Programs that Work
Student Achievement Guarantee in Education (SAGE)
Children Succeeding in School
Students performing at grade level or meeting state curriculum standards
Age of Child
Early Childhood (0-8)
Type of Setting
Out of School Time
Type of Service
Type of Outcome Addressed
Cognitive Development / School Performance
Evidence Level (What does this mean?)
The Student Achievement Guarantee in Education (SAGE) program is a statewide effort in Wisconsin to increase the academic achievement of children living in poverty. The key mechanism used to achieve this goal is a reduction of the student-teacher ratio in kindergarten through third grade to 15 to 1. In addition to class size reduction, schools participating in the program are expected to implement curricula with a rigorous academic focus, engage in professional development and accountability plans, and develop "lighted schoolhouse" before- and after-school programming with activities for both students and community members.
SAGE was created as a five-year pilot program in 1995. The program was phased in over a three-year period beginning with kindergarten and first grade in 21 school districts during the 1996-1997 school year, with the addition of second grade in the second year, and finally third grade in the third year. SAGE schools changed class sizes in a number of ways in order to meet program requirements, including
- a single classroom in which the student-teacher ratio is 15 to 1
- a shared-space classroom in which a classroom is fitted with a temporary wall, thereby generating two classroom spaces, each with one teacher and approximately 15 students
- a dual-teacher classroom in which two teachers work collaboratively with approximately 30 students
- a floating-teacher classroom in which the room ordinarily has between 16 and 20 students, except during reading, language arts, and mathematics instruction when another teacher joins the class to reduce the ratio to 15 to 1.
When SAGE began in 1996 the program was limited to districts with high poverty schools. In 2000 the law was changed to allow any Wisconsin school to apply. Approximately 500 schools participated in the 2005-06 school year.
A five-year evaluation of the SAGE demonstration project was conducted by the School of Education at the University of Wisconsin-Milwaukee (UWM). The evaluation involved 30 schools from 21 school districts across Wisconsin. SAGE was first implemented with 80 kindergarten, 96 first-grade, and 5 mixed kindergarten and first-grade classrooms at the start of the 1996-1997 school year. An additional 113 second-grade classrooms were added in 1997, and 88 third-grade classrooms in 1998. The UWM evaluation compared students in SAGE schools with a group of 14 to 17 non-SAGE comparison schools located in SAGE school districts. Comparison schools were not matched to SAGE schools, but resembled SAGE schools in terms of family income, achievement in reading, kindergarten through third-grade enrollment, and racial composition.
In the fall of 1996, 30 schools (seven of which were in Milwaukee) implemented the SAGE program. Over the course of the year, the study involved 3,614 students and 220 teachers. Across all years of the evaluation (1996-1997 through 2000-2001), SAGE and comparison schools were composed of approximately 25 percent African American students, 45 percent Caucasian students, 10 percent Native American students, and 20 percent students of other ethnicities (comparison schools tended to have a slightly higher percentage of Caucasian students). An average of approximately 55 percent of SAGE students qualified for free lunch, as compared with approximately 45 percent of comparison students, and a slightly higher percentage of SAGE students had repeated a grade, spoke English as a second language, or were classified as having special education needs.
The primary method of evaluation to determine the impact of class-size reduction was the annual administration of the Comprehensive Test of Basic Skills (CTBS) to both the SAGE and comparison schools. The complete CTBS battery includes subtests in reading, language arts, and mathematics. Although SAGE was implemented in kindergarten classrooms, students in kindergarten were not tested until the following year when they were first-grade students. Analyses of findings were conducted to assess the impact of SAGE participation on all students, as well as to compare performance of African American SAGE students to white SAGE students, and African American SAGE students to African American comparison students. Only those students who participated in the entire program, that is, who were present in the 1997-1998 SAGE and comparison first-grade classrooms, the 1998-1999 SAGE and comparison second-grade classrooms, and the 1999-2000 third-grade classrooms are included in the presented results. This group consisted of a total of 1,212 students, including 779 SAGE students and 433 comparison students. It should be noted that not all students completed all CTBS testing each year.
