I dreamed I stood in a studio
And watched two sculptors there;
The clay they used was a young child's mind
And they fashioned it with care.
One was a teacher; the tools he used
Were books, music and art;
One was a parent, who worked with a guiding hand
And a gentle loving heart.
Day after day the teacher toiled,
With touch that was deft and sure,
While the parent labored by his side
And polished and smoothed it o'er.
And when at last the task was done,
They were proud of the work they had wrought
For the things they had moulded into the child
Could neither be sold nor bought.
And each agreed they would have failed
If he had worked alone.
For behind the teacher stood the school
And behind the parent, the home.
By Cleo Victoria Swarat
The Child-Parent Center (CPC) Program is a center-based early intervention that provides comprehensive educational and family-support services to economically disadvantaged children and their parents from preschool to early elementary school.1 The CPC program was established in 1967 through funding from the landmark Elementary and Secondary Education Act of 1965. Title I of the Act provided grants to local public school districts serving high concentrations of low-income children for a broad set of programs and projects, as well as funds for personnel and instructional materials. These included, among others, preschool training, enrichment programs for the elementary grades, training for teachers, supplemental health and food services, programed instruction, additional teaching personnel to reduce class size, summer school, and remedial programs in reading and math. The Act also emphasized the development of innovative programs such that local school districts would "employ imaginative thinking and new approaches to meet the educational needs of poor children" (U. S. Senate, 1967, p. 1455). Establishment of the CPC program was based on this latter concept through implementation of comprehensive preschool education with a major emphasis on parent involvement.
Initially implemented in four sites and later expanded to 25, the CPC program was designed to serve families in high-poverty neighborhoods that were not being served by Head Start or other early childhood programs. The program has several distinctions. It is the second oldest (after Head Start) federally-funded early childhood program in the United States and is the oldest extended early childhood program. In 1976, it was selected as an exemplary education program by the Joint Dissemination Review Panel of the U. S. Office of Education. In 1998, it received a similar honor as part of the Title I Distinguished Schools National Recognition Program.2 The CPC program is unrelated to the Parent-Child Development Centers, a family-support program for children under the age of 3 funded through the U. S. Department of Health and Human Services.
Currently, the CPC program operates in 24 centers throughout the Chicago Public Schools; 19 provide services in preschool (ages 3 or 4) and kindergarten, 5 provide services in preschool but not kindergarten, and 13 centers implemented the primary-grade program in 1997-1998.3 The preschool and kindergarten components of the program are funded through Title I of the Improving America's School Act (replacing ESEA). Since 1977, the primary-grade portion of the program (also called the Expansion program) has been funded by Title I through the State of Illinois. 18 centers are in separate buildings proximate to the elementary school and 6 are in wings of the parent elementary school.
The major rationale of the program is that the foundation for school success is facilitated by the presence of a stable and enriched learning environment during the entire early childhood period (ages 3 to 9) and when parents are active participants in their children's education. Four program features are emphasized: early intervention, parent involvement, a structured language/basic skills learning approach, and program continuity between the preschool and early school-age years. The program theory is that children's readiness for school entry and beyond can be enriched through systematic language learning activities and opportunities for family support experiences through direct parent involvement in the centers. This theory is embodied in the following goal statement: "The Child-Parent Education Centers are designed to reach the child and parent early, develop language skills and self-confidence, and to demonstrate that these children, if given a chance, can meet successfully all the demands of today's technological, urban society" (cf. Naisbitt, 1968, p. A).
There are three conditions of eligibility. First, children must reside in school neighborhoods that receive federal Title I funds. Enrollment is reserved for children and families in most educational need as determined by a screening interview with center staff. Second, children must not be enrolled in another preschool program (e.g., Head Start). Third, parents must agree to participate in the program at least one-half day per week, though many parent do not often participate to this extent. To enroll children most-in-need, and reduce self-selection, the centers conduct extensive outreach activities such as distributing program descriptions in the community, visiting families door-to-door, and advertising locally. Several previous analyses have confirmed the substantial educational needs of program children and their parents (Chicago Public Schools, 1974, 1985, 1987; Schuster & Jennings, 1982).Program History
The development of the Chicago Child-Parent Centers can be traced to the middle of 1966 when Benjamin Willis, the General Superintendent of the Chicago Public Schools (CPS), asked Dr. Lorraine M. Sullivan, Superintendent of CPS District 8 and the program founder, to report on ways to improve student attendance in her district. District 8 was located in the center of the West Garfield Park community area and had one of the highest concentrations of poverty in the city. It bordered North Lawndale, another high-poverty area. The residential population in West Garfield Park (and in North Lawndale) was nearly 100% minority; 30% of the residents had incomes below the federal poverty level. The district schools were also among the most overcrowded. In preparing the report, a door-to-door survey of families was conducted concerning attitudes toward education and the district schools.
In the report to the General Superintendent, Dr. Sullivan recommended that the district address attendance and academic problems of students by focusing on the early childhood years when children and parents are most receptive to change. The major recommendations emphasized: (i) parent involvement in the early years of school, (ii) instructional approaches tailored to children's learning styles and designed to develop their speaking and listening skills, (iii) small class sizes to provide for individual attention, and (iv) attention to health and nutritional services (cf. Naisbitt, 1968). These principles were implemented through the establishment in May 1967 of four Child-Parent Education Centers (or Child-Parent Centers). Three were in District 8 and one site was just over the border in District 10 (serving North Lawndale) because of a lack of available land in District 8. Due to the lack of space in existing schools, these centers began as mobile units in close proximity to elementary schools. The units were comprised of six rooms including four classrooms, a parent room, and an administrative office. Each center had an enrollment of 120 and included a half-day (2 ½ hours) preschool and kindergarten program for 40 weeks plus an 8-week summer program. In 1960 and 1970 these community areas were among the lowest in family income in the city. In addition, the area was described as having the "most serious educational needs in Chicago: last year, only 8% of the sixth grade students in District 8 were reading at or above grade level" (Naisbitt, 1968, p. A).
In early 1967, Dr. Sullivan hired four principals to run the new centers--Wayne Hoffman, Helen Brennan, Debora Gordon, and John McGovern.4 Principals were given wide flexibility to hire their own staff and to develop their own instructional approaches. Among the curricula and materials the principals investigated were the High/Scope Perry Preschool Program, the Bereiter-Engleman Program, the Bank Street Program, the Gotkin Language Lotto, and several home-based parenting programs.
Originally, the centers were called clusters and were officially named within the next two years as Cole, Dickens, Hansberry, and Olive Child-Parent Education Centers. Because the centers were located in the lowest income areas of the city, Title I funding was granted from the beginning after approval by the Chicago Board of Education. Although not in the original plan, extended intervention services were available in first grade beginning in the fall of 1968 (for June 1968 kindergarten graduates), and they were expanded to third grade over the next several years. The Olive center provided services until sixth grade for several years (see Fuerst & Fuerst, 1993).
