A r c h i v e d  I n f o r m a t i o n

National Evaluation of The Even Start Family Literacy Program, 1998

Chapter 7: What Are the Educational and Developmental Outcomes?

This chapter begins by summarizing the educational and developmental outcomes of Even Start as found in the current study and, where relevant, in the first national evaluation. As described earlier in this report, the outcomes generally reflect (unless otherwise indicated) data collected for participants in the current evaluation and more specifically for those participants who remained in Even Start long enough to participate in at least two rounds of data collection (e.g., pretest, posttest, and/or follow-up).⁹⁰

It is important to note that the Sample Study component of the current evaluation depended upon local projects administering child and adult tests and submitting data on outcome measures. Staff from all Sample Study projects were initially trained in late summer 1994, and did not participate in any other training. In each of the following two program years, the continuing projects received a manual documenting the procedures for test administration. Many projects worked hard to ensure that the annual data collection procedures and submission were carried out smoothly and accurately, yet the quality of the data submitted by local projects was extremely variable. This reflects staff changes at the project level, both among the individuals responsible for test administration (either Even Start or collaborating agency staff) and those responsible for submitting annual data. Another characteristic of Sample Study data is that the number of families per project for whom we had valid outcome measures ranged from 3 to over 100. The combined effect of these caveats is that we must interpret our findings with considerable caution because the data may not capture the impact of participation in Even Start as accurately as we had hoped.

Also, as described in Chapter 2, when we contrasted demographic and other characteristics of those families for whom we have both pretest and posttest data (or those with multiple waves) to those families with only pretest data (or those with only one wave), we observed significant differences between these two groups. Specifically, families with multiple waves of data are more likely, on average, to be employed, have higher incomes, and speak languages other than English at home. Families with multiple waves also are less likely, on average, to be headed by single parents. Mothers in the multiple wave group have, on average, completed nearly one more year of schooling than mothers with only pretest data. As a result of these systematic differences, the results we describe below reflect a bias in favor of continuing participants who, on average, may have more experience with education and who may well have greater supports in the home. What this means for our analyses of data from the Sample Study participants is that we may well be overestimating the effects of participation in Even Start. This important caveat should be held in mind when reviewing this chapter and subsequent sections of this report.

We focus first on the current study and present a detailed discussion of outcomes for children, followed by outcomes in parenting education, adult education, and other types of progress indicators such as attainment of a GED and improvement in employment status.

In the current evaluation, information on educational and developmental outcomes for Even Start participants was collected from nearly 1,700 families participating in the forty-seven projects that continued to participate in the Sample Study in 1996-97 (described in more detail in Chapter 2).⁹¹ Data were collected for the 1994-95, 1995-96, and 1996-97 program years. Across all three program years, projects offered, on average, about ten months of instructional service.⁹² The findings described in this chapter summarize what we have learned about two different cohorts: families who entered Even Start in the fall of 1994 and families who entered in the fall of 1995 (Exhibit 7.1). For each group of families, we have information on outcome measures collected over a potential period of up to two years of program participation; in general, we have pretest scores, posttest scores (valid only when pre- and posttest administrations are separated by at least three months for the Preschool Inventory and two months for all other measures), and follow-up scores (also only valid with the same guidelines on test administration dates).⁹³

Footnotes:

⁹⁰ For the purpose of characterizing baseline status on several outcome measures, we do report pretest or entry scores (or wave one) for participants for whom we may not necessarily have posttest or followup scores (or multiple waves).

⁹¹ We received data from only forty-seven projects for the 1996-97 program year. We include data, however, from the fifty-three projects that had submitted data for the previous program year. The other Sample Study projects were no longer operational or had fulfilled their obligations to the Sample Study by completing the required waves of assessments on families.

⁹² The most recent re-authorization of Even Start requires projects to provide year-round services, with some services occurring during the summer months.

⁹³ Most of the findings reported in this chapter (and in Chapter 8) are based on pretest and posttest data collected across two program years-1994-95 and 1995-96, with some additional data collected during the 1996-97 program year. While we do have limited wave three data on some measures for children, there are so few adults (generally under 15 percent of those with two waves of data) that we base the majority of our analyses on difference scores between waves one and two. We report statistical significance and difference scores as indicated throughout this chapter.

-###-

[ Summary of Participation Rates for Other Participant Groups ]

[ What Were the Child Development Outcomes? ]