Preliminary Data Point to Promising Gains, Potential Problems
It may take several years to sort out how successful Even Start is in breaking down barriers that block the path to literacy for some parents and children, experts believe.
Although studies of other efforts to help children by bolstering their parents' literacy skills have been promising, Even Start's particular blend of early-childhood education, parenting training, and adult education has never been tested on such a large scale.
"The true test of the effectiveness of these programs,'' said Robert Popp, the director of research for the National Center for Family Literacy, will be "the persistence of effects as children go through elementary and high school.''
Preliminary data from the first stages of an Even Start evaluation show that the program is serving its intended population. Moreover, preschoolers involved appear to be reaping greater gains in skills associated with school readiness than they would without the program.
But the early data also highlight some implementation challenges faced by projects and signal some initial difficulties in involving families fully in all of the components.
The data are being gathered for a Congressionally mandated study being carried out through a contract awarded by the U.S. Education Department to Abt Associates Inc., with a subcontract to the òíã Research Corporation.
An unusual aspect of the study is that it relies heavily on grantees in gathering data.
Targeting Those in Need
An October 1991 report on projects funded in the program's first year and preliminary data from the second year--scheduled to be released in a report this summer--show the grantees are largely succeeding in implementing projects that serve the target population of disadvantaged, undereducated parents and their children.
The data, which cover about 120 projects in the 1989-90 and 1990-91 school years, show that among the adult participants:
76 percent had not finished high school.
70 percent had incomes under $10,000, and public assistance was the main source of income for 52 percent.
40 percent were single parents, and 65 percent were female.
More than half were age 29 or younger, and 15 percent were under 22.
43 percent of the parents were white, while 26 percent were black and 20 percent Hispanic.
The data show that projects, through collaborative arrangements within school districts and with other agencies, were offering the required project components as well as a wide range of support services and special events.
Still, the first- and second-year data also suggest that participation has varied widely among families and projects.
An estimated 95 percent of the families had at least one child participating in early-childhood education; 80 percent had at least one adult involved in parenting education; and 69 percent had at least one in adult basic education. All told, though, only about 54 percent of the families were fully participating in all three parts of the program.
"Adult basic education was the core service with the greatest amount of variability in implementation among projects,'' the first-year report concludes.
The researchers--who are soliciting feedback from grantees to verify and interpret the data--attribute the less-than-full participation to several factors. First, they say, it may reflect the timing of the data collection in cases where parents just entered the program or services were still being phased in. They also speculate that grantees may be unsure what to count as adult education, especially in programs that serve families mainly through home visits.
Families in Crisis
But the researchers say the data also suggest the adult-education component may be the hardest to sell.
Even when parents make a commitment to Even Start, said Stephen Murray, a vice president of RMC, "it may require a bit of encouragement or selling'' to prod their participation in adult education.
In addition, suggested Robert St. Pierre, a vice president of Abt and the project director for the evaluation, the programs reporting the lowest levels of participation may be serving "a much worse-off set of families'' who are "in crisis all the time.''
In those cases, helping parents to get other kinds of services may be "impinging somewhat on the educational focus of the program and perhaps delaying the time period before a family is fully able to participate,'' Mr. Murray explained.
Besides the issues related to recruiting, retaining, and motivating families, other barriers to implementation identified by the researchers included problems recruiting and training staff and coping with staff turnover, finding transportation for families, and communicating and coordinating with cooperating agencies.
Early Impact Seen
Previous studies have highlighted links between adult involvement in literacy training and improved outcomes for children. Data emerging from two longitudinal studies by the Center for Family Literacy involving 350 families in programs that combine education for parents and preschoolers also indicate the approach is paying off. (See Education Week, Sept. 4, 1991.)
The earliest data on Even Start outcomes, though far from complete, also appear encouraging.
The data, reflecting the second year of the program, show that 3- to 5-year-olds given tests using the Preschool Inventory "scored better than they would have just on the basis of normal development,'' said Mr. St. Pierre. Their gains, he added, were "pretty much commensurate with scores seen in other preschool programs'' geared for disadvantaged young children.
The test, developed for use in Head Start, takes about 15 minutes and measures such items as identification of objects, shapes, colors, and numerical concepts.
Some participants have raised concerns about the appropriateness of that or other tests for young children, project officials acknowledge. But they stress that the tests being used for the evaluation were meant to provide an overview of the children's skills and a basis for comparison with other programs. They were not meant to diagnose or track individual children, Mr. St. Pierre noted.
The first-year study noted that the national evaluation was funded about four months after Even Start grants were made, leading to a "somewhat slow start up for the evaluation.''
The shift this year from federal to state administration and a
transition between contractors have also caused some delays in
providing evaluation guidance to the newer grantees, federal officials
acknowledge. But they stress that in projects' first year, they are
more interested in basic descriptive data than in outcome