Although the genesis of the CPC program was in response to the identified needs of children and families in District 8, the social conditions of the middle 1960s were ripe for innovation in early childhood education. Project Head Start had begun about a year earlier as the first federally funded preschool program for low-income children. Scientific support also was growing for the primacy of early experience (Bloom, 1964) and for the large effects that modifications in the social environment could have in shaping child development (Hunt, 1961). Other community action programs were being developed in response to the urgent social problems of poverty, school failure, and crime that were disproportionately affecting inner-city minority children. Beginning in 1966, for example, the Model Cities Program began implementation in many large cities including North Lawndale, Woodlawn, Grand Boulevard, and Uptown neighborhoods of Chicago (Janowitz, 1967; Campbell, Marx, & Nystrand, 1969). While these efforts provided financial resources to coordinate and improve housing and general economic conditions at the community level, like those of the Child Parent Education Centers, these innovations were also designed to empower families who were distrustful and alienated from social and educational institutions.
Dr. Sullivan (1971) described the philosophy of the Child-Parent Centers as way to enhance the family-school relationship: "In a success-oriented environment in which young children can see themselves as important, they are 'turned on' for learning. Attitudes toward themselves and others, interest in learning, increased activity, conversation, and enthusiasm are all evidences of the change. Parents are increasingly aware of the role of the home in preparing children for school and have renewed hope that education will develop the full potential of their children" (p. 70). To accomplish this, the centers offered a structured program of parent involvement and language enrichment. Although physical health and psychological development were important goals of the program, promoting the basic skills of written and spoken language as well as numeracy became the primary focus over time. As noted by Naisbitt (1968), "as the program has evolved, the objectives have become more specific in the direction of providing a highly structured, instruction oriented educational program for pre-school children, with maximum emphasis on language and reading skills" (p. 1).5
Largely due to the initial success of children in the original sites, program expansion began in 1969. The four original CPC sites were located in the West Garfield Park, East Garfield Park, and North Lawndale neighborhoods on the west side of the city. Two more were added in 1969: Miller CPC in (East Garfield Park and Wheatley CPC in Riverdale. Together, these six sites are considered the original Child-Parent Centers. Five centers were added in 1970, seven in 1974, and six centers in 1975. The 25th and last CPC was founded in 1978. After one center closed in 1987 (Oakenwald North), there are currently 24 program sites.
From its inception until the spring of 1977, the CPC "activity" was administered completely in the centers for the duration of children's enrollment (up to six years). All funding was through ESEA Title I. In the fall of 1977, the funding and administrative structure of the program was divided. The primary-grade component (expansion program) was funded through Title I of the State of Illinois and was implemented in the parent elementary school serving the centers. The preschool and kindergarten components of the program continued to be implemented in the CPC sites. This is the current program configuration. Interestingly, the primary-grade component of the program was put in place in order to provide extra support to the original kindergarten graduates during their transition to first grade in 1968.
From the beginning, the Child-Parent Centers have served families from the most economically disadvantaged neighborhoods in Chicago. As indicated above, the original centers served among the most economically and educationally needy families. The context surrounding the centers were described at the time as [including] "dilapidated housing, boarded up buildings and vacant land, mixed industrial, retail, and residential use. The public schools are very large, some new and others very old and out-dated. Most are crowded, with playgrounds covered by mobile units. Broken windows are a common sight." (cf. Naisbitt , p. 2).
Figure 2 shows the community areas in which the Child-Parent Centers are located. In 1990, 22 of the 24 centers were located in one of the 20 poorest neighborhoods (Chicago Department of Public Health, 1994). The two other centers (Ferguson and Truth) are adjacent to the Cabrini Green public housing project and are similarly disadvantaged. Five of the six poorest neighborhoods have at least one Child-Parent Center. Herzl, Stockton, Truth, Woodson South, and Cole Centers do not currently operate kindergartens (the first four are not included in findings reported in this book).
Table 3 reports the U.S. Census poverty rates for the community areas in which the CPCs are located. From 1970 (near the beginning of the program) to 1990, the proportion of residents with family incomes below the federal poverty level was two to three time higher in CPC community areas than in areas without a CPC. The individual-level poverty rate for CPC community areas grew from 25.9% in 1970 to 40.5% in 1990, a 56% increase (Chicago Department of Health, 1994; Chicago Fact Book Consortium, 1995). Although the poverty rates for the nonCPC community areas were substantially below those of the CPC community areas, they also increased from 7.7% in 1970 to 16.9% in 1990, a 119% change. Of course, this large increase is partly due to the relatively low baseline rate in 1970.
A similar pattern of disadvantage occurred for unemployment. The unemployment rates in the CPC community areas were, respectively, 7.3%, 16%, and 16.9% for the decades of 1970 to 1990 compared to 3.7%, 9.3%, and 10.7% for the nonCPC community areas (Chicago Department of Health, 1994; Chicago Fact Book Consortium, 1995). These averages do not reflect the fact that some subareas (i.e., census tracts) had unemployment rates approaching 50%. A major reason for the worsening socioeconomic conditions in the neighborhoods surrounding the CPCs is the large decline in the residential population. For example, the community areas housing a Child-Parent Center experienced a 38% decline in population from 1970 to 1990 compared to only 9% for other Chicago community areas. Because the families that left were more socioeconomically advantaged than those that remained, poverty and unemployment became more concentrated and the employment sector diminished as well (Wilson, 1996).
The Chicago Fact Book Consortium (1995) starkly described the social and economic conditions of North Lawndale, one of the locations of the early Child-Parent Centers, during these changing times:
The newest residents of North Lawndale encountered a series of community catastrophes after 1960, which resulted in a stagnated economy and a deteriorating social fabric. First were the riots which came after the King assassination in 1968, during which substantial parts of the Roosevelt Road shopping strip were destroyed by fire. After that storeowners moved when insurance companies either canceled their policies or prohibitively increased their premiums. Another severe blow fell when the International Harvester Company's tractor works closed in 1969, with the loss of an estimated 3,400 jobs.
The riots, coupled with the racial turnover in North Lawndale between 1950 and 1970, purportedly resulted in the loss of 75% of its business establishments and 25% of its jobs. The department store and other retail facilities burned out or closed on Roosevelt Road were never replaced.
In 1974, Sears, Roebuck moved its headquarters to Sears Tower downtown, leaving behind a reduced facility employing 3,000. During the 1970s, 80% of the area manufacturing jobs disappeared, as Zenith and Sunbeam electronics factories shut down, and a Copenhagen snuff plant was closed. The closing of an Alden's catalogue store was a signal event in a sequence that wiped out 44% of the retail and service jobs in North Lawndale.
The downturn continued through the 1980s when Western Electric started closing down, to disappear completely by 1985. Two years later, without warning, Sears, Roebuck closed the Homan Avenue complex, resulting in the dismissal of 1,800 employees (p. 107).
