Search

The Even Start program, designed to demonstrate ways of implementing and delivering family literacy services, has had strong evaluation requirements since its inception. An analysis was conducted to determine the types of evaluations programs had conducted and their findings concerning Even Start outcomes. A convenience sample of 4 state evaluations and 118 local evaluations from 19 states were used for the analysis. Findings indicated that programs rarely used data for program management and improvement. Almost all evaluations reported some information on project implementation, including descriptions of project structure and activities, recruiting and retaining families, and working with collaborative agencies. More than three-quarters of the evaluations contained information about program outcomes, although none incorporated the use of control or equivalent comparison families. Twenty-four evaluations were selected for detailed examination of outcomes. These evaluations showed consistent positive gains on measures of child and adult development. Half the reports gauged the size of outcomes by making comparisons to national data, to evaluations of other early childhood or adult education programs, or to normal development. Because local studies did not provide much useful information to local project directors, State Coordinators, or to the Department of Education, several recommendations were made, including changes in Even Start legislation language, provision of guidance regarding the amount of funding needed for conducting evaluations focusing on program improvement, and the establishment of a community of local evaluators. (Includes a list of principles for the continuous improvement approach.) (KB)

A general interest in attrition, or loss of units from a study, stems in part from the observation that the infrequency of attention to attrition exacerbates problems of data interpretation. As a substudy of the national evaluation of Project Follow Through, the potential biasing effects of attrition of subjects from the sites were investigated. Policy attrition, the administrative dropping of a unit; program attrition, loss of subjects due to mobility, dislike of the treatment, etc.; and sample attrition, units omitted because of deficient data, have been defined, but were not distinguished in this research. The relationship between attrition, pretest score, and income level were studied. In the few cases where additional variance was accounted for by the interaction of pretest or income with treatment within-site, an interpretation of the possible biasing effects of attrition was made in the evaluation of the specific site. It is suggested that in cases where differential attrition is found, the offending variable be considered for use as a covariate. (BW)

Follow Through is a longitudinal program instituted by the U.S. Office of Education in which a variety of curriculum models have been implemented in a number of locations nationwide. Because services are provided from kindergarten through third grade, many children enter and leave the program at points not coterminous with the treatment itself. The purpose of this study was to assess the effects of attrition on the national Follow Through evaluation. Relationships between child characteristics such as ethnicity, sex, family income, and mother's education and attrition group membership were investigated to determine whether the findings of the major analysis should be qualified. (Author)

This project summary reviews and extends the findings of prior reports made by Abt Associates, Inc. (Cambridge, Massachusetts) on the Nutrition Education and Training (NET) program, synthesizes evaluation efforts in nutrition education, and presents a set of conclusions based on the evaluations of nutrition programs. Chapter 1 presents background information on the NET program and gives a brief overview of the report. Chapter 2 describes the organizational framework of most nutrition programs and relates nutrition education activities to their hypothesized effects. Chapter 3 describes how the NET program is intended to work and how it has been implemented and investigates the implementation of programs from 1978 to 1980. Chapter 3 also reviews the legislative mandate behind the national NET program and presents a status report on the program by summarizing findings from surveys of NET state coordinators and a sample of local project directors. Models of three state programs are presented to show the diversity of responses to the program at the state level. Chapter 4 addresses the impact of the NET programs in Georgia and Nebraska, comparing the programs to other nutrition education programs also funded through the NET program or through the Food and Nutrition Service (Department of Agriculture). Conclusions are drawn about the effects of the programs on childrens' nutrition-related knowledge, attitudes, and behaviors. The final chapter sets forth the major conclusions of the report. (FG)

One specific correction model suggested by Cohen and Cohen (1975) is applied to data collected in the evaluation of a large-scale quasi-experimental program (Project Follow Through), and the effects of different assumptions about test reliability on the analysis results and on the conclusions of the evaluators are examined. The study determines whether the application of reliability or "true score" corrections alters the results obtained via an analysis employing uncorrected covariates in such a fashion as to appreciably change the policy-oriented conclusions of an evaluator. The data on which this paper is based were collected for the 1976 Follow Through evaluation and include measures on a total of over 5,000 children who began the program in kindergarten (Fall 1971) and completed it in third-grade (Spring 1975). Results indicate that application of true-score corrections using three separate reliability estimates to covariates employed on analysis of covariance did not change the conclusions of the Follow Through evaluators. (RC)

Describes two-generation programs and how they differ from earlier single-focus approaches to serve children and families. Results from six programs are reviewed and indicate mixed and modest success in promoting the development of children and improving the parenting skills and economic self-sufficiency of parents. Recommendations are presented for program improvement and future research. (GR)

