent Variables: Student attendance, behavior, achievement as measured in Success for All studies.
Independent Variables: Family Support Team activities.
External Variables Controlled: The effects of the Family Support Team activities cannot be separated from the overall effects of the Success for All program.
Statement of Finding(s): Family Support Teams in Success for All Schools are successful in determining the need for and providing additional support for low-achieving students in order to improve their achievement.
Description of Finding(s): The following description is of a school in its fourth year of Success for All which serves one of the poorest communities in Baltimore (nearly all students qualify for free lunch). The school is not one of the fully-funded Success for All sites.
Family Support Team. Consists of guidance counselor, parent liaison, master teacher, attendance monitor, Success for All facilitator, principal.
School-Based Interventions. The guidance counselor was the cornerstone for school-based intervention efforts this year (1990-91). Thirty individual student cases were reviewed by the Family Support Team and on-going plans were developed for fifteen of those. The guidance counselor and the Success for All facilitator conducted a five-session group for parents whose children were having difficulty. The group had twelve members who attended regularly. In addition, the guidance counselor saw individual students and their families constantly throughout the year.
Attendance. The Family Support Team wrote and received a grant from the Fund for Educational Excellence that enabled them to hire a part-time attendance monitor, who has run a vigorous program to promote attendance. Attendance increased significantly in 1990-91.
Parent Involvement. The Family Support Team planned and carried out a "Happy Hour" party for staff and parents in January. More than 150 parents attended. The guidance counselor and SFA facilitator held a series of workshops for parents of pre-kindergarten children to promote emergent literacy skills and inform the parents about the pre-K and kindergarten programs. Lending libraries for parents were created, and parents were trained to participate in the Story Telling and Retelling program and in making Big Books.
The Family Support Team also wrote a grant proposal to develop a family literacy program at the school.
Service Integration. The Maryland Food Committee uses the school as a food distribution site. Parents and other community members who come in to receive the food take time to serve as volunteer listeners for children's reading. The Fellowship of Lights provided the school with a family therapy student for eight hours a week, and the student worked with 20 families during the year. The school has an architectural firm as a corporate sponsor. The firm provides weekly enrichment activities, a Boys' Club, and works to promote parent involvement in the school.
Are data from the study available? Yes
Project Director: Dolan, Lawrence J.
Institution: Johns Hopkins University, 3505 North Charles Street, Baltimore, MD 21218
PHONE: (410) 516-8800
FAX: (410) 516-8890
-###-
![[Center Description]](/icons/up.gif){kind=icon}
![[ Next Project ]](/icons/next.gif){kind=icon}