The Children At Risk program, which is administered by the Department of Public Instruction, was created in 1985 Wisconsin Act 29, the 1985-87 Biennial Budget Act, in an effort to reduce the number of students at risk of failing in or dropping out of school prior to receiving a high school diploma. Program services were first made available in the 1986-87 school year. Ten years later, the program's purpose remains the same, although its structure was changed significantly in 1993 Wisconsin Act 16. Available program funding has remained constant at $3.5 million in general purpose revenue (GPR) since fiscal year (FY) 1990-91.

The current program is delineated in s. 118.153, Wis. Stats. According to statutes:

• "children at risk" includes dropouts, habitual truants, parents, or adjudicated delinquents who are either in grades 5 through 12 and are two or more years behind their age group in basic skill levels or in grades 9 through 12 and are one or more years behind their age group in the number of high school credits attained; 

• each year, each school district must identify enrolled children at risk and develop a plan to meet their needs; 

• a school district must apply for aid under the program if in the previous school year it had 50 or more dropouts and a dropout rate exceeding 5 percent of its total high school enrollment; 

• a school district may apply for aid under the program if in the previous school year it had 40 or more dropouts; and 

• aid payments are based on the number of individual at-risk students who met designated performance criteria during the previous school year. 

The Legislative Audit Bureau is required biennially to audit school district eligibility, performance criteria, and state aid payments. We have completed four previous evaluations of the Children At Risk program. Our last evaluation, released in January 1995, found that since the program modifications enacted in 1993 Wisconsin Act 16 had been in effect for only one year, it was not possible to determine whether the program was having an effect on the dropout rate or the number of students graduating from high school. The program changes had, however, resulted in a significant reduction in the number of school districts eligible to receive program aid and the number of students identified as at-risk, as well as a decline in the number of these students who met performance criteria. 

Our current evaluation again reviews the available data and trends regarding school district eligibility for funding, pupil performance criteria, and state aid payments. We also reviewed dropout and graduation rate information that was available for the three complete school years since the changes to the program and analyzed whether there is a relationship between the trends and the Children At Risk program. We reviewed the services school district at-risk programs provide, as well as the extent to which school districts are contracting with private, nonprofit, nonsectarian agencies to provide programming for at-risk students. Finally, we developed options for the future of the Children At Risk program.

The Children At Risk program, as modified by 1993 Wisconsin Act 16, has been in effect for three full school years: 1993-94, 1994-95, and 1995-96. During this time period, it has become apparent that the requirements of the program as it exists today are not relevant to most school districts. The program can no longer be considered a statewide program designed to improve overall dropout and graduation rates for three reasons: 

• Only those school districts that applied for program aid were required to provide a formal program for at-risk students. Although all school districts are currently required to identify the children at risk enrolled and develop a plan describing how the district will meet their needs, previously all districts had to provide an at-risk program, whether or not they applied for and received funding under the program. 

• Only two school districts, Milwaukee Public Schools (MPS) and Racine Unified School District, were required to apply for aid in each year since the program was modified, and no more than 19 school districts have been required or eligible to apply for funding in any school year since 1993-94. 

• One school district, MPS, has received two-thirds of the $8.6 million distributed during the past three school years. For the 1995-96 school year, three districts-MPS, Racine, and Janesville-accounted for 87.8 percent of the funding received. The remainder, slightly more than $400,000, was distributed among the remaining 11 districts that applied for program aid. 

Because few school districts are affected by the Children At Risk program, it may not be reasonable to expect the program to have an effect on dropout and graduation rates throughout the state, as originally intended by the Legislature. However, some indication of the program's effectiveness may be reflected in the graduation and dropout rates of those districts that have received program funds in each of the three years since the program was modified. Of 427 school districts, 10 have received some funding in each of the three years. 

However, within those districts that have received funds for each school year since 1993-94, it is difficult to generalize as to whether the program has had any effect. First, the average graduation rate for all high school seniors in the ten districts has remained fairly constant, although the graduation rates of six of the ten districts increased. However, in only three districts-Appleton Area, Kenosha, and Oshkosh Area-did graduation rates consistently improve over the three-year period. In addition, the graduation rates in the three districts that receive the vast majority of available funding-MPS, Racine, and Janesville-declined. Second, although the average dropout rate for the ten districts has declined over the three years, there are no clear trends within the districts, and the dropout rates for all students in six of the ten districts increased. In only one district-MPS-did the dropout rate consistently decline over the three-year period. 

In addition, the manner in which the Department calculates graduation and dropout rates may result in data that are not indicative of the actual performance of students within the districts. The progress of individual students is not followed. Rather, the rates are based on total student enrollment, with the base number used in the calculations determined on the third Friday of September. No adjustment to the third Friday enrollment count is made if a student enters or leaves a school district during the year. Therefore, students who enter or leave a school district during the year are accounted for in only the final number of graduates or dropouts, but not in the original enrollment number. This can result in either an underestimation or an overestimation of the base used to determine rates, thereby skewing the results. 

Another measure of the program's effectiveness is the extent to which at-risk students who are enrolled in programs for which districts receive program funding meet the State's performance criteria. We found that in six of the ten districts, the percentage of at-risk students meeting performance criteria declined in both 1994-95 and 1995-96, compared to 1993-94. However, in two districts-MPS and Racine-the percentage of at-risk students meeting performance criteria improved in each of the three years reviewed. The improvement in MPS appears to be most directly attributable to a broader interpretation of student performance criteria than had been applied in the past the interpretation was supported by the Department and first fully employed by MPS for the 1994-95 school year. 

