Can schools solve problems they did not create? Washington State's "Schools for the 21st Century" Program.

Primary and secondary education reform efforts in the US over the past 40 years have often resulted in only partial successes at best. Major reform efforts, such as the National Defense Education Act of 1958 (NDEA), defined education system goals primarily in terms of Cold War strategy, largely ignoring community building and the promotion of social equity through education. Commissioned in response to both declining test scores and changing societal expectations from education, A Foundation Goes to School: The Ford Foundation Comprehensive School Improvement Program (1960-1970), (1) (CSIP), linked social welfare policy with educational policy. The CSIP concluded that dealing with issues affecting the environment within which schools operate (such as children living in poverty) would alleviate some of the demands placed on schools, and allow them to concentrate on issues directly related to student achievement.

The Schools for the 21st Century Program in the State of Washington tried to build on the conclusions of the CSIP study. Yet, our research on educational outcomes in Washington indicates that many important environmental influences on student performance are still not taken into account by education reformers. This might offer a partial explanation of why outcomes--in terms of standardized test performance--have not improved.

Reform Efforts in American Education

The foci of reform efforts in American education policy which arose in the early 19th century parallel many of the community-linked concerns that are currently being addressed in measures such as those in the Schools for the 21st Century Program. Horace Mann's "Common Schools" movement of the 1830s, for instance, insisted on education as the cornerstone of a strong democratic tradition, providing the individual with a sense of efficacy in the political as well as the private realm.

The late 19th and early 20th century witnessed efforts at educational reform driven by the changing demographic makeup of the population. In light of burgeoning immigrant populations, progressive reformers of the day placed a renewed emphasis on citizen participation in politics, requiring that these new denizens of America's cities develop a proper understanding and appreciation for "acceptable" methods of political participation. Education policy was cast as a strategic line of defense for American liberal traditions and society. The National Defense Education Act of 1958 (NDEA) can be interpreted as a continuation of this underlying policy goal, exported from domestic policy to the Cold War policy arena and redefined for that purpose.

The 1970s and 1980s witnessed a return to earlier approaches to education policy, with a renewed emphasis on individual benefits, but our conceptualization of individual and communal benefits had changed rather dramatically. The result has been two quite distinct approaches to school reform. The school choice approach stresses choices by and benefits for individuals, promoting market-oriented, competition-driven incentives for educators. Innovations include the privatization of education, vouchers, and open-enrollment magnet schools. While the school choice approach is grounded in the assumption that competition between schools will lead to innovation, improved student performance, and therefore general societal improvement, defenders of public schools reject this argument. They argue that school choice does not adequately address social equity concerns which, they believe, can be advanced by improving public schools through policies and innovations, such as Schools for the 21st Century.

While we share the objectives of those defending public schools, we question whether their approach, as manifested in Schools for the 21st Century, fully addresses the situation faced by today's schools. Rather than attempting to compile a list of the pressures and problems faced by school officials and teachers, we identify two issues that confront many school districts in the state: poverty, and significant numbers of at-risk students, commonly associated with limited English proficiency (LEP).

Schools for the 21st Century

In May of 1987 the Washington State Legislature created Schools for the 21st Century. The program reflected the following three core assumptions about the "restructuring" of the public school: reforms should be developed at the local level; reduction of State regulation is conducive to effective reform; effective reform requires time and collaboration on the part of administrators teachers, parents, and other concerned citizens. (2)

Local school districts developed a set of proposals for reform through a collaborative process involving teachers and administrators at the district level. To foster continual collaboration between administrators, teachers, community leaders, and parents, schools were allowed to pay teachers and administrators for an additional two-week period at the end of the school year to implement the program. State law authorized exemptions from regulations concerning number of days in the school year, student-teacher ratios, the "commingling" of funds and "salary lid compliance," to facilitate locally based reform efforts. The program also provided teachers with additional technology and allowed for distinct instructional methods to reflect the heterogeneous quality of student performance and the need for a more flexible classroom environment. Structural changes were also introduced, such as classrooms with students from two grade levels.

Seven school districts statewide (out of nearly 300) were chosen on the basis of geographical diversity and demonstrated need. The average median family income for the seven districts was $26.3 thousand compared to the state mean of $27.7 thousand, with five of the seven districts below the state average. In terms of parental education, often employed in explanations of student achievement, four of the seven school districts were below the state average. The mean poverty rate for school-children in the seven districts was 57 percent, with four of the seven districts above the state mean of 48.2 percent. In terms of district resources, the mean for the seven districts was $4.1 thousand per pupil; six of the seven districts were below the state mean of $4.6 thousand. The mean student-teacher ratio for the seven districts was 19 students per instructor, with five of the seven districts above the state mean of 18.7 students per instructor; one of the pilot program districts had a student-teacher ratio of 22:1.