A similar pattern of social upheaval and decline occurred in West Garfield Park:
Between 1960 and 1970, more than 40,000 new African-Americans were added to the population of 7,000. With 40,000 whites relocating outside the area. West Garfield Park became predominantly black during the 1970s, bringing the community to the most pivotal decade in its most recent history. All community indices relating to previous vitality were reversed. Between 1970 and 1980, the population decreased significantly by 30%....This change was accompanied by a drop in housing units of 28%.... Unemployment rose from 8 percent to nearly 21 percent. At the same time, the number of families living below the poverty line increased from 24 percent to 36 percent and female-headed households doubled to 58 percent. One major anomaly appeared as home ownership rose to 22 percent and subsequently continued to rise.
By 1980, the landscape was transformed dramatically as large empty lots, formerly occupied by small and medium-sized apartment buildings evidenced the loss of living space resulting from the withdrawl of investment, undermaintenance, and arson.
In 1980, three years before children's preschool participation in the Chicago Longitudinal Study, the poverty rate for residents in the 14 community areas served by the CPCs was 39% compared to 15% for the nonCPC community areas. Alternatively, 66% of children in the school attendance areas served by the CPCs resided in low-income families at the time of kindergarten entry in 1985. The average rate for the Chicago Public Schools was 42% (Chicago Public Schools, 1986).6 Thus, children and families served by the Child-Parent Centers are at high risk of experiencing educational and social difficulties, and few children in the U. S. are probably at higher risk.
Seventeen (17) Child-Parent Centers are currently located in the following 7 community areas: North Lawndale (4), West Garfield Park (3), East Garfield Park (2), Washington Park (2), Woodlawn (2), Grand Boulevard (2), and Near North Side (2). These areas typically have the highest concentration of poverty and unemployment in the city (Chicago Fact Book Consortium, 1995).
Child-Parent Centers and Head Start
Like Project Head Start, the CPCs provide comprehensive services, expect parent participation, and implement child-centered approaches to social and cognitive development for economically disadvantaged children. There are three distinguishing features of the Child-Parent Centers. As part of the school system, the CPCs are administrative centers housed in separate buildings or in wings of their parent elementary school. They also staff a parent resource room. Head Start programs typically contract with social service or community agencies, and not with school systems. They usually do not have staffed parent rooms in addition to classrooms, although they do provide extensive health sreening and services on site (which the CPCs do not). Second, eligibility for the CPCs is based primarily on neighborhood poverty; for Head Start it is primarily family-level poverty. Since both programs give preference to children in most educational need, this distinction is more illusory than real in practice. A third and most important difference is that the CPCs have historically provided up to six years of intervention services from ages 3 to 9 while Head Start is a preschool program. Thus, the CPC program provides the opportunity for a school-stable environment (i.e., minimal school transfer) during preschool and the early primary-grade years.
The two major CPC program components are described below for the years of program implementation experienced by the children in the Chicago Longitudinal Study. Figure 3 displays the organizational features of the CPCs for the 1983-1989 implementation. Exhibit I shows photographs of the 24 current Child-Parent Centers and some participating children and parents.7
Preschool and Kindergarten Component (1983-1986)
The preschool/kindergarten component in each Child Parent Center is administered under the supervision of the Head Teacher who is responsible for all aspects of program delivery. The program is implemented in a separate building in close proximity to the feeder elementary school or in a wing. Each center has its own budget and administrative operations. The major services coordinated by the Head Teacher are the child education program in the classroom, parent involvement, community outreach, and health & nutrition. The Head Teacher reports directly to the principal of the parent elementary school, which is the location of the primary-grade component (CPC expansion program). I describe the components in the following areas: staffing, structure of operations, curriculum philosophy and activities, parent involvement, health and nutrition, inservice training, and costs. Table 4 summarizes the key program features.
The Head Teacher is the program coordinator with overall responsibly for organizing and implementing program services. This primarily involves teaming with and supervising other primary staff including the Parent-Resource Teacher, the School-Community Representative, classroom teachers, and teacher aides (one for each classroom). Administrative support staff including a clerk and janitor also are funded as part of the program. Several speech therapists and school nurses also serve the centers but they service several centers at a time.
Structure of operations.
Centers offer a half-day preschool program in the morning and afternoon (three hours each) and a half-day (three hours) or all-day (six hours) kindergarten program at most sites (20 of 25 sites during 1983-1986). The centers operate on the regular nine-month school year calendar. An 8-week summer program was provided. Children enter preschool as three-year-olds (2 years, 9 months or older) in separate classrooms. The kindergartners typically are five-year-olds and also have separate classrooms. Although the centers vary in the number of children served (from 130 to 210), they typically house four to six classrooms plus a parent room and office for the Head Teacher or staff.
In preschool, class sizes were set at a maximum of 17 children. With a classroom teacher and a teacher aide for each class, the child-to-staff ratio averaged 17 to 2 (or about 8 to 1). In kindergarten, class sizes could be no larger than 25 children. With the teacher and teacher aide, the child-to-staff ratio averaged 25 to 2 (about 12 to 1). Of course, parent volunteers further lowered these ratios and this often was the case. The relatively small class sizes and the presence of several adults enable a relatively intensive child-centered approach to early childhood education.
Curriculum philosophy, materials, and activities.
The philosophy of the Child-Parent Centers has consistently emphasized the acquisition of basic knowledge and skills in language arts and math through a relatively structured but diverse set of learning experiences (e.g., whole class, small groups, individualized activities, field trips). While not ignoring the importance of psychological development and self-confidence, these affective learning outcomes were built into the reading and language-based instructional activities. For example, teachers provided frequent feedback and positive reinforcement as well as emphasized task accomplishment. The foundational skills of recognizing letters and numbers, oral communication, listening, and an appreciation for reading and drawing were of primary importance.
Although teachers had wide flexibility in selecting instructional materials for use in their classrooms, in the early 1980s they began using the curriculum program called the Chicago Early Assessment and Remediation Laboratory (EARLY; Chicago Board of Education, 1988) developed in the Board of Education's Department of Research and Evaluation (Naron & Perlman, 1981). The EARLY program contains a comprehensive set of instructional activities for preschool children and was designed to supplement other instructional materials. Because all centers used the program on a fairly regular basis, it is representative of the types of instructional activities in the preschool and to a lesser extent kindergarten. The cognitive domains of the activities are summarized below.
A. Body Image and Gross Motor Skills
1. Discrimination of body parts
2. Non-locomotor movement
3. Locomotor movement
5. Spatial integration
B. Perceptual-Motor and Arithmetic Skills
1. Discrimination of colors, shapes, sizes & familiar
2. Arithmetic such as one-on-one correspondence,
sequencing, counting, and number symbols
1. Auditory discrimination such as sound recognition
2. Sentence building with nouns, verbs, adjectives
3. Critical thinking such as story comprehension,
functional association, and verbal problem solving
Most of the activities were for use in small groups of four or five children. Each of the activities included seven operational components: objective, materials, procedure, cautions, suggestions for extension, the criterion, and the item measuring mastery of activity. Figure 4 shows two example activities.
