The Benefits of High-Quality Early Education
By Robert G. Lynch
Perry Preschool Project
Description: The Perry Preschool Project (Ypsilanti, Mich., 1962-1967) was created in the early 1960s by David Weikart, the then-special education director of the Ypsilanti Public Schools, to see if grade retention and widespread school failure could be prevented. Between 1962 and 1965, 123 African-American children with low IQs (in the 70 to 85 range) and from families with low socioeconomic status were randomly assigned to one of two groups: one enrolled in the new preschool program and one not.
Those enrolled in preschool attended for two school years at ages 3 and 4. For seven months of the year, the Project's four teachers had a total of 20 to 25 children in class for 2.5 hours per day and made weekly 1.5-hour home visits to each child and his or her mother. The curriculum and rich environment were well-conceived to provide children with experiences that build their language and foster pre-literacy, mathematics, logic, music, art, and social interaction skills. The daily routine that teachers established encouraged children to make choices and solve problems, thereby contributing to their cognitive and social development. The teachers were knowledgeable about early childhood development and education, received ongoing curriculum training and supervision, and communicated frequently with parents.
Results: Evaluations of the children were performed annually until the children reached 11, and then again at ages 14, 15, 19, 27, and 40. Table 1 summarizes some of the statistically significant outcomes of the preschool program.
Each time the children were evaluated, important benefits of the preschool program emerged. For example, by age 10 only 17 percent of the preschool children had been held back a grade or placed in special education compared to 38 percent of children who had not been placed in preschool. By age 14 the preschoolers had significantly higher achievement scores, and by age 19 they had higher literacy scores and grade-point averages.
The differences in achievement appear to have grown over time. By age 27, seventy-one percent of the preschoolers had graduated from high school versus 54 percent of those not placed in preschool. Seven percent of the preschoolers had been arrested five or more times as compared to 35 percent of those who had not participated in preschool. Seven percent of the preschoolers had been arrested for drug-related offenses compared with 25 percent of the non-preschoolers. By age 27, significantly fewer preschoolers had ever been arrested (57 percent versus 69 percent of the control group), and the average number of arrests was about half (2.3 life-time arrests versus 4.6 for the control group).
In addition, the children in the program had significantly better lifetime earnings. About 29 percent of preschoolers earned $2,000 or more per month compared to 7 percent of the non-preschoolers. The employment rate was 71 percent for the preschoolers compared to just 59 percent for the non-preschoolers. At age 27, average monthly earnings were 59 percent higher for the program participants ($1,219 versus $766 in 1993 dollars); 27 percent of preschoolers owned their own home, and 30 percent owned a second car. Only five percent of non-preschoolers owned their own home, and 13 percent owned a second car. Just 59 percent of preschoolers had received welfare or other social services in the past 10 years versus 80 percent of the non-preschoolers. More dramatically, only 15 percent of preschoolers were receiving public assistance at age 27 compared to 32 percent of the non-preschoolers. Finally, 57 percent of the female Perry Preschool participants were single mothers compared to 83 percent of the non-preschoolers. Preliminary evidence for the children at age 40 indicates that benefits continue to accrue: Preschool participants continue to have higher earnings, fewer arrests, greater home ownership, and less drug use.1
A benefit-cost analysis by Barnett (1993) found $108,002 in benefits and $12,356 in costs per preschool participant (in 1992 dollars), a benefit-cost ratio of 8.74-to-1 based on data collected up until the participants were 27 years old. Of the total benefits, the public received $88,433, and $19,570 accrued to the program participants. The benefits to the public included $70,381 saved by potential victims of crimes never committed (based on typical settlements for such crimes) and in reduced justice system costs; $8,846 in higher taxes paid because of participants' higher earnings; $7,155 saved in education costs due primarily to lower grade retention and use of special education; and $2,918 in lower welfare costs. These benefits were partly offset by $868 in increased costs for the public funding of higher education. The benefits to the program participants included $21,485 in higher earnings and fringe benefits and $738 in childcare offset by a loss of $2,653 in welfare payments.
Another benefit-cost analysis of the Perry Preschool Project also found substantial net benefits. Karoly et al. (1998) found $49,972 in benefits and $12,148 in program costs in 1996 dollars—a benefit-cost ratio of 4.1-to-1 based on data collected up until the participants were 27 years old. Estimates of benefits by Karoly et al., differ from those by Barnett mostly because they exclude the benefits that derive from reductions in the intangible losses due to crime: the pain and suffering that crime victims experience. Thus, Barnett calculates $70,381 in benefits from less crime, while Karoly et al. estimate the benefits from less criminal activity at just $20,885. The benefits from reductions in the intangible losses due to crime do not, for the most part, go to government. Thus, while there is a large difference in the overall benefit-cost ratios calculated by Barnett (1993) and Karoly et al. (1998), the benefit-cost ratios they calculate for government savings are very similar: 2.5-to-1 by Barnett and 2.1-to-1 by Karoly et al.