This study assessed the feasibility of conducting relatively low cost later effects studies of Follow Through (FT) programs without extensive new data collection, relying instead on existing data resources including the national FT data base, sponsor and site data bases, school and community records, and potential new data collection of modest scope. Six data sets were selected as meeting the criteria (adequate sample size, availability of a wide range of measures, and good local comparison groups). The following research questions were identified as being suitable for study using existing data bases: (1) What are the causal relationships among academic achievement, IQ, process measures, and affective measures? (2) Do achievement effects associated with FT participation increase, decrease, or stabilize after program completion? (3) What is the relative importance of the current academic year and the child's previous academic history as determinants of academic success? (4) What are the causal relationships between academic performance and school attendance in and after FT? (5) What is the relationship between post-FT academic performance and length of FT participation? (6) Is it possible to predict post-FT patterns on different measuring instruments? (7) What differences in post-FT performance exist among different models at the same site? (Author/MK)

This paper summarizes a report on a search for potential new educational approaches and assesses the readiness of these approaches to be implemented within Follow Through (FT) programs. Evaluation of potential FT sponsoring agencies focused on program implementation and maintenance, rather than on effectiveness in improving student outcomes. Three types of approaches were studied: home-based, special populations, and school-based. Eighteen potential sponsors were selected for in-depth study. They represented three basic approaches to education: behaviorist, cognitive developmental, and psychodynamic. Agencies were evaluated in terms of program base, targeted special populations, number of years in operation, targeted age/grade groups, institutional type, and validation by the Joint Dissemination Review Panel of the Educational Division of the Department of Health, Education, and Welfare. Readiness of potential sponsors to implement FT programs was assessed through studies of programs currently being sponsored by the agencies. Abstracts of current programs are included. (MK)

The original plans for the Follow Through (FT) program were to provide a program extending the Head Start community action model to the public schools which would serve all disadvantaged children. Because of the small budget allocation authorized by the Congress, the United States Office of Education reformulated FT as an experimental program of "planned variation" that was to introduce systematically a variety of educational approaches into kindergarten through third grade. The experiment was to determine the relative efficacy of each approach. Presently, FT is in a period of transition as it clarifies its future focus. This study consists of a search for potential new approaches and an assessment of the readiness of these approaches to be implemented within the FT program. This first part of a two-part report contains the methods used in identifying essential program characteristics, the characteristics themselves, the methods used to identify potential new sponsors, a summary of the potential sponsor's approaches, and a global assessment of their readiness for implementation. (Author/MK)

This document summarizes a study designed to assess the feasibility of conducting relatively low cost later effects studies of Follow Through (FT) programs without extensive new data collection, relying instead on existing data resources including the national FT data base, sponsor and site data bases, school and community records, and potential new data collection of modest scope. Six data sets were selected as meeting the criteria (adequate sample size, availability of a wide range of measures, and good local comparison groups). Research questions that can be addressed by existing data bases were identified as follows: (1) What are the causal relationships among academic achievement, IQ, process measures, and affective measures? (2) Do achievement effects associated with FT participation increase, decrease, or stabilize after program completion? (3) What is the relative importance of the current academic year and the child's previous academic history as determinants of academic success? (4) What are the causal relationships between academic performance and school attendance in and after FT? (5) What is the relationship between post-FT academic performance and length of FT participation? (6) Is it possible to predict post-FT patterns on different measuring instruments? (7) What differences in post-FT performance exist among different models at the same site? (Author/MK)

The Wiley and Harnischfeger school effects model treats quantity of instruction as a predictor of school performance. Despite the importance of the model for the study of school and program effects, including the incremental effects (if any) of Title I and other compensatory efforts, very little information is available on its practical application. Further, with or without the benefit of the model, little is known about the amounts and kinds of instructional services students actually receive in today's schools, either through locally or State supported regular instructional programs or Federally supported Title I programs. The research described in this document provides information in both of these areas. It includes a brief description of an instrument designed to obtain quantity and quality of schooling data, presents descriptive information about the amounts and kinds of services which various types of students receive, and addresses the following questions: (l) What amounts and kinds of services are actually delivered in today's elementary schools? (2) To what extent and in what way(s) does classroom instruction received by Title I students differ from that received by other students? (3) Does the amount of noncompensatory instruction received by Title I students appear to be related to the amount of compensatory instruction received? and (4) Does the delivery of compensatory services to Title I students differ in terms of group size and type of instruction from the delivery of noncompensatory services to these same students? (Author/GC)

This article considers the advantages and disadvantages of having evaluation users increase the specificity of evaluation requirements through a review of legislative requirements behind several congressionally mandated national-level evaluations, and an in-depth examination of one congressionally mandated evaluation in which very specific design parameters were included in the authorizing legislation. (Author/PN)