This is a review under sec. (Rule) 809.62, Stats., of a decision of the court of appeals, Davis v. Grover, 159 Wis. 2d 150, 464 N.W.2d 220 (Ct. App. 1990). The court of appeals reversed the decision of the Dane county circuit court, Judge Susan R. Steingass, and found that the Milwaukee Parental Choice Program (MPCP) violated art. IV, sec. 18 of the Wisconsin Constitution.1 The MPCP is a publicly funded program that permits selected children from low-income families to attend nonsectarian private schools at no cost to the student.
*512The scope of our inquiry is strictly confined to the specific issues raised on this review. We pass no judgment on the wisdom or desirability of the MPCP. The propriety of the program is most appropriately addressed by the legislature, not the judiciary.
Three issues are raised in this review. The first issue concerns whether the MPCP is a private or local bill which was enacted in violation of the procedural requirements mandated by art. IV, sec. 18 of the Wisconsin Constitution. We hold that the MPCP is not a private or local bill and, thus, is not subject to the procedural requirements of Wis. Const, art. IV, sec. 18.
The program was and remains politically controversial. As such, it was greatly debated in legislative committee public hearings and by the entire legislature. It is evident the program was not smuggled through the legislature. The purpose of this experimental legislation is to determine if it is possible to improve, through parental choice, the quality of education in Wisconsin for children of low-income families.2 Logically, the best location *513to test the program is in a city such as Milwaukee where the socio-economic disparities and educational problems are particularly great and the potential private educational choices are most abundant.
The second issue concerns whether the MPCP violates art. X, sec. 3 of the Wisconsin Constitution, which requires the establishment of uniform school districts. We hold that the MPCP does not violate art. X, sec. 3 of the Wisconsin Constitution because the participating private schools do not constitute "district schools," even though they receive some public monies to educate students participating in the program.
The third issue concerns whether the MPCP violates the public purpose doctrine which requires that public funds be spent only for public purposes. We hold that the MPCP does not violate the public purpose doctrine. We give great weight to legislative determinations of public policy. Sufficient safeguards are included in the program to ensure that participating private schools are under adequate governmental supervision reasonably necessary under the circumstances to attain the public purpose of improving educational quality. Further, the cost of education and the funds available for education are dependent upon the taxpayers' ability to fund an intensive public educational program. The amount of money allocated under this program to participating private schools for the education of a participating student is less than 40 percent of the full cost of educating that same student in the Milwaukee Public School (MPS) system. The total amount of public funds appropriated to fund this experimental program is inconsequential when compared to the total expenditures for public edu*514cation allocated to schools throughout the state of Wisconsin.
The relevant facts follow. The MPCP, as enacted into law, provides that a kindergarten through twelfth grade (K-12) student who resides in a city of the first class may attend, at no charge to the student, any nonsectarian private school located in the city if the following criteria are met:
(1) the family income does not exceed 175% of the poverty level;
(2) the pupil was enrolled in a public school in the city, was attending a private school under this program, or was not enrolled in school the previous year;
(3) the private school notifies the State Superintendent of its intent to participate in the program by June 30 of the previous school year;
(4) the private school complies with 42 U.S.C. sec. 2000d;3 and
(5) the private school meets all health and safety laws or codes that apply to public schools.
Section 119.23(2)(a), Stats. Additionally, private schools participating in the program must meet defined performance criteria4 and submit to financial and performance *515audits by the state.5 For each participating student, approximately $2,500 in state educational funding is diverted from the Milwaukee Public Schools (MPS) to the participating private school.
The legislature placed significant limitations on the scope of the program. The program limits the number of students that may participate in the program to no more than 1 percent of the school district's membership. Section 119.23(2)(b)l, Stats. This limitation makes the program available to approximately 1,000 Milwaukee students. The record reflects that participating students are selected on a random basis with preference afforded to students continuing in the program and their siblings. This narrowly defined and carefully monitored program provides that no private school may enroll more than 49 percent of its total enrollment under this program. Section 119.23(2)(b)2.
Since the goal of the MPCP legislation is to gather information to assist in identifying educational problems and solutions, a number of reporting and supervisory functions on the part of the State Superintendent as well as the Legislative Audit Bureau are statutorily required by the program. The State Superintendent must submit a report to each house of the legislature concerning achievement, attendance, discipline, and parental *516involvement under the program as compared to the public school system in general. Section 119.23(5)(d), Stats.
The State Superintendent is required to monitor the performance of students participating in the program and is given specific authority to prohibit participation in the program the following school year by any private school which does not meet the performance criteria. Section 119.23 (7) (b), Stats.
The State Superintendent is also authorized to conduct one or more financial or performance evaluation audits of the program. Section 119.23(9)(a), Stats. The Legislative Audit Bureau is further required to perform a financial and performance evaluation audit on the program. Section 119.23(9)(b). Clearly, the legislature included very particular and detailed reporting and supervisory requirements to test a new and innovative method of delivering education services to students of low-income families.
Governor Tommy Thompson first proposed a parental choice program in early 1988. The proposal was analyzed by the Legislative Fiscal Bureau, but was never considered by the legislature. In 1989, the governor again proposed a parental choice program, at which time the Legislature requested the Legislative Council to study the proposal.
In October 1989, the bill that led to the enactment of the Milwaukee Parental Choice Program was introduced by a bipartisan coalition of 47 members of the assembly and nine senate co-sponsors. The bill was referred to the Assembly Committee on Urban Education, which held a public hearing on the proposal. A broad array of persons and organizations, encompassing many of the interests represented in this case, appeared at the public hearing. Based on committee reports and the statements made at the public hearing, the commit*517tee recommended an amended version of the bill to the assembly. After considering a number of amendments to the bill, the assembly passed the bill.
The program, as passed by the assembly, was then considered by the senate and referred to the Committee on Educational Financing, Higher Education and Tourism. Subsequently, it was added to the senate budget adjustment bill, a multi-subject bill addressing numerous unrelated topics. The language of this component of the bill was preceded by the title, "Milwaukee Parental Choice Program." Following the addition of a fiscal amendment relating to the program, the entire budget bill was adopted by the senate. The assembly passed the budget bill without again considering the parental choice program.