Among other frequently used curriculum materials were Peabody Language Development Kits, Alpha Time, Bank Street Readers, Math Their Way, Language Lotto, and a variety of basal readers, but the latter were primarily for use in kindergarten. None of the instructional materials were used exclusively but rather in combination. Indeed, the centers had substantial funds to purchase instructional supplies and materials.
In kindergarten, classrooms largely followed the school system's Comprehensive Reading Program (CRP; Chicago Public Schools, 1985). The EARLY program was used less often in kindergarten. The CRP was a mastery learning approach to instruction. It was designed to facilitate the development of basic skills in language arts and math. Kindergarten classrooms allotted approximately 45 minutes per day in specific pre-reading and reading instruction. Social studies, science, and music and art activities also were included. Activities often centered around learning outcomes associated with the areas of word attack, comprehension, study skills, and literature with increasing complexity throughout the year. In addition, a variety of basal reading materials were used (e.g., Ginn, Houghton-Miflin, and Open Court readers).
As with preschool, children learned to read and write through a broad spectrum of experiences both in and outside the classroom. Field trips had special significance in the program and they were frequent during both preschool and kindergarten. Parent volunteers almost always went on these outings. For example, classes went to places such as the Museum of Science and Industry, the Lincoln Park or Brookfield Zoo, public libraries, the Art Institute, Union Station railroad, and to local businesses. Upon return, events and observations during the trip served as a basis for classroom and small group discussion (see Chicago Public Schools, 1987).
Based on teacher reports of instructional materials and classroom activities in the centers and on previous evaluations (Chicago Public Schools, 1985, 1987), I classified each of the 20 centers into one of three within-study instructional categories.8 Those that were relatively (i) teacher-oriented (or initiated) in their classroom activities, (ii) developmental, and (iii) undifferentiated (mixed) in that equal emphasis was given to teacher-oriented and developmental activities. Centers classified as teacher-oriented were those that emphasized large-group activities, academic skills, and a structured set of instruction materials (e.g., workbooks, basal readers). Pre-reading instruction was more frequent as well. Centers classified as developmental emphasized child-initiated activities in small-group activities, interest centers, and the use of materials such as the Peabody Language Kits and Bank Street Readers as well as the EARLY. Less emphasize was devoted to pre-reading activities and more to social development. Centers classified as undifferentiated had no distinct focus or were mixed such that teacher-directed and developmental activities were used equally.
Seven of the 20 centers were classified as primarily teacher-oriented, 10 were developmental, and 3 were classified as undifferentiated. These designations provide an overall within-study descriptor of implemented instructional approaches. Because all centers used a diversity of instructional activities, the categories are best viewed along a continuum from teacher-oriented to developmental. Most of the centers were in the middle range. Implementation occurred within the context of low adult-child ratios and extensive parent participation. Moreover, this classification best describes the preschool program. In kindergarten, most of the centers were relatively structured and oriented toward the development of academic skills.
As the program's title indicates, a central operating principle of the program is that parent involvement is the critical socializing force in children's development. Direct parent involvement in the program is expected to enhance parent-child interactions, parent and child attachment to school, social support among parents, and consequently promote children's school readiness and social adjustment. The centers make substantial efforts to involve parents in the education of their children, so much so that it is hard to overestimate. At least one-half day per week of parent involvement in the program is required.9 The unique feature of the parent program is the parent room, which is physically located in the center adjacent to the classrooms. The full-time parent-resource teacher organizes the parent room in order to implement parent educational activities, initiate interactions among parents, and foster parent-child interactions. With funds for materials, supplies, and speakers, areas of training include consumer education, nutrition, personal development, health and safety, and homemaking arts. Parents may also attend GED classes at the centers. Parent also serve on the School Advisory Council which assists staff in educational planning and implementation.
As shown in Table 5, a wide range of activities are encouraged in the program including parent room activities (e.g., arts & craft projects), classroom volunteering, participation in school activities, class field trips, helping to prepare breakfasts and lunches, and engaging in education and training activities. The diversity of activities are designed to accommodate parents' daily schedules and different needs. They help parents (a) understand themselves, (b) understand the importance of teaching their children at home, (c) feel comfortable in the role as volunteer, and (d) learn more about child development (Chicago Public Schools, 1983). Among the commercial programs used were Exploring Parenting, Parent Effectiveness Training, and Parents as Partners in Reading.
Examples of parent activities in the classroom are provided below (see Bureau of Early Childhood Programs, undated):
--Reading stories to small groups of children
--Changing the bulletin board
--Finishing activities begun by the teacher (e.g., a reading lesson)
--Arranging games for children who finish assignments early
--Listening to children describe his/her experiences
--Assisting in the planning of field experiments
--Conducting science experiments with small groups
--Working on craft projects with children in the interest center
--Practicing math activities one-on-one or in small groups.
School-Community Outreach Services.
In each center, the full-time school-community representative, who usually has grown up in the neighborhood, provides outreach services to families in three related areas. First, they identify families in the neighborhood to enroll in most educational need. They make door-to-door visits of likely participants, distribute brochures and advertisements of enrollment, and communicate with prospective families both formally and informally. Second, the school-community representative conducts a home or school visit to all enrolling families. One visit upon enrollment in the program is required. Additional visits occur on a most-in-need basis. Moreover, informal conferences are held between the parent and the school-community representative. Finally, the school-community representative mobilizes resources by referring families to community and social service agencies such as employment training and education, mental health services, and welfare services. The school-community representative also organizes transportation and related services for families.
Physical health and medical services.
Upon entry into the program, children undergo a health screening from a registered nurse on-site. Tests are given for vision and hearing. Enrolling parents are expected to provide records of their child's medical and immunization history. All children are required to have a physical and dental examination. Children in need of preventative services are referred to appropriate service agencies. Special medical and educational services such as speech therapy also are available. All children in the morning and all-day programs receive free breakfasts and lunches as part of the national programs offered by the U. S. Department of Agriculture. Children in afternoon programs receive only free lunches.
The CPC program also provided funds and time for staff development for the Head Teachers, classroom teachers and aides, the Parent-Resource Teacher, and the School-Community Representative. The Head Teacher typically provided inservice training to teachers and instructional staff and often invited outside speakers to present information on salient topics. Staff from the school system's Department of Early Childhood Programs also sponsored training activities for Head Teachers as well as other instructional staff on a regular basis.
Expenditures for the CPC program (in 1996 dollars) were estimated at $4,350 yearly per child for each of the half-day preschool program and the half-day kindergarten program. These amounts do not include expenditures for the free breakfast and lunch program (funded by the U. S. Department of Agriculture), or to implement full-day kindergarten programs in 14 centers (funded by Title II of ESEA).
