You are seeing this message because your web browser does not support basic web standards. Find out more about why this message is appearing and what you can do to make your experience on this site better.

www.HFRP.org

The Harvard Family Research Project separated from the Harvard Graduate School of Education to become the Global Family Research Project as of January 1, 2017. It is no longer affiliated with Harvard University.

Terms of Use ▼

Program Description

Overview Foundations, Inc. is a private, nonprofit organization that builds community and strategic partnerships to promote the expansion of educational programs that improve performance and enhance student achievement in school and during nonschool hours. Foundations operates extended-day enrichment programs before school, after school, and during the summer. Programs are implemented in many urban and rural schools in the mid-Atlantic and northeastern portions of the United States, serving children in all grade levels. Foundations also provides technical assistance and support services to schools, school districts, and other education and community organizations.
Start Date 1992
Scope national
Type after school, summer/vacation, before school
Location urban and rural
Setting public schools, community centers, faith-based organizations, recreation centers
Participants pre-K through 12th grade students
Number of Sites/Grantees 87 sites
Number Served Over 4,000 (2002–2003)
Components The Foundations' programs feature curricula that are focused on content-rich experiences with daily activities emphasizing academic subjects and the physical, social, and emotional development of participants.

Each yearlong curriculum manual focuses on a theme designed to reinforce academic skills in reading, writing, and mathematics. Sample themes include travel, discovery, celebrations, and inventions. Daily lessons incorporate technology whenever possible and also include recreational activities and time for social interaction. Lessons feature cooperative group work as well as independent projects.

All curricula are literature-based and involve studying award-winning books by a broad array of well-known authors. The books and materials expose students to many different cultures in the U.S. and foreign countries. The lessons accommodate multiple teaching and learning styles. Students learn by reading, writing, listening, reasoning, and through hands-on activities. Each curriculum unit is aligned with widely accepted national standards.

The curricula also take into consideration the need for children to interact with other children and adults, to play, and to receive support and encouragement. For this reason, participants spend time engaged in other activities including field trips, homework assistance, and computer lab time. The program also emphasizes family involvement, providing opportunities for families to volunteer in the classroom or participate in other ways.

The student-to-adult ratio in the program is approximately ten to one. The program employs staff with prior experience working with children and teachers and program coordinators who have an associates, bachelors, or masters' degree in education or a related field. A sliding scale is used to determine program fees based on each family's ability to pay. Staff attempt to communicate regularly with participants' day school teachers, thereby providing some coordination of after school activities with the experiences students have throughout the day; this is particularly true for the homework assistance portion of the program.

In addition to operating programs, Foundations develops technical assistance materials and curricula. Foundations is currently developing a guide under a grant from the Charles Stewart Mott Foundation to help after school providers and practitioners merge academic content with after school programming. Academic Content, After-School Style includes chapters that outline curricula and standards in English, math, and science and chapters on reading and literacy support, homework and helping in general, working with English as a second language learners, planning and assessment, and resources for children and youth from kindergarten through high school.

Foundations is also developing eight middle school modules. Each guide contains 40 lessons designed to support academic, social, and interpersonal growth for pre-teen and teenage participants. The manuals can be used in after school and summer programs in a wide variety of settings.
Funding Level $8.3 million annually (2002)
Funding Sources 21st Century Community Learning Centers federal grants, tuition, Schwartz Foundation, Lenfest Foundation, Annenberg Foundation, Annie E. Casey Foundation, William Penn Foundation, Geraldine Dodge Foundation, Nelson Foundation, Abell Foundation, Auerbach Foundation, Charles Stewart Mott Foundation, Kansas Health Foundation, Phoebe W. Haas Foundation, Toppel Family Foundation, and a number of government, corporate, community, and private entities


Evaluation

Overview Foundations' programs have been evaluated annually since 1997–1998. We have profiled the initial evaluation reports (covering the 1997–1998 and 1998–1999 school years), as well as the most recent available evaluation (of the 2001–2002 school year). Two intermediate evaluation reports by RAND Education look at the 2000–2001 school year and are available on request from RAND. For the 1997–1998 and 1998–1999 evaluation reports, Foundations after school programs operating in six elementary schools were evaluated to determine if the programs had an impact on academic achievement. The 2001–2002 evaluation also focused on student achievement on standardized tests among students participating in Foundations after school programs, but looked at 19 elementary schools in Pennsylvania, New Jersey, and Florida.