The governor signed the bill, but vetoed a sunset provision included in the program which would have limited the effective period of the program to a five-year time span. Thereafter, the MPCP was enacted into law under ch. 119, Stats., the chapter applicable to the school system in cities of the first class.
Lonzetta Davis, et al. (Davis), representing families of participating students and private schools participating in the program, initiated this action challenging a number of regulatory actions taken by State Superintendent of Public Instruction Herbert Grover (Superintendent Grover).6 Davis believed Superintendent Grover's actions were designed to frustrate the MPCP and exceeded his authority as State Superintendent.
Felmers O. Chaney, et al. (Chaney), representing various school administration organizations and the *518National Association for the Advancement of Colored People, intervened, challenging the MPCP on state constitutional grounds; namely, that it violates Wis. Const, art. IV, sec. 18 (private/local legislation clause), Wis. Const, art. X, sec. 3 (uniform district schools clause), and the public purpose doctrine.
The State of Wisconsin, acting on its own behalf, argues that the MPCP is constitutional in all respects.
The circuit court found the MPCP constitutional and that Superintendent Grover's actions exceeded his regulatory authority. Chaney filed an appeal on the constitutional issues with the court of appeals. Superintendent Grover did not appeal the circuit court's decision on the regulatory issues.
The court of appeals reversed the decision of the circuit court and held that the MPCP violated the private/local legislation clause of Wis. Const, art. IV, sec. 18. It did not reach the uniformity clause and public purpose doctrine issues.
No injunction was ever issued against the Milwaukee Parental Choice Program, which continues to operate unaffected by the pending litigation.
The issues presented in this case involve questions of law. On review, this court decides questions of law independently without deference to the decisions of the trial court and court of appeals. Ball v. District No. 4, Area Bd., 117 Wis. 2d 529, 537, 345 N.W.2d 389 (1984). We now address each of these issues separately.
I. THE PRIVATE/LOCAL LEGISLATION CLAUSE
Article IV, sec. 18 of the Wisconsin Constitution states:
*519No private or local bill which may be passed by the legislature shall embrace more than one subject, and that shall be expressed in the title.
It was adopted as part of the original Wisconsin Constitution of 1848 and has remained unchanged. In previous cases, we have explained that art. IV, sec. 18 has three underlying purposes:
1) [T]o encourage the legislature to devote its time to the state at large, its primary responsibility; 2) to avoid the specter of favoritism and discrimination, a potential which is inherent in laws of limited applicability; and 3) to alert the public through its elected representatives to the real nature and subject matter of legislation under consideration.
Milwaukee Brewers v. Department of Health & Social Services, 130 Wis. 2d 79, 107-08, 387 N.W.2d 254 (1986). The requirements of art. IV, sec. 18 are prescribed to ensure accountability of the legislature to the public and to "guard against the danger of legislation, affecting private or local interests, being smuggled through the legislature." Milwaukee County v. Isenring, 109 Wis. 9, 23, 85 N.W. 131 (1901). In Brookfield v. Milwaukee Sewerage, 144 Wis. 2d 896, 426 N.W.2d 591 (1988), we further examined legislative accountability. Section 18 also recognizes the need to avoid "internal logrolling"7 on the part of the legislature. Multi-subject bills by their nature are subject to a greater susceptibility of smuggling and logrolling. They intermingle a variety *520of unrelated legislation which singly may not have the support of the majority and, thus, tend to reduce accountability to the public. Nevertheless, the fact that a multi-subject bill contains a program such as the MPCP does not necessarily condemn the process in which the program was enacted as unconstitutional.
The determination of whether a bill violates Wis. Const, art. IV, sec. 18 involves a two-fold analysis. We must first address whether the process in which the bill was enacted deserves a presumption of constitutionality. Second, we must address whether the bill is private or local. If the bill is found to be private or local, then the requirements of art. IV, sec. 18 apply; namely, that the legislation must be a single subject bill and the title of the bill must clearly reflect the subject.
The general rule in Wisconsin is that a statute is presumed to be constitutional and "the burden of establishing the unconstitutionality of a statute is on the person attacking it, who must overcome the strong presumption in favor of its validity." ABC Auto Sales v. Marcus, 255 Wis. 325, 330, 38 N.W.2d 708 (1949). This presumption of constitutionality was recognized in the art. IV, sec. 18 context in Soo Line R. Co. v. Transportation Dep't, 101 Wis. 2d 64, 76, 303 N.W.2d 626 (1981). However, we explained in Brookfield v. Milwaukee Sewerage, 144 Wis. 2d 896, 912-13, 426 N.W.2d 591 (1988), that a distinction exists between assessing the constitutionality of the substance of legislation and assessing the constitutionality of the process in which the legislation was enacted. In Brookfield, we stated:
In the sec. 18 context, the point of the rules listed in the text is to determine whether some sham or artifice is being perpetrated by smuggling through a local *521bill in the sheep's clothing of a statewide interest or a general bill.. . .
By contrast to sec. 18, under equal protection the legislature is not being accused of violating a constitutionally mandated procedural rule. Therefore, because the legislature is now presumed to have "intelligently participate[d] in considering such bill . . .." (Isenring, 109 Wis. at 23) this court is not seeking to determine whether a sham has been perpetrated. Consequently, this court has repeatedly stated that a law attacked on equal protection grounds is entitled to a presumption of constitutionality, see, e.g., Laufenberg v. Cosmetology Examining Board, 87 Wis. 2d 175, 181, 274 N.W.2d 618 (1979), which presumption attends the use of the rational basis test.
Thus, although both sec. 18 and equal protection seek to determine whether one group is being accorded favored status, the difference between the sec. 18 and the equal protection contexts is this: In sec. 18 cases, because the legislature is alleged to have violated a law of constitutional stature which mandates the form in which bills must pass, the court will not indulge in a presumption of constitutionality, for to do so would make a mockery of the procedural constitutional requirement. . ..
By contrast, in equal protection, as stated above, the court will presume constitutionality . . . given the quite different purposes of sec. 18 and equal protection.