The economic benefits of the Perry Preschool Project were probably under-estimated by both Barnett (1993) and Karoly et al. (1998). For example, neither of these benefit-cost analyses calculate the likely positive effects on the children born to participants who have higher earnings and employment and lower incarceration rates.2 Other savings to taxpayers and boons to government budgets, such as reductions in public healthcare expenditures, likely resulted from the program, but these benefits were not calculated either.
The Chicago Child-Parent Center Program
Description: The Chicago Child-Parent Centers (Chicago, Ill., 1967 to present) serve children from low socioeconomic status families. Twenty-three centers, all of which operate next to or within a wing of a Chicago public elementary school, provide half-day (three hour) preschool services for children aged 3 or 4. Nineteen of these centers also provide half-day or full-day kindergartens. Both the preschool and the kindergarten operate throughout the school year and for eight weeks in the summer. Thirteen of the centers provide additional educational services through the third grade when children typically reach 9 years of age. Annually, over 5,000 children are now attending the centers.
Each center is run by a head teacher who oversees the child education, parent involvement, community outreach, health, and nutritional prorams; that head teacher reports directly to the principal of the partner elementary school. Other staff include classroom teachers and aides, a parent resource teacher, and a school-community representative. Nurses, speech therapists, and other specialists are shared with the partner elementary school. Preschool and kindergarten classes have a teacher certified in early childhood education and a full-time aide; preschool classes typically have 17 children and kindergarten classes typically have 25. The child-to-adult ratio is often much lower because of the presence of parent volunteers. (Parents are required to volunteer at the center for one-half day per week, but not all centers have attained this level of involvement.) Each center selects its own curriculum, but all emphasize basic language and reading skills, as well as social and psychological development. Teachers and aides receive regular in-service training through the Chicago Public Schools' Department of Early Childhood Programs. The centers do home visits and strongly encourage parental involvement in classroom activities, field trips (e.g., to the Museum of Science and Industry and the zoo), and adult education classes. The centers also provide free breakfasts, lunches, and health services, including vision and hearing tests.
Results: Several different studies have followed large samples—typically 1,000 or more students—and compared outcomes for center students to non-center students. Data in these various studies have been collected periodically, with one ongoing study having analyzed data for as long as 19 years, or until the students were 22 years old. Table 2 summarizes some of the statistically significant outcomes of the preschool program.
Fuerst and Fuerst (1993) reported that center preschool students had higher scores on achievement tests at grade two as well as significantly higher graduation rates (62 percent versus 49 percent) than non-center students. Reynolds (1994) found that center preschool children scored higher on achievement tests than other comparable groups of children every year from kindergarten to 7th grade. In addition, that study found center preschool children had less need for special education (12 percent versus 22 percent) and significantly lower rates of grade retention (24 percent versus 34 percent).
The Chicago Longitudinal Study (CLS) has been following nearly all of the 1,150 students who attended center preschools in 1983–85 and center kindergartens in 1985–86 and comparing them to a control group of 389 children of the same age who met the eligibility criteria for participation in the intervention program and came from families of low socioeconomic status. The CLS has demonstrated that numerous benefits have been generated by the centers; presented here are just the benefits produced by the center's preschool program. For example, the study found that the center preschool children had significantly higher achievement test scores at ages 5, 6, 9, and 14. These children also spent less time in special education through age 18 (0.7 years versus 1.4 years), and had lower grade retention at ages 9 and 15 (19 percent and 23 percent versus 26 percent and 38 percent). Between the ages of 4 and 17, five percent of the preschool children had been victims of abuse or neglect compared to 10 percent of the nonparticipating group. Delinquency rates were significantly lower for the center preschool children through ages 13 and 14. By age 18, only 17 percent of center preschool children had been charged with serious crimes compared to 25 percent for non-center children—and charges for violent offenses were brought against nine percent of center preschool children but 15 percent of non-center children. Parental involvement with the schools was much higher among the parents of center preschool children than it was for the parents of non-center children. By ages 20 and 22, the high school graduation rates for center preschool children were 50 percent and 65 percent compared to just 39 percent and 54 percent for non-center children.
Reynolds et al. (2002) carried out a benefit-cost analysis of the Chicago Child-Parent Center Program. For the preschool program alone, they identified $47,759 in benefits and $6,692 in total costs in 1998 dollars—a benefit-cost ratio of 7.1-to-1 based on data collectd up until the participants were 22 years old. The benefits derived mainly from reduced public education expenditures due to lower grade retention and use of special education, reduced costs to the criminal justice system and victims of crime due to lower crime rates, increased income tax revenue due to projected higher lifetime earnings of center participants, and higher projected earnings of center preschool participants.
Once again, the benefits of the program were underestimated. For example, the benefits from less pain and suffering as a result of fewer crime victims were not included, nor were the likely gains from improved health, fewer pregnancies, and other positive life changes. Moreover, the likely benefits to offspring of center participants were not calculated.
Robert G. Lynch is associated professor and chairman of the department of economics at Washington College. His most recent book is Rethinking Growth Strategies—How State and Local Taxes and Services Affect Economic Development.
1. Schweinhart (2004).
2. See Rolnick and Grunewald (2003).