In addition to evaluating Foundations students' performance on standardized tests and comparing it with non-Foundations students, the organization conducts twice yearly program evaluations using the Quality Assurance System (QAS). The QAS developed by Foundations assesses program implementation and pinpoints areas for improvement. The QAS uses multiple data sources to score the Elements (written as rubrics) that comprise each program Building Block. This assessment system gives a detailed picture of program strengths and weaknesses and has applicability for a broad array of extended-day programs. The QAS tool is available through Foundations, Inc.
Evaluator RAND Education (1999 report)

Gansk & Associates (2002 report)
Evaluations Profiled Foundations School-Age Enrichment Program: Evaluation of Student Achievement (RAND Education)

Improvements in Math and Reading Scores of Students Who Did and Did Not Participate in the Foundations After School Enrichment Program During the 2001–2002 School Year (Gansk & Associates)
Evaluations Planned A 2002–2003 evaluation is being conducted with students in grades 1–5.
Report Availability Hamilton, L. S., Le, V., Klein, S. P. (1999). Foundations School-Age Enrichment Program: Evaluation of student achievement. Santa Monica, CA: RAND Education.

Le, V., & Hamilton, L. S. (2001a). Examining test score gains among participants of the Foundations after-school program (PM-1178-EDU). Santa Monica, CA: RAND Education.

Le, V., & Hamilton, L. S. (2001b). Achievement gains in math and reading by participants of the Foundations after-school enrichment program (PM-1265-EDU). Santa Monica, CA: RAND Education.

Klein, S. P., & Bolus, R. (2002). Improvements in math and reading scores of students who did and did not participate in the Foundations After School Enrichment Program during the 2001–2002 school year. Santa Monica, CA: Gansk & Associates. Available at: www.communityschools.org/foundations.pdf (Acrobat file).


Contacts

Evaluation Vi-Nhuan Le
RAND Education
1700 Main St.
Santa Monica, CA 90407
Tel: 310-393-0411 ext. 7574

Stephen P. Klein, Ph.D.
GANSK & Associates
120 Ocean Park Blvd., #609
Santa Monica, CA 90405
Tel: 310-396-3030
Email: rbolus@netzero.net
Program Rhe McLaughlin
Director of Evaluation
Foundations, Inc.
Moorestown West Corporate Center
2 Executive Dr., Ste. 1
Moorestown, NJ 08057-4245
Tel: 856-533-1600
Email: rmclaughlin@foundationsinc.org
Profile Updated May 27, 2003

Evaluation 1 (RAND Education): Foundations School-Age Enrichment Program: Evaluation of Student Achievement


Evaluation Description

Evaluation Purpose To determine whether Foundations participants improve in the academic skill areas of reading and math, and if so, to what extent.
Evaluation Design Quasi-Experimental and Non-Experimental: Data were collected from six elementary schools. At two of the six schools included in the evaluation, data were collected both from fourth grade program participants and fourth grade nonprogram participants. One of these schools was used both years while the other was replaced by another school during the second year. This nonparticipant comparison group, while not the result of random assignment, appeared to be similar to the participant sample in most respects, including ethnicity and socioeconomic status. Data were also collected from program participants in grades one through five at all six elementary schools.
Data Collection Methods Secondary Source/Data Review: Report cards for program participants at all six schools were collected during the second year of the evaluation, 1998–1999.

Tests/Assessments: The Terra Nova Reading/Language Arts and Mathematics Computation Tests were administered twice (pretest and posttest) each year to all program participants at the six schools and to the fourth grade nonparticipant comparison group at two schools. The Reading/Language Arts Test contains several reading passages followed by questions. The test also includes language, vocabulary, writing, and editing skills items, all in multiple-choice format. The Mathematics Computation Test contains problems involving arithmetic operations appropriate to each grade level. There are no word problems on the math test. Data were collected in February 1997, June 1998, November 1998, and June 1999.
Data Collection Timeframe Data were collected during the 1997–1998 and 1998–1999 school years.


Findings:
Summative/Outcome Findings

Academic During the first year of the study, the scores of fourth grade participants increased more than fourth grade nonparticipants. In all three subjects-reading, language arts, and math-participants had significantly greater gains in scores from the pretest to the posttest (p<.05).

Differences between the participant and comparison groups in reading and language arts score gains were greater than differences in math score gains.