Primary-Grade Component (Follow-on Program)
Beginning in the fall of 1977, the primary-grade component was implemented in the elementary schools serving the Child-Parent Centers. This follow-on to earlier intervention is implemented as a vertical expansion of the preschool and kindergarten component. Previously, all program operations were implemented in the centers and were administered by the principals (now Head Teachers) in their respective sites as part of the federally funded program. Thus, one major difference between the preschool/kindergarten and primary-grade components is that the latter is implemented in the parent elementary school rather than in the Center. In addition, all children are eligible for and participate in the primary-grade program regardless of whether they participated in preschool and kindergarten. Moreover, program coordination is streamlined. The Head Teacher and the Parent-Resource Teacher are combined into one position (curriculum parent-resource teacher). The school-community representative serves the whole school instead of just of the Center. The six schools serving the original CPCs (opened in 1967-1969) provided three years of service (first to third grade); the remaining centers provided services two years of service (first and second grade).
The Curriculum Parent-Resource Teachers in each school supervised all program operations in the elementary school and coordinates the teachers and teacher aides in the program classrooms. They also works directly with the school-community representative and other auxiliary staff in the school. The Curriculum Parent-Resource Teacher's responsibilities are to (i) coordinate the primary level classroom program in the school, (ii) order materials and supplies, (iii) conduct on-going parent programs, including coordinating parent classroom involvement, (v) assist teachers in implementing individualized instructional activities, and (v) providing and organizing inservice training. The teacher reports directly to the principal and also cooperates with the Head Teacher in the affiliated preschool/kindergarten component.
Structure of operations.
Program schools offer an all-day program in 24 sites during the 1986-1989 implementation (one closed in 1987). Children entered first grade as six-year-olds. Although all first and second graders (third graders in selected schools) participated in the program, each classroom had at least 50% composed of children from the CPC preschool/kindergarten component. The number of rooms per school devoted to the program varied from 4 to 18 per school and included from 90 to 420 students.
The program was designed to enrich the primary-grade classroom experience. The main innovation was a modification in organization. In each grade, class sizes were reduced to a maximum of 25 children and each teacher was provided with a teacher aide. With a classroom teacher and aide for each class, the child-to-staff ratio averaged 25 to 2 (or about 12 to 1). These reduced class sizes and ratios required the additional funding per year of approximately 40 full-time teachers and 220 teacher aides across schools. Parent volunteers further lowered these ratios. The reduced class sizes in the program contrasted significantly with classrooms in other schools, which typically had 30 or more students usually without teacher aides.
Curriculum philosophy, approach, and materials.
The instructional philosophy of the CPC program emphasized the development of basic skills in language arts, math, and social studies in first and second grade with greater attention to science beginning in third grade. All schools implemented the CRP, and reading comprehension was the explicit objective of the curriculum. The instructional approaches in the classroom did not differ markedly between CPC and nonCPC schools. Because CPC classrooms had fewer students, more adults, and a wider array of instructional supplies and materials, CPC program children experienced a more intensive, individualized education than their comparison-group counterparts.
CRP was a directed reading program organized to include developmental reading (basal readers), corrective reading, remedial reading, recreational reading, and enrichment reading. A five-step directed reading lesson included the following components: (i) preparation, (ii) reading, (iii) interpretation, (iv) development of skills, and (v) extension of experiences. In each of the grades, 1 hour in developmental reading with a basal reader was offered and 30 minutes for skill emphasis. As part of the skill emphasis, Chicago Mastery Learning materials or other supplementary materials were used. One day per week, an hour of recreational enrichment reading was provided. In the program sites, a variety of basal reading programs were used including Ginn (seven sites), Houghton-Mifflin (six), Holt, Reinhart, & Winston (5), Harcourt Brace (3), and Open Court (3).
Learning goals were as follows. By the end of first grade, students were expected to successfully complete activities designed to identify the main ideas of written stories and pictures and to draw conclusions from them, identify short vowel sounds in words, and recognize plural forms of words, compound words and contractions as well as identify parts of books. In grade 2, students engaged in activities to identify synonyms, antonyms, and homonyms; recognize cause and effect relationships; and identify character traits setting, and plots in written stories. At the end of grade 3, students were expected to identify word meanings, obtain information from graphs and charts, make predictions, use a dictionary, and identify facts and opinions.
The main difference between the parent programs between the two program components is that parent involvement is strongly encouraged but not required in the primary grades. Otherwise, the Curriculum Parent Resource Teacher implements activities in the elementary-school parent resource room in a similar fashion to the preschool/kindergarten component. For example, each teacher has a budget for supplies, materials, and equipment to conduct parent education and training activities, encourages home support for learning, and organizes and participates in field trips.
The school community representative serves the primary-grade program by conducting home visits, meeting with parents and children at school about educational activities, provide referrals to social service agencies, monitoring student attendance, and assisting the Curriculum Parent Resource Teacher and the Head Teacher in the Child Parent Center. Because the School-Community Representative serves all children in the school, the intensity of service delivery to program participation is lower than in preschool and kindergarten.
Health and nutrition.
A school-wide nurse is available to assist children with identified health problems. The curriculum-parent resource teacher also monitors children's needs and refers them to school-wide services or to community agencies. Speech therapists, school psychologists, and social workers also are available as part of the school-wide coverage. Children received free breakfasts and lunches if they are eligible.
Like the Head Teacher, the Curriculum Parent Resource Teacher provides in-service training to classroom teachers and aides in the expansion classrooms. Staff from the Department of Early Childhood Programs provide in-services to the curriculum parent resource teachers and have orientation meetings with principals of the elementary schools.
Expenditures for the primary-grade program were approximately $1,500 anually per child above and beyond the regular school program (1996 dollars). This translates to $3,000 for two years and $4,500 for three years of participation. The largest portion of this expenditure was to reduce class sizes and to provide teacher aides for each class. These amounts do not include expenditures for the free breakfast and lunch program (funded by the U. S. Department of Agriculture) provided to eligible students.
The CPC program was designed to serve children during the formative early childhood years, ages 3 to 9. Due to differences in age of entry and exit, substantial variation in exposure to intervention occurs. This was due to administrative selection, family self-selection, and mobility. Children enrolling in the six original centers at age 3 can enroll up to six years while those entering at age 4 receive up to 5 years. Children entering a later CPC at ages 3 or 4 can participate for up to 4 or 5 years. Because any child can enroll in the follow-on (expansion program) in grades 1 to 3, this opportunity contributed to variation in program participation.
Over the years, the implementation of the CPC program has been investigated extensively. The Educational Testing Service (Marco & Landes, 1971) completed the first process evaluation of the CPC program for the 1969-1970 implementation period. Based on surveys and interviews of staff and school administrators, the CPC program earned the highest overall quality rating of the 24 Title I Chicago programs. The program was rated high on general quality, administrative desirability, social consequences and moderately high on efficiency. Especially high were ratings on the items "fostering pupil interest" and in "stimulating good relations among the school, home, and community."