The graduation rate of at-risk seniors, which is one of the program's performance criteria, is another measure of the program's effectiveness. This rate increased from 49.2 percent in 1993-94 to 60.2 percent in 1995-96. The major reason for the significant increase in the average graduation rate is an increase in the rate within MPS between the 1993-94 and 1994-95 school years.

Overall, it is difficult to reach a definitive conclusion regarding the effectiveness of the Children At Risk program. On the one hand, it would appear that some of the districts that receive funding have shown improvements in either their graduation rates or their dropout rates. On the other hand, only one district-Appleton Area-showed an improvement in both its graduation rate and its dropout rate. At the same time, however, the percentage of Appleton Area's at-risk students that met the State's performance criteria under the Children At Risk program declined substantially.

Furthermore, while some may argue that MPS, which has the highest dropout rate among the ten districts, has seen a decline in this rate, it is questionable whether the decline can be attributed to the Children At Risk program. On the one hand, Children At Risk program legislation requires MPS to provide special programs to at-risk students and, therefore, the decline in the district's dropout rate may be linked to this program. On the other hand, other measures of student performance have shown varied results.

It may not be reasonable to expect a direct connection between student performance and the Children At Risk program, especially because of department actions and the limited amount of funding available. First, although the funding available through the program has been intended to serve as an incentive for districts to improve the retention and performance of their at-risk students, the Department has done little to reinforce this aspect of the program. Rather, the Department:

• worked with districts, in the first year following changes to the program, to maximize both the number of students identified as at-risk and the number of students who met the established performance criteria. The Department allowed districts that applied for aid under the program to apply a broad interpretation of how to measure the criteria, resulting in an increase in the funding provided rather than the increased retention of students.

• does not ensure the accuracy of the information provided to it by those districts applying for funds. As a result, although the Department had already paid school districts in March 1996 for the 1994-95 school year, and had prepared an estimate of payments to be made this spring for the 1995-96 school year based on these reports, the Department was unaware of several inaccuracies that a simple review of the reports would have noted.

We include a recommendation that the Department provide additional oversight of the Children At Risk program.

In addition, the limited funding available calls into question whether the Children At Risk program can be expected to have a significant effect on dropout and graduation rates. Children At Risk funding for school year 1995-96 represented at most 3.3 percent of the cost per pupil. The average per pupil program aid received was $167, which is an amount that cannot be expected to purchase a wide range of services for a district's at-risk students. It appears, therefore, that the program has become little more than another source of revenue for the participating districts.

The Children At Risk program has undergone several minor and one major change since it was established, in an effort to create a stronger program designed to address concerns about the State's dropout and graduation rates. Whether any additional major changes to the program would further promote its purposes is not clear, given the limited amount of funding involved. The Legislature could, however, maintain the program requirements as they currently exist while making changes to the funding formula. Options regarding funding include:

• establishing a competitive grant program through which school districts develop and implement innovative at-risk programs to replace the current program. Eligibility for grants could be determined based on dropout rates, as is currently the case, but program funding could be awarded for implementation of at-risk programming activities, teacher training, and funding at-risk program personnel.

• establishing a base level of funding for the three districts-MPS, Racine, and Kenosha-that are routinely required to apply for funds under the program. Because of high dropout rates, these school districts have the greatest need to develop programming in an attempt to prevent dropouts and encourage students to stay in school to attain a high school diploma.

• eliminating program funding. Other funding is available to serve "high risk" students, a group that includes at-risk students, including $143.3 million in federal funds and an additional $62.6 million in state funds in FY 1996-97.

If the Legislature maintained all other aspects of the program as they currently exist, those districts with the most difficult dropout problems would continue to be required to provide programming. In addition, they would also maintain the ability to contract with private agencies to provide at-risk educational services. Currently, however, MPS is the only school district that contracts with private agencies to provide educational services to its at-risk student population.

We found that MPS's reliance on these agencies, referred to as partnership schools, has increased over time. For example, from the 1986-87 to the 1996-97 school years:

• the number of contracts with partnership schools increased from 7 to 18;

• the number of seats within the partnership schools increased from 279 to 1,605; and

• payments to the partnership schools increased from $1 million to a projected $8.1 million for the 1996-97 school year.

In reviewing the performance of at-risk students within MPS over a two-year period, we found that a larger percentage of at-risk students enrolled in the partnership schools met the State's performance criteria under the Children At Risk program than do at-risk students enrolled in programs administered directly by the district. For example, 63.1 percent of at-risk students enrolled in the partnership schools during school year 1995-96 met three of the State's five performance criteria, compared to 45.2 percent of at-risk students enrolled in MPS-administered programs.

We also compared the educational progress made by students in both partnership schools and regular MPS programs over a two-year period to determine whether they graduated on time. We found that a higher percentage of students enrolled in partnership schools graduated than did students in MPS schools.

We identified several contract management concerns about the relationship between MPS and its partnership schools, many of which were also identified by an internal audit conducted by the district board in 1995. In addition, during three of the four years we reviewed, MPS contracted with a private school using Children At Risk contracting authority although the program under contract did not serve identified at-risk students. As a result, an estimated $633,033 was paid to the school under the program's auspices. To the extent that these funds were provided to the district by the State, they should be repaid to the State by the district; however, because MPS does not differentiate between funding sources in making such payments, it is not possible to determine whether state funds were used. Nevertheless, it is clear that MPS violated state statutes when it used its contracting authority under the Children At Risk program to provide services to other students.