The gain in reading score from the pretest to the posttest for the Foundations participants was .63 standard deviations as compared to only .09 standard deviations for the comparison group.

In contrast, the gain in math score from the pretest to the posttest for the participants was .77 standard deviations as compared to .4 standard deviations for the comparison group.

Foundations students began the year with scores only slightly higher than the comparison group, but their posttest scores were substantially higher, with the Foundations students scoring well above the national median and comparison students well below it.

During the second year of the study, the gains in scores of fourth grade participants and fourth grade comparison group students were nearly identical, with participants achieving slightly higher gains.

The gain in reading scores for fourth grade participants and comparison group students was much lower the second year of the study than the first year.

In the second year of the study, fourth grade Foundations participants had substantially higher pretest scores in language arts and math than their counterparts in the comparison group that could not be solely accounted for by students being in their second year of program participation, suggesting systematic differences between the two groups.

During the first year, Foundations students at all six schools in grades 1–3 experienced average increases in scores from the pretest to the posttest that were quite large, approximately one standard deviation. These results exclude grade 2 reading and language arts gains, which were .33 and .34 standard deviations, respectively.

The smallest gains accrued to fifth graders, with scores actually declining in math and language arts by .08 and .1 standard deviations, respectively.

During the second year, Foundations students at all six schools generally demonstrated gains from the pretest to the posttest. First and second graders demonstrated the largest gains with first graders improving more than one standard deviation in all subjects and second graders improving .5 standard deviations in reading, greater than .8 standard deviations in language arts, and almost one standard deviation in math.

Improvements from pretest to posttest were smaller for third graders, but were not statistically significant.

Fifth graders showed the smallest improvements, actually declining by .05 standard deviations in language arts in the second year.

Improved performance on the standardized tests was not limited to particular subskills or item types. Foundations students in all grades showed substantial and comparable gains in the two skills measured by the math test-add whole numbers and subtract whole numbers-during both study years (with the exception of third graders in 1998–1999 who only showed gains in subtracting).

During the first year, in reading and language arts, substantial improvement was seen in analyzing text during the first year. Substantial improvement was also seen in editing skills in second, third, and fourth grades, writing strategies in third grade, and sentence structure in fourth grade.

Smaller, yet statistically significant, gains were also made in other reading/language arts areas. During the second year, gains were observed in every subskill among first and second graders. Third and fourth graders showed smaller improvements in editing skills, basic understanding, and analyzing text. Fifth graders failed to improve in a variety of skill areas.

Birth date did not have any association with score gains among Foundations students. In other words, for any given grade, the age of participants was not found to be associated with test score gains.

Gender was related to reading and language arts performance during the first year. Females exhibited larger gains than males. On average, the female language arts gain was .77 standard deviation units as compared to an average of .43 standard deviation units for male participants. No gender difference was found in the second year of the study.

There were significant differences among schools during both years, which tended to be consistent across all subject matter areas. Improvement at one school was more than one half of one standard deviation larger in every subject than improvement at any other school during year one and another school had consistently lower levels of improvement. The difference in these schools persisted during the second year of the study.

During the second year, participants' grades in math, language arts, and reading improved, although the gain was statistically significant only in math and only at one school. Other nonstatistically significant improvements were evident in participants work habits, ability to get along with others, and ability to work independently.

For students who participated in Foundations programs over both years of the study, in most cases, the scores increased, particularly in math. Often, the second-year pretest scores, however, were actually lower than the first-year posttest scores suggesting that skills declined over the summer. For third graders in 1997–1998, this was true-math and language arts scores for more than half of the group (only 11 students) were higher in June 1998 than in November 1998.

Evaluation 2 (Gansk & Associates): Improvements in Math and Reading Scores of Students Who Did and Did Not Participate in the Foundations After School Enrichment Program During the 2001–2002 School Year


Evaluation Description

Evaluation Purpose To determine whether Foundations participants improve in the academic skill areas of reading and math, and if so, to what extent. In addition, gains found during the 2001–2002 evaluation are compared to gains found in previous years' evaluations conducted by RAND Education (see Reports Available above).
Evaluation Design Quasi-Experimental: Data were collected from students involved in Foundations enrichment programs in 19 schools located in three counties in Pennsylvania (Philadelphia, Delaware, and Franklin), Cumberland County, New Jersey, and Palm Beach County, Florida. Almost all children in these programs were in first through fifth grades. Results for Foundations program participants were compared to a national norm group and to students at seven of the nineteen schools with comparable skills and background characteristics who were not in a Foundations program. Only those students completing both the pretest and posttest on either test were included in the analysis (n=384 for math and 406 for reading for the program group, n=583 for math and 646 for the comparison group).