A further study of SAGE was conducted by Webb et al. (2004), and it assessed program effects on state assessments of students in grades three and four. The original 30 SAGE schools (approximately 4,900 students) and 10 to 17 comparison schools in each grade (approximately 7,800 students) took part in the study. Data used in the analysis included the CTBS scores in reading, language arts, and mathematics collected by UWM researchers, the grade three Wisconsin Reading Comprehension Test (WRCT), and the grade four Wisconsin Knowledge and Concepts Examination (WKCE).
Key Evaluation Findings
SAGE was evaluated over numerous years at different grade levels from first through fourth grades. Overall, findings suggest that SAGE participants performed significantly better than comparison children on the majority of reading, language arts, and mathematics achievement tests at all grade levels. There is evidence of a cumulative beneficial effect of the intervention over multiple years.
Specifically, in reports published in each year of the evaluation (1997 through 2001), the UWM evaluation team found the following:
Achievement Outcome Findings, 1997 (Implementation Year One)
- After one year, when controlling for pretest scores, family income, school attendance, and race/ethnicity, SAGE first-grade students scored significantly higher on the total score and on reading, language arts, and mathematics subtests of the CTBS than did comparison students.
- African American SAGE students scored lower on the CTBS pretest total score than did African American comparison students, but they scored significantly higher than comparison students on the posttest.
- African American male SAGE students attained comparable or higher change scores from pre- to posttest than did African American female SAGE students. Conversely, in comparison schools, change scores for females exceeded change scores for males on every subtest as well as on the total score.
Achievement Outcome Findings, 1998 (Implementation Year Two)
- Second-grade classrooms were included in 1997-1998, and for all analyses, SAGE classrooms scored significantly higher than comparison classrooms. The results suggest that the positive first-grade effects of the SAGE program were maintained but did not increase in second grade.
- For first-grade classrooms, African American SAGE students scored significantly higher than African American comparison students on every posttest and on the total CTBS score. They also made significantly larger gains than comparison students on every pre- to posttest change score.
Achievement Outcome Findings, 1999 (Implementation Year Three)
- For first-, second-, and third-grade classrooms in 1998-1999, SAGE students significantly outscored comparison students on all posttests, with the exception of reading scores in second grade.
- African American SAGE students made significantly larger gains than African American comparison students from pre- to posttest and significantly outscored comparison students on all posttest scores.
Achievement Outcome Findings, 2000 (Implementation Year Four)
- First-grade students were not tested in 1999-2000; results were limited to second- and third-grade students.
- Second-grade SAGE students showed a significant achievement advantage over comparison group students for all subtests (when the first-grade pretest was used to adjust for achievement differences).
- African American SAGE students made significantly larger gains than African American comparison students from pre- to posttest and surpassed comparison students on all posttests.
- While African American students, as a group, scored significantly lower than Caucasian students in both SAGE and comparison schools, the gap between African American and Caucasian students was larger in comparison schools.
- SAGE third-grade students showed significant improvement over their comparison school counterparts from the beginning of first grade to the end of third grade across all academic areas.
- Using the first-grade pretest as the baseline score, African American SAGE students scored significantly higher than African American comparison students on the total score and on every CTBS subtest except for language arts.
- Gains made by African American versus Caucasian students were significantly better in SAGE schools from the beginning of first grade to the end of third grade. The opposite pattern was observed in comparison schools.
Achievement Outcome Findings, 2001 (Implementation Year Five)
- First- and second-grade students were not assessed in 2000-2001, limiting results to third graders (who had participated since kindergarten).
- Third-grade SAGE students scored significantly higher than comparison students on the total CTBS score and on all subtests.
- For African American students, statistically significant change scores from first grade to third grade were found on all CTBS tests, with SAGE students outperforming comparison students. The only exception was language arts, for which no statistically significant effects were found.
- African American students continued to score significantly lower than Caucasian students on total scale score and on all subtests, regardless of whether they were SAGE or comparison school students. No significant differences in the gains made by African American students versus white students were observed for this group of students.