Pilot Project Results: The State's Findings

A 1993 interim report by the Washington State Board of Education (WSBE) to the Washington State Legislature documented results of the program. A large part consisted of self-evaluation by the participating schools and school districts. Each assessed self-identified goals as achieved, partially achieved, not achieved, or "no longer a goal."

The goals identified in this process fall into three general categories. First, administrative changes--such as greater teacher empowerment, merit pay, staff training, and staff co-ordination efforts. Second, educational innovations--to create outcome-based education, introduce more technology into the classroom, and instill a spirit of "lifelong learning" into students. Third, changes in governance--involving efforts to collaborate with parents, community members, and university faculty to develop appropriate educational programs. In all three cases, there emerged no generally accepted empirical method for measuring success; and no basis for establishing a causal relationship between these efforts and increased student achievement.

Three general obstacles to reform were outlined in the report. First, a consensus-based approach to education policy making requires a great deal of time. Second, there was resistance to policy changes from both teachers and community members. Third, it proved difficult to develop consensus among the various stakeholders (school and district administrators, teachers, parents, interested members of the community).

The magnitude of these obstacles was left rather vague in the interim report, which allowed for a positive evaluation of the program to date. This conclusion seems to beg at least some of the important questions. In this essay we try to pose these questions by addressing indicators the WSBE interim report did not consider, namely, the change in aggregate student performance on standardized examinations in the schools participating in the three-year program.

Results of the Pilot Project: Our Findings

We studied aggregate district-level outcomes on standardized tests for the school districts within the program compared with those not selected. The basis for comparison was the nationally standardized Iowa Test of Basic Skills (ITBS), administered annually in most districts in Washington State. It consists of four testing areas: mathematics, reading, verbal, and battery (resource skills). A comparison of the change in mean score

[Part 1 of 2]

IOWA TEST OF BASIC SKILLS: AVERAGE SCORES FOR 4th AND 8th GRADES IN SCHOOL
DISTRICTS IN THE SCHOOLS FOR THE 21ST CENTURY PROGRAM: 1989-1990

    4th GRADE        BATTERY      MATHS      READING      VERBAL
 DISTRICT, COUNTY   1989  1990  1989  1990  1989  1990  1989  1990


CAMAS, CLARK          61    67    58    63    58    59    62    64
MOSES LAKE, GRANT     50    47    49    45    46    42    50    47
SEATTLE, KING         53    53    52    52    48    48    49    51
NORTH MASON, MASON    53    56    56    63    48    53    46    49
COLTON, WHITMAN       36    60    49    66    32    55    40    39
YAKIMA, YAKIMA        48    37    45    38    48    37    45    38
GRANGER, YAKIMA       28    22    37    38    21    20    27    21
STATE MEAN          52.8  53.4  51.3  51.6  51.9  52.0  48.6  49.4
S.E.                0.73  0.71  0.58  0.57  0.76  0.73  0.65  0.59

[Part 2 of 2]

IOWA TEST OF BASIC SKILLS: AVERAGE SCORES FOR 4th AND 8th GRADES IN SCHOOL
DISTRICTS IN THE SCHOOLS FOR THE 21ST CENTURY PROGRAM: 1989-1990

    4th GRADE       COMPOSITE
 DISTRICT, COUNTY   1989  1990


CAMAS, CLARK        59.8  63.3
MOSES LAKE, GRANT   48.8  45.3
SEATTLE, KING       50.5  51.0
NORTH MASON, MASON  50.8  55.3
COLTON, WHITMAN     39.3  55.0
YAKIMA, YAKIMA      46.5  37.5
GRANGER, YAKIMA     28.3  25.3
STATE MEAN          51.2  51.6
S.E.                 0.6   0.6

[Part 1 of 2]

    8th GRADE        BATTERY      MATHS      READING      VERBAL
 DISTRICT, COUNTY   1989  1990  1989  1990  1989  1990  1989  1990


CAMAS, CLARK          55    45    55    43    55    46    52    46
MOSES LAKE, GRANT     58    58    58    58    53    52    49    53
SEATTLE, KING         53    50    52    50    50    49    50    47
NORTH MASON, MASON    59    54    49    47    62    56    59    47
COLTON, WHITMAN       60    42    60    46    63    41    58    42
YAKIMA, YAKIMA        49    44    45    37    51    50    46    43
GRANGER, YAKIMA       29    24    29    28    24    19    43    32
STATE MEAN          55.7  55.0  53.9  54.0  57.9  56.7  51.9  50.5
S.E.                0.70  0.75  0.69  0.73  0.71  0.78  0.62  0.66