Brookfield, 144 Wis. 2d at 918-19 n.6. In Brookfield, there was no indication that the legislature had adequately considered or discussed the legislation in question that was passed as part of the budget bill. The record in the present case is replete with evidence that the MPCP was introduced by a significant number of *522legislators and was debated extensively by the legislature and its various committees and agencies. The program was proposed in several consecutive years. The Assembly Committee on Urban Education held a public hearing on the proposed program. The program was passed as a separate, single subject bill by the assembly. Unfortunately, the senate included it as part of the multi-subject budget bill, thereby creating the problem we address here.
We are aware that time constraints sometimes force legislators to pass a variety of worthy legislation in one multi-subject package. However, multi-subject bills reduce accountability to the public and are very susceptible to the charge of violating the procedural requirements of Wis. Const, art. IV, sec. 18. The legislature could avoid litigatory challenges of this nature by using separate, single subject bills for legislation that is not plainly of statewide concern.
However, we find no evidence in this case that suggests the program was smuggled or logrolled through the legislature without the benefit of deliberate legislative consideration.8 As mentioned earlier, the MPCP legisla*523tion was passed by the assembly as a single subject bill. Even though the senate included the MPCP as part of the budget bill, the budget bill was debated by the senate and the senate specifically amended the MPCP prior to enactment of the budget bill. Clearly, the legislature "intelligently participate^] in considering" this program. Id. Therefore, under the circumstances of this case, it is proper for us to apply a presumption of constitutionality to the process in which the MPCP was enacted into law.9 Applying a presumption of constitutionality in this case was expressly authorized by the Brookfield court where we stated:
*524[U]nder sec. 18, full scrutiny of the legislature, rather than the substituted process of smuggling through is the best determinant of need.
Just as we seek not to err on the one hand by employing an inappropriate standard of deference through presuming constitutionality where such a presumption would render sec. 18 meaningless, so equally we seek not to err on the other hand by substituting our judgment for that of an attentive legislature. . . .
If such legislation is passed after full consideration . . . that will be the proper time to engage in the presumption of constitutionality. . ..
Brookfield, 144 Wis. 2d at 918-19 n.6 (emphasis added). The burden of overcoming this presumption of constitutionality falls upon Chaney, et al., the parties attacking the statute.
Even though we conclude that there is no indication that the MPCP was smuggled or logrolled through the legislature without due consideration and we apply a presumption of constitutionality to such process, our analysis does not end here. Article IV, sec. 18 specifies certain procedural requirements that must be satisfied if legislation is found to be private or local. The previous discussion concerning legislative consideration is only relevant to the presumption of constitutionality portion of the analysis. It has no effect on our determination of whether the MPCP is a private or local bill. We now turn to the determination of whether the MPCP is private or local legislation.
This court has developed three prongs of analysis for cases involving a challenge to legislation as being private or local. The first prong of analysis involves legislation that is specific on its face as to particular people, places or things that allegedly runs afoul of art. IV, sec. 18. See Milwaukee County v. Isenring, 109 Wis. 9, 85 *525N.W. 131 (1901); Monka v. State Conservation Comm., 202 Wis. 39, 231 N.W. 273 (1930); Soo Line R. Co. v. Transportation Dep't, 101 Wis. 2d 64, 303 N.W.2d 626 (1981); and Milwaukee Brewers v. DH&SS, 130 Wis. 2d 79, 387 N.W.2d 254 (1986). These cases explain that "such legislation is private or local within the meaning of sec. 18 and therefore prohibited unless the general subject matter of the provision relates to a state responsibility of statewide dimension and its enactment will have a direct and immediate effect on a specific statewide concern or interest." Brookfield, 144 Wis. 2d at 911.
The second prong of analysis involves legislation that is not specific on its face, but which involves classifications and allegedly runs afoul of the specific prohibitions of art. IV, sec. 31, which was adopted as an aid in a sec. 18 analysis. Section 31 explains specific areas in which the legislature is prohibited from enacting any special or private laws. The resolution of these cases depends on whether the legislation "falls into the category of matters upon which the legislature is competent to legislate pursuant to sec. 32 notwithstanding the prohibition of sec. 31." Id.
The third, and final, prong of analysis involves legislation that is not specific on its face, involves classifications, does not violate the provisions of sec. 31, but allegedly runs afoul of sec. 18. See Brookfield v. Milwaukee Sewerage, 144 Wis. 2d 896, 426 N.W.2d 591 (1988). A statute creating a closed classification can be the same as legislation that is specific on its face to a certain locality. In Brookfield, we determined that such cases must be analyzed consistent with the classification concepts developed in cases under art. IV, secs. 31 and 32. Id. at 912.
*526Five primary elements comprise the Brookfield test. These elements are as follows:
First, the classification employed by the legislature must be based on substantial distinctions which make one class really different from another.
Second, the classification adopted must be germane to the purpose of the law.
Third, the classification must not be based on existing circumstances only. Instead, the classification must be subject to being open, such that other cities could join the class.
Fourth, when a law applies to a class, it must apply equally to all members of the class.
. . .[F]ifth, the characteristics of each class should be so far different from those of the other classes so as to reasonably suggest at least the propriety, having regard to the public good, of substantially different legislation.
Brookfield, 144 Wis. 2d at 907-09. While these tests are similar to those used in the equal protection context, they are necessarily differently applied because sec. 18 and equal protection address quite different concerns.
The rationale for using the five-factor test was aptly explained in Brookfield, 144 Wis. 2d at 912-14 n.5. We shall not endeavor a reexplanation of that rationale here. We shall state only that sec. 18 addresses the form in which the legislation is enacted and not the substance of the legislation. In the classification legislation context, it is necessary to use the five-factor test to determine exactly what the substance of the legislation is in order to determine whether the procedural requirements of Wis. Const, art. IV, sec. 18 apply. Thus, although the *527five-factor test "is used in both a sec. 18 context and an equal protection context, the tests are necessarily differently applied, given the quite different purposes of sec. 18 and equal protection." Id. at 913 n.5.