Results of later evaluations have confirmed that the program has consistently served the intended target population and successfully provided the expected educational and family support services. Specifically, these evaluations have found that (i) the population of at-risk families served matched the intended target population (Stenner, 1974), (ii) rates of child attendance regularly exceeded 92%, which is 4 to 6 percentage-points higher than other Title I programs (Institute for the Development of Educational Auditing [IDEA], 1973, 1974; Schuster & Jennings, 1982), (iii) centers implemented their programs similarly in their attention to a basic skills approach and a focus on child development and parent involvement, and (iv) the distinctive elements of the program were observed and were present to a greater degree than instructional programs at nonCPC sites (Chicago Public Schools, 1985, 1987; Conrad & Eash, 1983, IDEA, 1974).
Conrad and Eash (1983), for example, found that CPC classrooms were rated by observers as significantly higher than nonCPC classrooms in child-centeredness (i.e., small class size, individual and small group work) and in basic skills and evaluation orientation. CPC classrooms also were judged to have somewhat but not significantly more enriched material environments and greater parent classroom participation. Relative to nonCPC parents, CPC parents displayed higher home control and support for their children than nonCPC parents. In the IDEA study, parental program involvement significantly discriminated between CPC participation and nonparticipation as well as led to changes in children's school achievement. Schuster and Jennings (1982) found that over 80% of CPC parents indicated in surveys that they visited the center or had been involved in the instructional program. A review of parent-room records at selected sites revealed that over 50% of parents were actively involved in their child's center for at least two days per month.
Program implementation during the early to middle 1980s, the years of operation covered by the study sample, indicated a similar pattern of successful service delivery (see Chicago Public Schools, 1985, 1987; Reynolds, 1994), thus reinforcing the basic program elements of early parent involvement and a basic skills approach to early education and language development. In assessing kindergarten programs in the Child-Parent Centers, an evaluation by Nancy Mavrogenes (Chicago Public Schools, 1987) indicated that reading activities (e.g., decoding) compromised 50% of all activities observed and writing activities compromised 23%. Field trips were common and they usually involved extensive preparation and discussion during class. Parents of CPC children also were found to be more involved in the center than other parents. Their involvement frequently included engaging in activities in the parent room, going on class field trips, tutoring students, and performing clerical tasks.10
Finally, parents, staff, and school administrators have indicated strong support for early intervention and program continuity exemplified in the 6-year duration of the program. Reports of the CPC program also have identified two major areas for improvement. Because the centers select their own instructional materials, use of a uniform curriculum has been recommended (Schuster & Jennings, 1982). To address this recommendation, the centers began using the EARLY materials in the early 1980s. The second recommendation was to increase parent participation in the program. Both program staff and parents have commented over the years that a greater proportion of parents should participate in the program (Chicago Public Schools, 1987, IDEA, 1974). One advantage of requiring parent involvement is that more parents participate but this stated policy leads to the expectation that all parents will participate. It has been observed that while all parents do not fully participate in the program, participation is high relative to programs that both do and do not require participation (Chicago Public Schools, 1987; Schuster & Jennings, 1982).
Early Studies of Program Impact
Although there have been many reports documenting the success of the CPC program in promoting children's school achievement, controlled studies have been rare. The initial outcome evaluations were mandated monitoring studies and performance audits by contractors at IDEA during 1971-1972 and 1972-1973 (IDEA, 1973, 1974; Stenner & Mueller, 1973). Among the major findings were that CPC participants scored at or above the national average in language and math tests in kindergarten and that the vast majority of participants scored in the range of "ready for first grade" on the Metropolitan Readiness Test. Children who remained in the Child-Parent Centers in first to third grades maintained their level of performance relative to the national average and increased their performance advantage relative to Chicago students generally as well as students in other Title I programs. Post-kindergarten academic performance was reported only for students that continued in their participation in the program. The fourth and fifth grade scores of CPC students with extensive program participation were substantially higher than the Chicago average and Title I students but they scored, on average, 6 to 8 months lower than the national average.
Nevertheless, the findings that programs for educationally disadvantaged children in the poorest inner-city neighborhoods can effect substantial performance gains led to further on-site observation and analysis which resulted in the program's selection in 1974 as one of 12 outstanding reading programs by the Right to Read Office of the U. S. Office of Education. In 1976, the CPC program was selected as one of 33 exemplary programs by the Joint Dissemination Review Panel of the U. S. Office of Education.
Two controlled studies of the full program (preschool/kindergarten and primary-grade components) were completed after the mid 1970s. Investigating the longer-term effects of extensive participation in the six original CPCs for 684 students, Fuerst and Fuerst (1993) found that those who participated in the CPCs for four or more years from 1967 to 1977 scored, on average, 7 to 8 months higher in eighth-grade reading achievement and 4 to 5 months higher in math achievement than students who did not participate in the program but who attended schools serving a similar composition of students. A significantly higher proportion of CPC students (62%) graduated from high school compared to 49% for the no-treatment comparison group. Only program girls, however, had a significantly higher graduation rate than their same-sex peers (74% versus 57%), although the direction of influence for boys favored participants (49% versus 40%). The effects of the full program relative to participation in only the preschool and kindergarten component or in the follow-on component were not investigated.
Conrad and Eash (1983) investigated the effects of participation in both the preschool/kindergarten and CPC expansion program in four sites from 1977 to 1982 under the reconfigured program structure (primary component transferred to the elementary school). Relative to a nonCPC comparison group matched on family and school demographic characteristics, 227 children who participated in the CPCs in preschool or kindergarten had significantly higher scores on several indicators of kindergarten achievement. Similar to Fuerst and Fuerst's (1993) findings, participation in the CPC expansion program after preschool and kindergarten was associated with significantly higher scores in reading and math achievement at age 8 and with higher scores on locus of control. Nevertheless, these later analyses included only 54 students. The kindergarten findings were similar to comparative outcome evaluations of later implementations (Chicago Public Schools, 1985, 1987).
Overall, these later comparative evaluations largely supported the findings of the early monitoring reports that participation in the CPC program was associated with significantly higher scores on achievement tests and ratings of social psychological development. By the end of kindergarten, CPC children have consistently been at or above the national average in reading readiness, language development, and math achievement and appear to maintain their scholastic advantages to a greater degree than most other early childhood interventions, presumably because of participation in follow-on program as well as the focus on parent involvement and language skills. None of the earlier evaluations, however, investigated the effectiveness of the program at all centers nor the differential effects of timing and duration of intervention exposure. These earlier studies also did not include a comprehensive set of child outcome variables that are common in contemporary evaluations of early childhood intervention. Moreover, sample attrition after kindergarten was not taken into account in analyses. On-going, prospective longitudinal studies are usually recommended for evaluating the effects of prevention programs and of programs that receive substantial priority in funding. They offer many advantages over retrospective studies that rely on administrative records as well as time-limited studies that provide snapshots of program effectiveness.