The study by Webb et al. (2004) reported:
- By reanalyzing the data collected by the University of Milwaukee research team, Webb et al. (2004) replicated the findings that the SAGE program had a significant cumulative effect from the beginning of first grade through third grade in all three CTBS content areas (reading, language arts, and mathematics). The greatest cumulative effect was in mathematics, with a more than 13-point difference in favor of the SAGE students. These data indicate that there is a benefit to students for being in SAGE classrooms over multiple years.
- These effects were not sustained into fourth grade on the WKCE in reading, language arts, or mathematics tests, with SAGE students scoring statistically equivalent to comparison students. Similarly, no cumulative effect for the SAGE program was found for the grade 3 WRCT. The results on the grade 3 WRCT and the grade 3 CTBS reading test produced conflicting results for the effectiveness of SAGE; WRCT results indicated no significant effect, whereas the CTBS results indicated significant cumulative effects.
Public or private elementary schools
SAGE is funded by the state of Wisconsin and administered by the Department of Public Instruction. Wisconsin state law limits SAGE funding to $2,000 per low-income student, an amount that has been unchanged since the program’s inception. Since 2000-2001, thirty-three schools have dropped out of the program, many citing financial issues. If the $2,000 per student had been increased by the same Consumer Price Index as district revenue limits, the aid amount would become $2,405 per student for 2005 and $2,529 for 2007.
The SAGE program consists of four components: class-size reduction (maximum 15 to 1 student-to-teacher ratio), rigorous curriculum, professional development, and "lighted schoolhouse."
- Only the class-size-reduction portion of the program has been uniformly implemented and evaluated among SAGE program sites.
- A "rigorous curriculum" is defined by SAGE as employing a curriculum that meets state standards. Schools are permitted to determine the specific curriculum that best fits their needs.
- Each SAGE school must create a staff professional development and accountability program.
- Every SAGE school is required to implement a "lighted schoolhouse" program, providing before- and after-school extended-hours programming. Schools are expected to tailor this program to meet their resources and needs.
The class size reduction component of SAGE has no special staffing needs, other than the specific ratio of 15 students per teacher. Classrooms are taught by the regular classroom teachers.
Issues to Consider
This program was given a "promising" rating. While the evaluations found positive program effects on outcomes of reading, language arts, and mathematics, SAGE and comparison schools were not particularly well-matched, and data on group characteristics indicate that there were meaningful discrepancies between the two populations of students. As such, it is possible that observed outcomes may have been influenced by these preexisting differences. Further, observed changes after the start of the program show a consistently unstable level of enrollment in both the SAGE and comparison schools. For example, during the second year of the program, only 42.3 percent and 44.3 percent of the student population at study and comparison schools, respectively, were ongoing. Of the students at study and comparison schools, 18.8 percent and 18.6 percent, respectively, withdrew, and 39.0 percent and 38.0 percent of students, respectively, were new enrollees. Researchers did not conduct an analysis to determine what impact this might have had on test scores.
It is also important to consider that there is a possible selection bias regarding the way in which schools were designated as SAGE participants. Participating districts were allowed to select the SAGE school from among all of the schools in the district that met SAGE inclusion criteria. No information is given on how districts made this selection. It is possible that districts selected certain schools based on their anticipated success and higher level of program participation or higher level of school functioning (as could possibly be inferred by the baseline discrepancies between SAGE and comparison schools). Without any further analysis of this factor, it cannot be ruled out as a potential contributor to the demonstrated outcomes.
Overall, analyses of test results at the classroom level suggest that students in smaller classrooms tended to score significantly higher in language arts, mathematics, and reading as well as the total CTBS score after adjusting for individual pretest results, socioeconomic status, and attendance. In other words, classrooms with fewer students were more likely to have higher class average achievement scores and were more likely to contribute to closing the achievement gap between African American and white students than were classrooms with a higher number of students.