Notwithstanding the fact that the title of sec. 119.23, Stats., expressly mentions Milwaukee, the text of the MPCP as well as its placement in the statutes suggests that it involves a classification and should be analyzed under Brookfield rather than Milwaukee Brewers. The MPCP applies to any school district in a city of the first class. It is not limited to Milwaukee because Madison presently meets the population requirement and could become a city of the first class by a simple declaration. While the title of legislation expressly refers to Milwaukee, titles of statutes are not part of the statute itself.10 We find no reason why this rule should not encompass legislative bills as well. Therefore, the MPCP is similar to the statute in Brookfield in that it involves a classification and not expressly a specific person, place or thing. Thus, we are required to apply the Brookfield five-factor test to determine whether the MPCP is private or local legislation.11
The first element of the Brookfield test requires that "the classification employed by the legislature must be *528based on substantial distinctions which make one class really different from another." The MPCP does not create a new classification, but involves a classification that has consistently been recognized and accepted by this court: namely, cities of the first class. "Cities of the first class” is defined under sec. 62.05, Stats., as cities with a population of 150,000 or more. Presently, Milwaukee is the only city to declare itself a city of the first class in the state of Wisconsin.
In Brookfield, we acknowledged that the mere size of a particular city does not necessarily justify treating that city differently than any other city in the state. Brookfield, 144 Wis. 2d at 916. However, cities of the first class, by virtue of their large population and concentration of poverty, are substantially distinct from other cities. In Camasco Realty Co. v. Milwaukee, 242 Wis. 357, 377, 8 N.W.2d 372 (1943), where the challenged law pertained to cities of the first class, we noted that "the requirements of a metropolitan city like Milwaukee as against the smaller municipal corporations of the state are so obvious that any other result would be opposed to the public welfare." In State ex rel. Nyberg v. Bd. of School Directors of the City of Milwaukee, 190 Wis. 570, 577, 209 N.W. 683 (1926), this court upheld a statute regarding first class city school districts and stated that "there is a substantial basis for classifying for school purposes the large communities embraced in cities of the first class as established under our law and the smaller communities of the state."
School districts located in areas with monumentally oppressive poverty problems as found in first class cities have particular educational problems as well. These problems were recognized also in Kukor v. Grover, 148 Wis. 2d 469, 482-83, 436 N.W.2d 568 (1989). As demon*529strated by dropout rates, welfare statistics, and population data, the Milwaukee Public School District has significantly greater education and poverty problems than any other school district in the state.
Various statistical analyses, while not entirely consistent, dramatically show the need for legislative attention. The dropout rate for the Milwaukee Public Schools is higher than any other area in the state. For example, in the 1988-89 school year, the dropout rate for students in grades 9-12 in the MPS reached 14.4 percent.12 In contrast, the public school dropout rate for the state at large during the 1988-89 school year was 3.11 percent, with no county, other than Milwaukee County, having a dropout rate of greater than 4.3 percent.13
During the 1988-89 fiscal year, Wisconsin spent $2.4 billion, or $499.57 per capita, on public welfare. Wisconsin ranked sixth among all states for welfare-related expenditures.14 In 1988, over 50 percent of the general public assistance in Wisconsin was spent in Milwaukee County alone and the city of Milwaukee comprises about two-thirds of the population of Milwaukee County. Furthermore, of the $485 million spent in Wisconsin in 1988 for Aid to Families with Dependent Children, $213 million was allocated to Milwaukee County.15
The statistical data clearly illustrates that the socioeconomic disparities and the educational problems are *530greater in the large urban area of Milwaukee than any other part of Wisconsin. By definition, first class cities encompass large urban cities in Wisconsin, such as the city of Milwaukee. Therefore, we find that the classification of first class cities is based on substantial distinctions which make the class really different from all others. The first element of the Brookfield test is satisfied.
The second element of the Brookfield test requires that "the classification adopted must be germane to the purpose of the law." Both the trial court and the court of appeals concluded that the only reasonable inference to be drawn from the MPCP was that it was an experiment intended to address a perceived problem of inadequate educational opportunities for disadvantaged children. Davis, 159 Wis. 2d 164-65. We agree with this conclusion.
Improving the quality of education in Wisconsin is, without a doubt, a matter of statewide importance. It is apparent that on a national scale the educational needs of many students are not being met by the present educational structure and options. Average School Aptitude Test (SAT) scores fell from 978 in 1960 to just 870 in 1980.16 Nearly 25 percent of public high school students drop out before graduation and the dropout rates for minorities often reach 50 percent. These are some of the highest dropout rates in the western world.17
The educational problems that the nation is experiencing are also evident in the Milwaukee Public Schools, where 55-60 percent of MPS students do not graduate from high school or do not graduate in a six-year period of time. A recent report by the Greater Mil*531waukee Education Trust states that only 40-45 percent of the students who start high school in the MPS graduate in four, five or six years. This completion rate is down from 57 percent in 1984. Of those who do graduate from high school, 36 percent graduate with a "D" average.18 Students of MPS, in general, score below the national average on the basic skills tests, and minority students score dramatically below the average. The grade point average (GPA) on a scale of 4.0 for MPS students in general is 1.60, whereas the GPA for African-American students in the MPS is just 1.31.19
The consequences of school dropouts and inadequate education are shocking. High school dropouts comprise 75 percent of the prison population and 80 percent of the families receiving Aid for Families with Dependent Children. Only 55 percent of the male dropouts under age thirty have jobs and only 20 percent have full-time jobs.20
Recently, researchers have attempted to discover the reasons underlying inadequate public instruction. A Brookings Institution study examined data from more than 60,000 students in 1,000 public and private schools to test the relationship between 220 different variables. The study concluded that the three most important factors that affected student achievement were student ability, school organization, and family background. Chubb & Moe, Politics, Markets & America's Schools 140 (1990). The factor which is most amenable to legislative efforts appears to be school organization. In this *532respect, the researchers found that "by itself, autonomy from bureaucracy is capable of making the difference between effective and ineffective organizations — organizations that would differ by a year in their contributions to student achievement."21 Id. at 181. We find especially interesting the study's conclusion that the educational credentials of teachers, teachers' scores on competency tests, how teachers are paid and other formal qualities do not make a significant difference on student achievement. Id. at 186.