Chicago Longitudinal Study of the Child-Parent Centers
The Chicago Longitudinal Study (Reynolds, 1989, 1991; Reynolds, Bezruczko, Mavrogenes, & Hagemann, 1996) was designed as a prospective investigation of the scholastic development of 1,539 children who graduated from government-funded kindergarten programs in the Chicago Public Schools in 1986. The study includes all children (n = 1,150) who enrolled in the 20 CPCs with preschool and kindergarten programs beginning in the fall of 1983 and were kindergarten graduates. Children enrolled at ages 3 or 4 and could continue their participation up to age 9 in the spring of 1989 (end of third grade). Some children received services from the CPCs in kindergarten (but not in preschool) and participated in the primary-grade component. Most of these children were classified as CPC participants. Six of the 20 centers offered the intervention through third grade while the other centers offered it through second grade. Consequently, the relationship between duration of participation and school adjustment can be investigated.
The original comparison group included an additional 389 children who graduated in 1986 from government-funded all-day kindergarten programs from seven schools participating in the Chicago Effective Schools Project (CESP). Five were randomly selected and the other two were located in the CPCs or their parent elementary school. The schools participated in the Chicago Effective Schools Project, a school system program to meet the needs of high-risk children. Children served by these schools matched the poverty characteristics of the CPCs and, like CPC participants, they were eligible for and participated in government-funded programs. This nonCPC comparison group had no systematic intervention experiences from preschool to third grade, although some enrolled in Head Start and/or the primary-grade component. Thus, children in the comparison group had intervention experiences that were fairly representative of low-income families in Chicago in the middle 1980s and probably more extensive than many low-income families in other cities.
Patterns of enrollment.
Table 6 displays the program characteristics of the study sample in kindergarten and in the follow-up period at age 14.11 Program group 1 includes children in the six original CPCs and they participated for up to six years (preschool to third grade). All original centers have all-day kindergarten programs; 5 are located in wings of the parent elementary school. Program group 2 includes children in the newer CPCs, in which follow-on participation is limited to two years (second grade). A majority of these centers have all day kindergartens and are in separate buildings near the parent elementary school.
Program groups 3 and 4 did not participate in the CPC preschool program and together they served as the preschool comparison group. However, group 3 did participate in kindergarten programs in the six original Child-Parent Centers, and they received a similar pattern of services as children who began in preschool. Group 4 constituted the CESP (nonCPC) comparison group and included 389 children without participation in the CPCs in preschool or kindergarten.12 Because all children who attended program schools were eligible for the primary-grade component, some children in the nonCPC comparison group did participate in the CPC follow-on program. Thus, there was significant variation in participation in preschool, kindergarten, and in the primary grade components of the CPC program. This variation enabled comparisons of the timing and duration of intervention exposure (see Appendix A for the characteristics of the study sample over time).
Table 7 shows the distribution of program and comparison-group children across the 25 sites at the beginning of the study in 1985-86. The centers and their parent elementary schools are listed by year of implementation. The six centers opened in 1967-1969 provide up to six years of intervention services; those opened in 1970 and later provide up to five years of service.
Defining the experiences of the nonCPC comparison group.
The nonCPC comparison group included children who enrolled in seven randomly selected schools participating in the all-day kindergarten program of the Chicago Effective Schools Project (CESP). Two of the schools were elementary feeder schools for children from the Child-Parent Centers and five were not. In contrast to children in the Child-Parent Centers, children in the CESP schools did not receive systematic intervention services from preschool to the primary grades. They matched the racial and socioeconomic composition of children from the Child-Parent Centers and were eligible for and received early childhood intervention services. About one-fifth of the nonCPC comparison group did enroll in Head Start, however. Consequently, all comparison groups in this study received alternative intervention rather than no intervention.
Begun in 1981 and administered by the Office of Equality of Educational Opportunity, the CESP was designed to improve the achievement patterns of educationally disadvantaged students in racially isolated schools. The 27 schools in the project were selected primarily because they enroll a large proportion of low-achieving students as determined by scores on the Iowa Tests of Basic Skills. Other criteria included student mobility, attendance rates, socioeconomic status, and percentage of minority students enrolled (Chicago Board of Education, 1984). Generally, the CESP was designed to target key areas of school improvement as informed by research on effective schools. These areas included the following: school leadership, instruction, school climate staff development, parental involvement, and assessment of school progress. Implementation of an all-day kindergarten program, which began in 1984, was a manifestation of these principles in that the program provided reduced class sizes with teacher aides, extra funds for instructional materials ($650 per room), and staff in-service training. Parents also were encouraged to participate in children's schooling but without the resources of a staffed parent resource room.
Table 8 shows the sample characteristics of the original CPC program and comparison groups prior to participation in the follow-on program. The CPC program group includes children with any participation in the program from preschool to Grade 3. At the beginning of the study, the groups were similar on several characteristics including school poverty, sex, race, and SES as proxied by eligibility for free lunch. CPC participants had, on average, fewer brothers and sisters (2.4 versus 2.7) and their parents had a higher rate of high school graduation (60% versus 51%). These differences will be taken into account in estimating program effects.
Of the original sample of 1,539 children, 1,164 (76.5%) were active in the Chicago Public Schools at age 14 (77% of program group and 72% of comparison group); 1,070 ( 70%) were active in Chicago at age 15. Although CPC participants were more likely to be active in the study sample, the differences were not significant for the most part (see Appendix B), and no selective attrition has occurred on youth outcomes in this study (see Appendix B) or in previous ones (Reynolds, 1994, 1995; Reynolds & Temple, 1995). The lone exception was that selective attrition did occur for measured family outcomes. Program participants (in preschool) whose parents provided family outcome data had higher word analysis scores in kindergarten than those did not provide data (attrition sample; see Appendix B for details).
Children in the comparison group of this quasi-experimental study did not enroll in the CPCs primarily because they did not live in a neighborhood of the center. Thus, geographic location rather than family motivation or other self-selection factors determined nonparticipation. Reynolds and Temple (1995) found that preschool participation can be predicted with 86% accuracy from child, family, and school-level information. Program site characteristics were the best predictors of program participation and effects persisted when they were included in the model. As shown in Table 8, some children from the original comparison group did participate in the CPC follow-on program, and they were counted as participants.13
Why did children who enrolled in the CPCs leave before the end of the program? Some parents preferred to send their children to regular school programs. They enrolled their children in preschool and kindergarten with the intention of moving to other schools afterwards. Other parents moved out of the school neighborhood due to professional (e.g., job change) or personal reasons. Children participated for six years instead of five years because of administrative selection, since six years were only offered in six schools while the remaining schools offered a maximum of five years. Thus, for the most part, children did not leave the program after kindergarten because of poor performance or because of superior performance.
Summary of Evidence in the Chicago Longitudinal Study
For the past decade, the Chicago Longitudinal Study has investigated the effects of different levels of program participation. A brief summary of previous findings is provided below. Table 9 reports some of the major findings for school achievement, grade retention, and special education placement.