Only the class-size reduction component of SAGE has been explicitly evaluated, as participating classrooms closely adhered to the 15 to 1 student/teacher ratio. The remaining three initiatives—rigorous curriculum, professional development, and lighted schoolhouse—were not directly examined in the UWM evaluation. The evaluation team felt that because these three initiatives were loosely executed and not uniformly implemented across sites, a single standard of measure was not possible. The UWM authors indicated that they are fairly comfortable in concluding that the impact of reduction in class size is most likely the primary cause of observed student outcomes. However, as these additional initiatives were implemented to some extent in some of the SAGE schools, they cannot be entirely ruled out as contributing to the findings. As such, it is not possible to ascertain whether class-size reduction alone would generate the test-score gains observed, or whether all four program components are required to produce these gains.
In addition, the results indicate that although SAGE students outperform comparison students in terms of overall test scores throughout the duration of the program, after the first grade, comparison students begin to narrow the achievement gap. This calls into question whether gains initially made by program participation will translate into a continued long-term advantage and enhanced performance on the part of SAGE students.
This analysis also did not take into consideration the different approaches that schools and teachers used to reduce class size. It may be that some approaches are more effective than others for increasing performance on the state assessments. It is also feasible that when the SAGE program is administered in specific ways, students in primary grades are given an advantage that is not as meaningful for students in later grades.
Study results raise the possibility that SAGE has a more positive effect on African American students than on white students. The same finding was not evident when students from low-income families were compared with students from high-income families.
Various school districts throughout Wisconsin are currently implementing SAGE. A current site list is available at http://dpi.wi.gov/sage/schools.html.
Janice S. Zmrazek
SAGE Program Coordinator
Wisconsin Department of Public Instruction
P.O. Box 7841
Madison, WI 53707-7841
phone: (608) 266-2489
The SAGE Web site, http://dpi.wi.gov/sage/index.html, provides detailed information about SAGE guidelines, regulations, budget, and history.
Maier, Peter, Alex Molnar, Stephen Percy, Phillip Smith, John Zahorik, Greg Giglio, Sally Hochstein, Lisa Radtke, Laura Roskos, Mark Schill, and Kathy Shields,
First Year Results of the Student Achievement Guarantee in Education Program,
Milwaukee, Wisc.: Center for Urban Initiatives and Research, University of Wisconsin-Milwaukee, December 1997.
Molnar, Alex, Philip Smith, John Zahorik, Anke Halbach, Karen Ehrle, Lawrence M. Hoffman, and Beverly Cross, 2000-2001 Evaluation Results of the Student Achievement Guarantee in Education (SAGE) Program, Milwaukee, Wisc.: School of Education, University of Wisconsin-Milwaukee, December 2001.
Molnar, Alex, Philip Smith, John Zahorik, Karen Ehrle, Anke Halbach, and Barbara Kuehl, 1999-2000 Results of the Student Achievement Guarantee in Education (SAGE) Program Evaluation, Milwaukee, Wisc.: School of Education, University of Wisconsin-Milwaukee, December 2000.
Molnar, Alex, Philip Smith, John Zahorik, Lee Breese, Karen Ehrle, Anke Halbach, Amanda Palmer, Alan Silverman, and William Harvey, 1997-1998 Evaluation Results of the Student Achievement Guarantee in Education (SAGE) Program, Milwaukee, Wisc.: School of Education, University of Wisconsin-Milwaukee, December 1998.
Molnar, Alex, Philip Smith, John Zahorik, Lee Breese, Karen Ehrle, Anke Halbach, Amanda Palmer, and Betsy Schoeller, 1998-1999 Evaluation Results of the Student Achievement Guarantee in Education (SAGE) Program, Milwaukee, Wisc.: School of Education, University of Wisconsin-Milwaukee, December 1999.
Webb, Norman L., Robert H. Meyer, Adam Gamoran, and Jianbin Fu, Participation in the Student Achievement Guarantee in Education (SAGE) Program and Performance on State Assessments at Grade 3 and Grade 4 for Three Cohorts of Students-Grade 1 Students in 1996-97, 1997-98, and 1998-99, Milwaukee, Wisc.: Wisconsin Center for Education Research, February 9, 2004.