In response to the conclusions reached by the Brookings Institution study and others, the MPCP was drafted to include two main features to help fulfill the *533statewide purpose of improving education. The first feature empowers selected low-income parents to choose the educational opportunities that they deem best for their children. Concerned parents have the greatest incentive to see that their children receive the best education possible. Parental choice allows parents to send their children to nonsectarian private schools which, except for the statutory responsibilities of the State Superintendent, are autonomously operated free from the bureaucracy of the public school system. In so providing, the program will engender educational success competition between the public and private educational sectors for students of low-income families.
However, the program is not an abandonment of the public school system. Rather, the MPCP would affect at most only 1 percent of the students in the MPS, giving the program a very small window of opportunity to test the effectiveness of an alternative to the MPS.
Furthermore, the MPCP contains a second feature which not only should benefit the MPS but also the state at large. The second main feature of the MPCP creates an extensive data compilation and reporting process which the state can use to measure the effects of choice and competition in education. The experimental nature of the program is evident from these detailed compilation and reporting requirements.
The experimental nature of the program can also be inferred from the fact that the program, as originally drafted, would have been effective for only a five-year period of time. However, in a partial veto, the governor removed the five-year time limit. It is unclear whether the governor felt that the time limitation was too short or too long. It is apparent, though, that the governor and the legislature directed the gathering of extensive infor*534mation for the purpose of reacting to this experimental program.
The success of the program is dependent upon the participation of numerous and diverse nonsectarian private schools such that the fate of the program does not rest on the operations of one or a few schools. The record indicates that at least nine private schools in Milwaukee filed an intent to participate in the MPCP when it was first implemented. We assume no other city in Wisconsin offers as many private schools as Milwaukee. The significant availability of private schools is so necessary to a reliable sampling of alternative educational methods that it distinguishes a first class city such as Milwaukee from all other communities.22 This experiment tests a theory of education. The possible failure in one or more private schools may be the fault of the school rather than the program's concept. Therefore, locating the program in a first class city such as Milwaukee where numerous and diverse private schools exist will enable the legislature to determine which, if any, of the private schools *535were most effective and why they are particularly successful in their mission of education.
We conclude that the classification of first class cities is germane to the purpose of the law. Clearly, improving the quality of education and educational opportunities in Wisconsin is a matter of statewide importance. The best location to experiment with legislation aimed at improving the quality of education is in a first class city, a large urban area where the socio-economic and educational disparities are greatest and the private educational choices are most abundant. The experimental nature of the MPCP places this case in direct contrast to Brookfield where we found no relationship between Milwaukee county's size and the challenged financing scheme. See Brookfield, 144 Wis. 2d at 920. Therefore, the second element of the Brookfield test is satisfied.
The third element of the Brookfield test requires that the classification not be based only on existing circumstances. Rather, "the classification must be subject to being open, such that other cities could join the class." Granted, the title of the statute is "Milwaukee Parental Choice Program." However, the statute is located in ch. 119, Stats., which addresses first class city schools and is applicable, by virtue of sec. 119.01, Stats., to cities of the first class. There are two requirements for a city to be of the first class. The city must have a population of at least 150,000 and the city's mayor must make an official proclamation that the city is of the first class. See sec. 62.05, Stats.
Presently, Milwaukee, with a population of 628,088, is the only city in Wisconsin which is officially a first class city. However, it is not the only city in Wisconsin which qualifies for such status, nor is the classification *536limited only to Milwaukee. Madison is large enough to qualify as a city of the first class. Madison has a population of 191,262. If the mayor of Madison officially declares Madison to be a first class city, it will be subject to all legislation affecting cities of the first class, including the parental choice program. Therefore, we conclude that the classification is subject to being open and is not based only on existing circumstances. The third element of the Brookfield test is satisfied.
The fourth element of the Brookfield test requires that the law be applied equally to all members of the class. As mentioned earlier, there is only one member of the class at the present time. Milwaukee is the only official first class city. However, if Madison or any other qualifying city were to become an official first class city, then there appears nothing to indicate that the benefits and obligations of the MPCP would not equally apply to these additional members. Therefore, we find that the law would apply equally to all cities of the first class. The fourth element of the Brookfield test is also satisfied.
The fifth, and final, element of the Brookfield test which is applicable to the present case requires that "the characteristics of each class should be so far different from those of the other classes so as to reasonably suggest at least the propriety, having regard to the public good, of substantially different legislation." The satisfaction of this element has already been addressed. Supra at 527-529. The immense disparity in the socio-economic conditions and educational problems in the MPS as well as the greatest potential private educational choices in the urban area of Milwaukee create the ideal testing ground for experimental legislation such as the MPCP. *537Therefore, we find that the MPCP also satisfies the fifth element of the Brookfield test.
The MPCP satisfies all elements of the Brookfield classification test. Therefore, we hold that the MPCP is not a private or local bill within the meaning of Wis. Const, art. IV, sec. 18 and, thus, not subject to its procedural requirements. We emphasize that the MPCP is not a private or local bill because it satisfies the applicable tests, not because of the amount of legislative consideration afforded to it.
II. THE UNIFORMITY CLAUSE
Wisconsin Constitution art. X, sec. 3 states:
The legislature shall provide by law for the establishment of district schools, which shall be as nearly uniform as practicable; and such schools shall be free and without charge for tuition to all children between the ages of 4 and 20 years.
This court has stated on several occasions that the requirement of uniformity "applies to the districts after they are formed, — to the 'character of instruction' given, — rather than to the means by which they are established and their boundaries fixed." Kukor v. Grover, 148 Wis. 2d 469, 486, 436 N.W.2d 568 (1989) (citing State ex rel. Zilisch v. Auer, 197 Wis. 284, 289-90, 223 N.W. 123 (1928)). Furthermore, the Kukor court concluded that "character of instruction" refers to that of "district schools" and is legislatively regulated by sec. 121.02, Stats.
Chaney argues that the MPCP violates the uniformity clause of Wis. Const, art. X, sec. 3. The thrust of Chaney's argument involves two steps: (1) the partici*538pating private schools are "district schools" within the meaning of the uniformity clause; and (2) by offering a "character of instruction" that is different from the one found under the mandate of sec. 121.02, the participating private schools violate the uniformity clause. The key to this argument is whether private schools participating in the MPCP are considered "district schools" for the purposes of the uniformity clause.