[Part 2 of 2]

    8th GRADE       COMPOSITE
 DISTRICT, COUNTY   1989  1990


CAMAS, CLARK        54.2  45.0
MOSES LAKE, GRANT   54.5  55.3
SEATTLE, KING       51.3  49.0
NORTH MASON, MASON  57.3  51.0
COLTON, WHITMAN     60.3  42.8
YAKIMA, YAKIMA      47.8  43.5
GRANGER, YAKIMA     31.3  27.8
STATE MEAN          54.8  54.0
S.E.                 0.7   0.7

performance on these standardized tests for the districts within and outside of the reform program yielded mixed results, at best.

All seven participating districts administered the ITBS exams in 1989 and 1990, the second and third years of the program. At the fourth grade level, of the districts participating in the program, the proportion testing at or above the state-wide average increased slightly in three of the four testing areas. In terms of composite fourth grade test performance, four of the seven districts in the program had a greater improvement than districts not in the program.

At the eighth grade level, the program did not appear to have the intended effect. At this level, aggregate district-level performance fell in all testing areas during the period under study. Only one of the seven participating districts showed improved composite eighth grade performance. The other six declined on average seven percentile points, while non-participating districts declined on average less than onehalf percentile.

These results suggest that aspects of education reform are not being sufficiently considered. First, we would expect that the reform initiative should lead to higher performance at both the fourth and eighth grades, or at the very least yield results that keep pace with school districts not involved in the program. This has not occurred. Overall, many of the 21st Century school districts were doing less well in 1990 than they had done in 1989.

Implicit in the education literature is the assumption that reform policies are equally applicable across grade levels, that the same pedagogical and structural variables play an equal role across grade levels with respect to educational outcomes. The assumption was present in the Washington pilot program; it is called into question by the ITBS results.

An analysis of these differing results at the fourth and eighth grade levels leads to some revealing findings. At the fourth grade level, external characteristics (e.g. parental education, percentage of school district children living in poverty, and percentage of children classified as "at-risk") are most important in predicting test scores. At the eighth grade level, while external social characteristics remain important, factors internal to the school organization (such as student-teacher ratios and dollars allocated per teacher) play a much larger role. We would suggest that fourth grade students are generally more dependent on parental involvement and hence external factors likely play a prominent role. By the time a student has entered the eighth grade, however, we would suggest that he or she is far less dependent on parental socialization; and that factors within the control of schools and districts become increasingly important. These differences should make us question the assumption that uniform application of education reforms across grade levels will yield similar results, or, indeed, that any single quick fix of the schools could possibly address the underlying problems.

Concluding Remarks

We do not propose that the answer to education performance lies in adopting the school choice approach, which has recently moved to the forefront of the education policy debate. Without entering that debate here, we feel safe in concluding that its resolution will not remove public school reform from its central place in American education policy debate. Proposals for schools reform will need to give hope of real improved student performance at a cost acceptable to an electorate increasingly hostile to inchoate policy initiatives imposing higher taxes yet often yielding few tangible benefits.

Public schools are increasingly expected to deal with issues of social equity as well as providing for individual student achievement. Schools for the 21st Century represents an attempt to deal with this dual concern. Paradoxically, in its attempt to incorporate past lessons in education policy innovation, the program may fail to recognize the possible limitations to education reform.

If the attempt to integrate social welfare policy with education policy results in the stagnation or decline of student achievement, then other approaches may be needed--even if the Ford Foundation study is correct in concluding that socio-economic disadvantage is related to student achievement. One aspect of such an approach signalled by our results is the fact that the effects of socio-economic factors on student achievement are not uniform across grade levels.

Even more fundamentally, however, it remains unclear to what extent school-related policy can achieve improved student outcomes--independent of a concomitant wider effort on the part of welfare policy makers. The Washington State approach recognizes the need for such a linked, cross-policy effort, but when applied sought to overcome only a limited number of obstacles, essentially by bringing a wide range of participants into education policy discussion.

Washington State's Schools for the 21st Century program's accomplishments fall short of its self-identified objectives. In this instance, it is not clear that the Schools for the 21st Century program simultaneously dealt with social equity issues and the improvement of student performance. We are certainly not suggesting that social equity should be ignored, but that it will have to be approached in a manner that is complimentary to the educational efforts of public educators and their pupils.

(1) . Meade, E.J., Jr., Ford Foundation, 1972.

(2) . Anderson, K. and J.A. Billings, 1993 Report to the Legislature on Schools for the 21st Century Program, Washington State Board of Education, 1993