Studies of the effects of preschool participation on school competence outcomes indicate that children who participated for one or two years in the CPC preschool had higher reading and math achievement test scores and lower rates of grade retention and special education placement up to age 12 (end of sixth grade) (Reynolds, 1995; Reynolds & Temple, 1995; Reynolds, Mehana, & Temple, 1995). Consistently significant differences in favor of preschool participants also have been found for parent participation in school (Reynolds, 1995) but less so for classroom adjustment (Reynolds, 1995) and children's perceived competence (Reynolds et al., 1995).
While both one and two years of participation were found to be significantly associated with positive adjustment, the estimated effects of the second year of preschool were mixed. They were significant and educationally meaningful only for kindergarten entry and exit achievement and for special education placement in the early elementary grades (Reynolds, 1995).
Primary-grade intervention alone.
Reynolds (1994) found some support for the short-term effectiveness of intervention beginning in kindergarten and continuing to third grade compared to kindergarten-only intervention. The 76 children who participated in the program from kindergarten to third grade scored higher than the 191 children in the no-treatment comparison group in reading achievement, math achievement, and had a lower rate of grade retention at the end of the program (third grade). This advantage persisted at the 1-year follow-up for reading achievement (Ms = 102.6 versus 99.6) and for grade retention (M rate = 18.7% versus 25.9%). No significant differences were detected at the 2-year follow-up for any measure.
Extended intervention to second and third grade.
Recent analyses up to age 12 indicate that while preschool intervention contributes significantly to positive adjustment over time, extended intervention appears to add significantly to the effect of preschool and kindergarten intervention (Reynolds, 1994). At the end of the program in third grade and at the 2-year follow-up, duration of program participation was significantly associated with higher reading achievement, higher math achievement, and a lower rate of grade retention. Relative to program participation ceasing in kindergarten, participation that extends to second or third grade was associated with higher reading and math achievement, and lower rates of grade retention and special education placement at the 4-year follow-up. Notably, the estimated effect of participating in three years of the follow-on component was higher than participation in two years. Recent findings at age 13 supports these results (Reynolds & Temple, 1998; see also Reynolds, 1998d).
Mediators (pathways) of effects.
Two studies have specifically addressed the mediators or pathways through which program participation influence later performance. Based on a subsample of children in the six original CPCs, Reynolds (1992a) found that cognitive readiness at school entry (measured by the Iowa Tests of Basic Skills) was the primary mediator of the effects of preschool participation on grade 3 achievement for children in the six original CPCs. Parent participation in school, school mobility, and teacher ratings of classroom adjustment did not directly mediate the effects of program participation. Parent participation in school mediated preschool effectiveness indirectly through school mobility.
A follow-up study of the sample in sixth grade (Reynolds et al., 1996) used a confirmatory structural modeling approach and found that both cognitive readiness at school entry and parent participation in school (teacher and parent ratings) independently mediated the effects of preschool participation on reading and math achievement. Teacher ratings of classroom adjustment in first grade and avoidance of school mobility and grade retention also contributed to the transmission of preschool effects. In support of the mediating role of parent participation in school, Reynolds (1994) found that parent participation in school significantly explained differential performance in reading and math achievement among children who participated in the CPC program for different lengths of time from preschool to second and third grade.
Evidence supporting the social adjustment and school support hypotheses of program effectiveness is limited. In previous studies (Reynolds, 1992a; Reynolds et al., 1996), teacher ratings of classroom adjustment were influenced by preschool participation only through cognitive readiness at school entry. School mobility (an indicator of school support) also did not directly mediate the effects of program participation on school achievement. The motivational advantage hypotheses has not been investigated. Alternative hypotheses explaining the effects of extended early intervention also have not been investigated.
Moderators of effects.
Previous studies have not extensively investigated program effects by subgroup. Several studies (Reynolds, 1995; Reynolds, 1994; Reynolds et al., 1996) reported no gender differences in the effect of preschool or extended participation on school achievement and school adjustment--girls and boys benefitted equally. Children experiencing a higher number of environmental risk factors (e.g., growing up in a high poverty neighborhood, with a mother with limited education, and with large family sizes) were as likely to benefit from the program as children experiencing relatively few environmental risks (Reynolds, 1998a). More refined analyses involving parent education, school attributes, and program characteristics have not been completed.
The present study extends on these earlier analyses by systematically investigating the effects of timing and duration of intervention exposure, the pathways of program effectiveness, and the moderators of program effectiveness for the ages 14 and 15 follow-up sample. No prior studies in the CLS sample have done this and I am aware of no studies of large-scale programs that have done so.
Summary of Methodological Issues
The present study reports new outcome data for the active Chicago sample. The methodological issues inherent in this longitudinal project have been addressed extensively. The principle threats to the validity of findings are selection bias into the program and nonrandom attrition from the sample. Selection bias due to the quasi-experimental design of the study has been extensively investigated. Selection bias into the preschool intervention appears to be small and does not affect estimates of program impact (Reynolds & Temple, 1995; Reynolds et al. 1996).14
As shown in Table 9, the estimated effects of preschool participation have been found to be educationally meaningful (values of >= .20 standard deviations or increases in the rate of success of 20% or more) regardless of sample size, analytic method, comparison group, and adjustments for selection bias and measurement error. Concerning selection bias associated with participation in the follow-on intervention, Reynolds and Temple (1998) found that although children who participated in the follow-on program had higher achievement test scores than the non-follow-on group prior to the follow-on program, preprogram growth rates in achievement from the beginning to the end of kindergarten were equivalent between groups. This finding renders the selection-maturation explanation of observed effects unlikely. Program effects persisted up to seventh grade even after differences in kindergarten pretest scores and kindergarten growth rates, and family demographics were taken into account.
Sample attrition also has had minimal effects on the findings of the Chicago Longitudinal Study. Most of the original study sample was active in the Chicago Public schools at age 14 (eighth grade) and no selective attrition among groups were found, although program participants were more likely than comparison-group participants to be active in the study sample after controlling for differences in child and family factors (See Appendix B). Reynolds and Temple (1998) found that taking into account sample selection due to attrition increased rather than decreased the size of the effects of extended intervention.
Compared to many other studies, the generalizability of the findings of the Chicago Longitudinal Study is high. This large sample of children participated in an established, government-funded program for different lengths of time. Consequently, generalizability extends to low-income children in many public programs in central cities. As indicated in the introduction, research on the long-term effects of such programs is quite limited. That the program and comparison groups are well matched on several background characteristics and equally eligible for early childhood programs further strengthens the interpretability of findings and thus increases their significance for public policy.
In summary, the CPC program is a center-based early intervention that provides comprehensive educational-support and family-support services to children and families in the poorest neighborhoods in Chicago during ages 3 to 9. As one of the earliest educational innovations of the 1965 Title I Act, the program has a long history of successful program implementation. In the past decade, the Chicago Longitudinal Study has investigated the effects of program participation for a 1983 to 1989 cohort, and the conditions under which long-term effects may be achieved. Previous studies using a variety of analytic approaches have consistently found that program participation is significantly associated with greater school achievement and fewer problem behaviors. Longer-term effects across a wide range of youth and family development are yet to be determined. This is the goal of the rest of the volume.