In Comstock v. Jt. School Dist. No. 1, 65 Wis. 631, 636-37, 27 N.W. 829 (1886), this court held that a statute allowing school districts to determine whether to admit nonresident school children did not violate the uniformity clause. In that case, we declared that "when the legislature has provided for each such child the privileges of a district school, which he or she may freely enjoy, the constitutional requirement in that behalf is complied with." Id. at 636-37. Thereafter, the legislature is free to act as it deems proper.
This sentiment was reiterated in several subsequent cases and most recently in Kukor, 148 Wis. 2d at 496-97. In Kukor, we found that a statutory school finance system did not violate Wis. Const, art. X, sec. 3 because every Wisconsin student has an opportunity to attend a public school with uniform character of instruction.
The MPCP unambiguously refers to nonsectarian private schools. "Private school" is a defined term under sec. 115.001(3r), Stats., and means "an institution with a private educational program that meets all of the criteria under s. 118.165(1) or is determined to be a private school by the state superintendent under s. 118.167." We assume that the legislature was aware of this statutory meaning and intended to use "private school" in the MPCP as a statutory term of art.
Similar to the legislation in Kukor, the MPCP in no way deprives any student the opportunity to attend a *539public school with a uniform character of education. Even these students participating in the program may withdraw at any time and return to a public school. The uniformity clause clearly was intended to assure certain minimal educational opportunities for the children of Wisconsin. It does not require the legislature to ensure that all of the children in Wisconsin receive a free uniform basic education. Rather, the uniformity clause requires the legislature to provide the opportunity for all children in Wisconsin to receive a free uniform basic education. The legislature has done so. The MPCP merely reflects a legislative desire to do more than that which is constitutionally mandated.
Therefore, we hold that the private schools participating in the MPCP do not constitute "district schools" for purposes of the uniformity clause. The legislature has fulfilled its constitutional duty to provide for the basic education of our children. Their experimental attempts to improve upon that foundation in no way denies any student the opportunity to receive the basic education in the public school system.
Nevertheless, the MPS argues that the method which the state has chosen to fund the program indicates that the legislature considered this program part of the basic public education delivery system and, thus, subject to Wis. Const, art. X, sec. 3 requirements of uniformity. As noted earlier, participating private schools receive public monies under the MPCP for the education of participating students. Chaney argues that a school supported by public taxation is a "public school" by definition under sec. 115.01, Stats.
Under this theory, any school that accepted public monies would become a "district school" which is subject *540to Wis. Const. art. X, sec. 3. However, this theory flies directly in the face of past decisions by this court. In State ex rel. Warren v. Reuter, 44 Wis. 2d 201, 216, 170 N.W.2d 790 (1969), we held that the appropriation of public funds to a private entity need only be accompanied by such controls as are necessary to fulfill the public purpose required. Depending on the circumstances, these controls do not necessarily have to be the same as those regulating similar public agencies. A more detailed analysis of this area is presented in the following issue.
In no case have we held that the mere appropriation of public monies to a private school transforms that school into a public school. We decline the opportunity to adopt such a conclusion here.
III. THE PUBLIC PURPOSE DOCTRINE
Chaney also argues that the public purpose doctrine prohibits the legislature from authorizing the expenditure of public funds for the basic education of students to private schools without adequate supervision and controls. Therefore, Chaney concludes that the MPCP violates the public purpose doctrine because the program lacks adequate supervision and controls.
Although the public purpose doctrine is not an express provision of the Wisconsin Constitution, this court has long held that public expenditures may be made only for public purposes. Reuter, 44 Wis. 2d at 211. In Reuter, we stated, "[w]e need not go into the origin or the validity of the doctrine which commands that public funds can only be used for public purposes. The doctrine is beyond contention." Id.
*541In considering questions of "public purpose," a legislative determination of public purpose should be given great weight because " 'the hierarchy of community values is best determined by the will of the electorate' and that 'legislative decisions are more representative of popular opinion because individuals have greater access to their legislative representatives.' " State ex rel. Bowman v. Barczak, 34 Wis. 2d 57, 65, 148 N.W.2d 683 (1967) (citations omitted). Without clear evidence of unconstitutionality, "the court cannot further weigh the adequacy of the need or the wisdom of the method" chosen by the legislature to satisfy the public purpose. State ex rel. Warren v. Nusbaum, 59 Wis. 2d 391, 414, 208 N.W.2d 780 (1973).
No party disputes that education constitutes a valid public purpose, nor that private schools may be employed to further that purpose. Rather, the parties dispute whether the private schools participating in the MPCP are under proper government control and supervision, as required by Wisconsin Indus. Sch. for Girls v. Clark Co., 103 Wis. 651, 668, 79 N.W. 422 (1899).
Chaney and, particularly, Superintendent Grover contend the controls in the MPCP over participating private schools are woefully inadequate and insist that these schools be subject to the stricter requirements of sec. 121.02, Stats. MPCP advocates, on the other hand, believe the statutory controls applicable to private schools coupled with parental involvement suffice to ensure the public purpose is met. The circuit court agreed with the MPCP advocates' contention, as we do.
The present situation is similar to that faced by this court in Reuter. In Reuter, we upheld an appropriation of public funds to the Marquette School of Medicine for *542the purpose of providing quality medical education. Reuter, 44 Wis. 2d at 207. To test the propriety of expending public monies to a private institution for public purpose, this court must determine whether the private institution is under reasonable regulations for control and accountability to secure public interests. Id. at 215-16. "Only such control and accountability as is reasonably necessary under the circumstances to attach the public purpose is required." Id. at 216.
Chaney attempts to distinguish the present situation from Reuter in two main ways. First, Chaney argues that private schools participating in the MPCP may do whatever they want with the public money that they receive, whereas the funds in Reuter were earmarked for "medical education, teaching and research." Chaney is facially correct in that no express limitations exist on the use of the funds paid to private schools through the MPCP. However, the private schools must still provide their students with an education. It simply does not matter how the school spends the money so long as it gives the participating student an education that complies with sec. 118.165, Stats., in return for the money. Public schools face a similar situation. While the use of certain state aid to school districts is limited under sec. 121.007, Stats., the public schools must continue to provide a basic education to its students regardless of how and to what extent its programs and investments are funded.