The Foundations group and the comparison group had nearly identical mean pretest scores on both math and reading (standard deviations were also nearly identical). Some of the comparison group students were selected from the same schools as Foundations students, while others were selected from demographically comparable schools. The comparison group students represent one to two classes of students from the selected schools. Depending on the size of the school, it may or may not have been possible to select two classrooms, in which case only one classroom was selected. From these classrooms, the evaluators collected data on all students not participating in Foundations programming.
Data Collection Methods Tests/Assessments: Students were administered the CTB/McGraw-Hill CAT-5 Mathematics and Reading Comprehension tests in both the fall of 2001 and again in the spring of 2002. Beginning in the 1999–2000 school year, Foundations switched from the CTB/McGraw-Hill Terra Nova tests of Mathematics Computation and Reading/Language Arts to the CAT-5. The switch was undertaken because of concerns that the repeated use of the same measure across years might yield spurious score gains that were not associated with a commensurate increase in ability. The CAT-5 Math Concepts and Application test assesses a broader range of skills and knowledge than does the Mathematics Computation portion of the Terra Nova. Because of time constraints, only the Reading Comprehension section of the CAT-5 was administered, meaning that other language arts skills and other aspects of reading proficiency (such as word analysis) were not assessed.

CTB/McGraw-Hill uses an item response theory (IRT) methodology to convert raw scores into scale scores, thereby putting scores from different forms of the CAT-5 onto the same 100 to 900 point scale of measurement. This means that the evaluators could combine and average scale scores across grade levels. Differences in mean scale scores from the pretest to the posttest can be translated into effect sizes by dividing the difference by the standard deviation of the pretest scores. CTB/McGraw-Hill also computes normal curve equivalent scores (NCEs) for each student on each section, allowing evaluators to assess how far above or below the mean these students' scores fall relative to a national norm group of students who are at the same grade level. The evaluators converted NCEs to their corresponding National Percentile (NP) scores to assess whether the improvements in the students' scores during the year were keeping pace with the improvement in the national norm group during a corresponding number of months.
Data Collection Timeframe Data were collected during the 2001–2002 school year.


Findings:
Summative/Outcome Findings

Academic There were highly statistically significant improvements in both reading and math scores between pretest and posttest (p<.001). For scale scores, the effect sizes for math (mean effect size = .91 across grade levels) were generally greater than those for reading (mean effect size = .76 across grade levels).

Foundations students were progressing faster than the national norm group in math skill development and they were keeping pace with the norm group in reading. (Foundations students were at the 51st percentile in math at the pretest and the 61st percentile at the posttest; for reading they were at the 56th percentile at the pretest and 58th percentile at the posttest.)

A regression analysis that controlled for grade level and school found that the Foundations students' overall average improvement in scale scores between the pretest and posttest was significantly greater than among non-Foundations comparison group students (p<.001). While Foundations students gained an average of 45 scale score points in math and 38 scale score points in reading, comparison group students gained only 26 and 17 points, respectively. A t-test comparison of the differences in mean gain scores produced statistically significant effects (p<.05) for almost all grade levels and subjects. (The only exceptions being grade 4 in math and grade 5 in reading, which did not show statistically significant differences.)

The effect size for Foundations students in math was .91, while for comparison group students the effect size was .52 (a difference of .39 standard deviation units). The effect size for Foundations students in reading was .76, while for comparison group students the effect size was .35 (a difference of .41 standard deviation units).

The mean effect sizes in the 2001–2002 school year were larger than those found in the analysis of the 2000–2001 school year data. Specifically, the mean math and reading effect sizes across all five grade levels during the 2001–2002 school year were .91 and .76, respectively. The corresponding values for the 2000–2001 school year were .51 and .49.

Younger children tended to show larger improvements (as measured by effect sizes) than older students. For example, in 2001–2002, the effect sizes for grades one and two were 1.20 and 1.38, respectively. The corresponding effect sizes for grades four and five were .34 and .88, respectively.

© 2016 Presidents and Fellows of Harvard College
Published by Harvard Family Research Project