Second, Chaney argues that private schools participating in the MPCP have no duty to demonstrate any institutional quality, whereas Marquette University was accredited by an independent national organization as well as federal and state agencies. See Reuter, 44 Wis. 2d at 217. In effect, Chaney is challenging the quality of *543education provided by the private schools participating in the program.
The MPCP specifically allows participating students to attend a "nonsectarian private school." See sec. 119.23(2)(a), Stats. "Private school" has an express statutory definition under sec. 115.001 (3r), Stats., which requires the institution to meet all of the criteria under secs. 118.165(1) or 118.167, Stats.
Under sec. 118.165, Stats., a private school must:
(1) be organized to primarily provide private or religious-based education;
(2) be privately controlled;
(3) provide at least 875 hours of instruction each school year;
(4) provide a sequentially progressive curriculum of fundamental instructions in reading, language arts, mathematics, social studies, science, and health;
(5) not be operated or instituted for the purpose of avoiding or circumventing compulsory school attendance; and
(6) have pupils return home not less than two months of each year unless the institution is also licensed as a child welfare agency.
Even though private schools are not subject to the same amount of controls which are applicable to public schools, they are subject to a significant amount of regulation which is geared toward providing a sequentially progressive curriculum. This issue is uniquely complicated, however, by the underlying thesis of the MPCP that less bureaucracy coupled with parental choice improves educational quality.
*544Keenly aware of this potential problem, the legislature included within the MPCP sufficient supervision and control measures. The State Superintendent is required to annually report to the legislature comparing the students participating in the MPCP with students in the MPS. The report includes data on academic achievement, daily attendance, percentage of dropouts, and percentage of pupils suspended and expelled. The State Superintendent is authorized to conduct financial and performance audits on the program, and the Legislative Audit Bureau is mandated to perform financial and performance evaluation. We believe that these detailed reports and evaluations in conjunction with the private school requirements under secs. 118.165(1) and 118.167, Stats., provide sufficient and reasonable control under the circumstances to attain the public purpose to which this legislation is directed.
Control is also fashioned within the MPCP in the form of parental choice. Parents generally know their children better than anyone. The program allows participating parents to choose a school with an environment that matches their child's personality, with a curriculum that matches their child's interest and needs, and with a location that is convenient. If the private school does not meet the parents' expectations, the parents may remove the child from the school and go elsewhere. In this way, parental choice preserves accountability for the best interests of the children.
In Wisconsin v. Yoder, 406 U.S. 205 (1972), the United States Supreme Court also recognized the importance and the strong tradition of parental choice in education. Using a balancing of interests test, the Yoder Court held that the First and Fourteenth Amendments *545prevent the state from compelling Amish parents to cause their children to attend formed high school to age sixteen. Id. at 234. In so deciding, it stated:
Providing public schools ranks at the very apex of the function of a State. Yet even this paramount responsibility . . . yield[s] to the right of parents to provide an equivalent education in a privately operated system.
Id. at 213. Yoder involved the protection of the Religion Clauses, whereas the present case involves purely secular considerations. However, the Yoder Court declared that purely secular considerations "may not be interposed as a barrier to reasonable state regulation of education." Id. at 215 (emphasis added). We have determined in this case that the reporting and private school requirements applicable to the MPCP provide sufficient and reasonable state control under the circumstances.
Further, the cost of education and the funds available for education are dependent upon the taxpayers' ability to fund an intensive public educational program. The amount of money allocated to a private school participating in the MPCP to educate a participating student is less than 40 percent of the full cost of educating that same student in the MPS. Each of the participating private schools is willing to accept the responsibility of educating a child for the $2,500 granted by the state.23 In *546contrast, it costs the MPS an average of $6,451 to educate each student.24 At most, $2.5 million of public funds will be appropriated to fund this experimental legislation. This amount is inconsequential compared to the more than $6.4 billion that is annually expended for public education in Wisconsin.25 The amount of money to fund the MPCP represents only about four one-hundredths of one percent (.04 percent) of the public money allocated for public education throughout the state. Therefore, we hold that the MPCP does not violate the public purpose doctrine because the MPCP contains sufficient and reasonable controls to attain its public purpose.
We conclude that the Milwaukee Parental Choice Program passes constitutional scrutiny in all issues presented before this court. Accordingly, we reverse the decision of the court of appeals.
By the Court. — The decision of the court of appeals is reversed.
Article IV, sec. 18 of the Wisconsin Constitution states:
No private or local bill which may be passed by the legislature shall embrace more than one subject, and that shall he expressed in the title.
Wisconsin is the first state in the nation to experiment with a parental choice program which involves the use of private schools as an alternative to the public school system. The program is an attempt to identify factors which could improve the quality of education. Clearly, the program is not only of statewide importance but national significance as well because education of our citizens knows no boundaries and other states could benefit from the knowledge resulting from this innovative experiment.
The citizens of Wisconsin have a long and proud tradition of striving for excellence and an improved quality of life. Our state flag proudly displays our motto in its statement of "Forward." The forum of education is just one area in which Wisconsin demonstrates its excellence and innovation. The University of Wisconsin System is widely recognized as one of the nation's leading systems of public higher education. Furthermore, Wisconsin was a pioneer in the establishment of vocational and technical *513schools. The MPCP represents another illustration of Wisconsin's innovation and willingness to lead the nation in its attempts to further improve the quality of education and life.
42 U.S.C. sec. 2000d states:
No person in the United States shall, on the ground of race, color, or national origin, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance.
Under sec. 119.23(7)(a), Stats., each private school participating in the program must meet at least one of the following standards:
*5151. At least 70% of the pupils in the program advance one grade level each year.
2. The private school's average attendance rate for the pupils in the program is at least 90%.
3. At least 80% of the pupils in the program demonstrate significant academic progress.
4. At least 70% of the families of pupils in the program meet parent involvement criteria established by the private school.
See secs. 119.23(7)(b), (8), and (9), Stats.
Superintendent Grover attempted to require private schools that wished to participate in the program to execute complex forms certifying that they met numerous requirements in excess of those specified under sec. 118.165, Stats., or in the MPCP.
"Logrolling” is the legislative practice of embracing in one bill several distinct matters, none of which could singly obtain the assent of the legislature, and then procuring its passage by a combination of the minorities in favor of each of the separate measures into a majority that will adopt them all. Black's Law Dictionary 942 (6th ed. 1990).
Justice Bablitch's dissenting opinion quite adamantly argues that, "[i]f there were sufficient votes in the Senate to pass the bill as a separate piece of legislation, the Senate would have done so, thereby avoiding any possible constitutional challenge under this section. Given that they did not do so, it is clear that the votes were not there to pass it as a separate piece of legislation." Bablitch dissent at 572. While we share his concern about the potential for logrolling, Justice Bablitch presumes a fact that is not supported by the record. There is no indication in the record that there were insufficient votes to pass the MPCP as a separate piece of legislation in the Senate. A plausible alternative explanation could include the Senate's concern that a worthy *523piece of legislation may be thwarted by the close of a legislative session. Without adequate evidence in the record, we are less inclined to presume the evil that Justice Bablitch so strongly suggests.
We are quite concerned about the dissent's indictment of the legislature's integrity. The legislative branch exists to provide an essential and valued function. Legislators are elected by the public to represent the public's interest. Presumably, they are elected based on many factors, including their wisdom and integrity. We are unwilling to attack that integrity unless evidence exists to the contrary.
The circumstances of the present case allow us to presume constitutionality for the process in which this legislation was enacted. Justice Bablitch's analogy to the granting of a liquor license is so dissimilar to the MPCP legislation that it merits very little discussion. See Bablitch dissent at 574-75. We shall point out only that the granting of a liquor license concerns only one or few individuals and is not likely to grasp the attention of the legislature. In contrast, improving educational quality is a statewide concern and, as mentioned, the record is replete with evidence that the MPCP received a substantial amount of serious deliberation by the legislature.
Section 990.001(6), Stats.; Wisconsin Valley Imp. Co. v. Public Service Comm., 9 Wis. 2d 606, 618, 101 N.W.2d 798 (1960).
The court of appeals suggests that we adopt a modified Brewers test to accommodate "experimental" legislation. Davis, 159 Wis. 2d at 167. It is their contention that the nature of the experiment, not the classification, should be subject to the test for a private or local bill. However, creating such a test would unnecessarily further complicate this area of law. In this case, the classification tests adequately address and incorporate the nature of experimental legislation.
Milwaukee Public Schools, Indicators of Educational Effectiveness 12 (1990).
State of Wisconsin, Legislative Reference Bureau, 1991-92 Blue Book 615 (1991).
State of Wisconsin, Legislative Reference Bureau, 1991-92 Blue Book 797 (1991).
State of Wisconsin, Legislative Reference Bureau, 1989-90 Blue Book 841 (1990).
U.S. Department of Education, Center of Education Studies, The Condition of Education: A Statistical Report 20 (1987).
Bast & Wittmann, The Case for Education Choice (1990).
See Gretchen Schuldt, Many black freshmen at less than 'D', Milwaukee Sentinel, Apr. 23, 1991, at 1A.
Milwaukee Public Schools, Indicators of Educational Effectiveness (1990).
Public Policy Forum, Public Schooling in the Milwaukee Metropolitan Area, 37-39 (1988).
Chubb & Moe's conclusion that school organization can directly affect student achievement has been criticized by some commentators. For example, Professor John F. Witte of the University of Wisconsin-Madison Department of Political Science states that the comprehensive measure of school organization incorporates fifty variables and, thus, makes Chubb & Moe's analysis problematic and their combined inference totally unconvincing. Witte, "Public Subsidies for Private Schools: Implications for Wisconsin's Reform Efforts," the Robert M. LaFollette Institute of Public Affairs, University of Wisconsin-Madison 21 (1991). Professor Witte contends that the immense number of variables associated with school organization makes it almost impossible to isolate effects of specific organizational practices. Id.
However, in the absence of a constitutional challenge, it is not for us to determine the propriety of choosing one approach over another. This task is more appropriately undertaken by the legislature who is better equipped and possesses greater resources to hold public hearings and grasp public sentiment. As we stated in State ex rel. Bowman v. Barczak, 34 Wis. 2d 57, 65, 148 N.W.2d 683 (1967), "legislative decisions are more representative of popular opinion because individuals have greater access to their legislative representatives."
The scope of the MPCP was necessarily limited to the boundaries of a first class city. Such restriction in the scope of an experiment is necessary for the controlled, orderly, and efficient administration of the experiment. We do not conclude, as does Justice Bablitch's dissent, that "the authors of this legislation intended to benefit only private schools located within the city [of Milwaukee]. ” Bablitch dissent at 569. Rather, the intended beneficiary of the program is the state at large, which can learn from the results of the program. The fact that students participating in the MPCP may only attend private schools located within the first class city is merely a consequence of the boundaries of the experiment. Transportation costs require that available private schools be in the proximity of the students' residence. It would be impractical and absurd to transport a student participating in the program from Milwaukee to a La Crosse or even a Waukesha private school.
We hasten to note that the program does not appear to offer any financial advantage or windfall to the participating private schools. There is no evidence that the modest $2,500 per student that is received by participating private schools covers the cost of educating the student. The legislature determined the amount to be paid to the participating private schools without evidence of the actual cost incurred by the private school to *546provide an education to each enrolled student. Because $2,500 is less than 40 percent of the cost of educating a student in the MPS, we must assume that the participating private schools are either more efficient than public schools or discounting some of the cost to educate the students participating in the program.
M. Fisher, "Fiscal Accountability in Milwaukee's Public Elementary Schools," Wisconsin Policy Research Institute Report, Vol. 3, no. 4 (Sept. 1990).
State of Wisconsin, Legislative Reference Bureau, 1991-92 Blue Book 620 (1991).