dissenting in part and concurring in part.
¶ 47 I concur with the majority on two points and disagree with it on two others. First, I agree with the majority that Kotter-man v. Killian, 193 Ariz. 273, 972 P.2d 606 (1999), controls the issues Appellants raise under Article 2, Section 12 and Article 9, Section 10 of the Arizona Constitution and that the use of the term “taxes” in the purpose statement to A.R.S. § 43-1183 does not change that result. Appellants have preserved their arguments on those issues and they are best made to the Arizona Supreme Court. Second, I concur with the majority’s conclusion that § 43-1183 by itself does not commandeer our public school system or subject that system to sectarian control in violation of Sections 20 and 26 of the Arizona Enabling Act.
¶ 48 I dissent from the majority’s conclusion that § 43-1183 does not violate the Establishment Clause of the First Amendment to the United States Constitution under Zel-man v. Simmons-H arris, 536 U.S. 639, 122 S.Ct. 2460,153 L.Ed.2d 604 (2002). Dismissal of the complaint was not appropriate because on this record the tax credit program is unconstitutional under the First Amendment both as to secular purpose and neutrality towards religion. Since there are genuine disputes of fact whether a tax credit program giving aid solely to private charitable and educational institutions which directly discriminate against children based on their religion is predominately serving a secular purpose or is neutral with respect to religion, the judgment should be reversed. However, while I conclude that the tax credit scheme meets one element of Establishment Clause analysis, whether it qualifies as a program of true private choice, I do so based on the universe of choices available to parents as mandated by Zelman.
I. Procedural Status of the Case and Standard of Review
¶ 49 The majority’s analysis appears in a factual vacuum because neither the superior court nor the majority properly address the procedural status of this case, the factual record, and how those affect our standard of review. The superior court did not address the factual record because it dismissed the complaint on the assumption that Kotterman controlled this case except for the issue of whether the tax scheme violated the State’s obligation to maintain a general and uniform non-sectarian public school system.11 The superior court’s assumption was erroneous because, as the majority acknowledges, Zel-man controls our Establishment Clause analysis. We are bound by United States Supreme Court decisions on federal constitutional issues when the Court issues a decision that differs from a prior Arizona Supreme Court decision on the same subject. Hernandez-Gomez v. Volkswagen of Am., Inc., 201 Ariz. 141, 143-44, ¶ 8, 32 P.3d 424, 426-27 (App.2001); State v. Superior Court, 2 Ariz.App. 458, 460, 409 P.2d 742, 744 (1966). See also Rodriguez de Quijas v. Shearson/Am. Express, Inc., 490 U.S. 477, 484, 109 S.Ct. 1917, 104 L.Ed.2d 526 (1989) (if precedent of Court directly applies but appears to rest on reasons rejected in other lines of cases, court of appeals must follow case which directly controls). Zelman, decided in 2002, is a United State Supreme Court case regarding indirect government aid to religious schools under the Establishment Clause. Kotterman, decided in 1999, is an Arizona case regarding the same subject. Thus, because the two cases concern the same subject and Zelman was decided after Kotterman, Zelman is the controlling precedent on the Establishment Clause issue.
*417¶ 50 While recognizing that Zelman controls the Establishment Clause issue, the majority proceeds with an analysis of Zelman with only a limited discussion of the factual record presented in documents attached to the motions papers. That status and record affect our review. In opposition to the motion to dismiss, the Appellants attached various documents to their memoranda, supporting allegations in the complaint. Those attachments included documents that were attached to briefs filed in a related case in the United States Court of Appeals for the Ninth Circuit, including Department of Revenue reports analyzing School Tuition Organizations (“STOs”) and scholarships offered as well as the amounts of tax credits the STOs controlled. When papers are attached to a response to a motion to dismiss and the superior court considers those documents or does not strike them, the motion to dismiss is treated as a motion for summary judgment. Ariz. R. Civ. P. 12(b); Vasquez v. State, 220 Ariz. 304, 308, ¶ 8, 206 P.3d 753, 757 (App.2008) (“Because the court considered ‘matters outside the pleading,’ it should have treated the motion [to dismiss] as one for summary judgment.”) (citation omitted); see Jacobson v. AEG Capital Corp., 50 F.3d 1493, 1496 (9th Cir.1995) (“If matters outside the pleadings are submitted, the motion to dismiss under Federal Rule of Civil Procedure 12(b)(6) is treated as one for summary judgment under Federal Rule of Civil Procedure 56.”).
¶ 51 Summary judgment may be granted when “there is no genuine issue as to any material fact and [] the moving party is entitled to judgment as a matter of law.” Ariz. R. Civ. P. 56(c). In reviewing a motion for summary judgment, we determine de novo whether any genuine issue of material fact exists and whether the trial court properly applied the law. Eller Media Co. v. City of Tucson, 198 Ariz. 127, 130, ¶ 4, 7 P.3d 136, 139 (App.2000) (citation omitted). We review the decision on the record made in the superior court. Phoenix Baptist Hosp. & Med. Ctr., Inc. v. Aiken, 179 Ariz. 289, 292, 877 P.2d 1345, 1348 (App.1994) (citation omitted). We view the facts and the inferences to be drawn from those facts in the light most favorable to the party against whom judgment was entered. Prince v. City of Apache Junction, 185 Ariz. 43, 45, 912 P.2d 47, 49 (App.1996) (citation omitted). We also consider legal questions de novo. Inch v. McPherson, 176 Ariz. 132, 136, 859 P.2d 755, 759 (App.1993) (citation omitted). Accordingly, before addressing the law, we must summarize the factual record under the appropriate standards.
II. Factual Background12
¶ 52 In their amended complaint, Appellants alleged that pursuant to § 43-1183, through 2011 all corporations that pay Arizona state income tax can receive a tax credit against such taxes for cash contributions made to STOs, if the contributions were approved by the Arizona Department of Revenue (“DOR”).13 According to the complaint, those contributions would diminish the State’s general fund revenue and forty-six percent of that fund is expended annually to finance Arizona’s public school system. While DOR’s discretion to approve the contributions is limited only by the aggregate cap annually imposed on tax credits under the statute,14 there is no cap on any individual corporate contribution. The contribution can exceed the taxes owed by the corporation to the state that tax year and carry forward the excess credit to offset taxes for up to five consecutive years. A.R.S. § 43-1183(F). In *418contrast, the scheme caps the aggregate amount of tax credits per fiscal year to $10 million for fiscal year 2006-07 with the cap to increase by twenty percent each and every future year over the prior year. According to the DOR, the aggregate cap is $12 million for fiscal year 2008, $14.4 million for fiscal year 2009, $17.28 million for fiscal year 2010 and $20,736 million for fiscal year 2011. Pursuant to A.R.S. § 43 — 1183(C)(3), DOR must approve all proposed contributions on a first-come, first-serve basis, provided the contributions do not exceed the aggregate maximum contribution for the fiscal year.
¶ 53 The tax credit money can be used only for nonpublic schools. Any STO must use at least ninety percent of its annual revenue to award scholarship tuition grants to students attending non-public elementary or secondary schools in Arizona. An STO may award scholarships only to children whose family incomes do not exceed one hundred eighty-five percent of the income limit used to qualify a child for school lunches under the National School Lunch and Child Nutrition Acts, approximately $70,000 per year for a family of four. A.R.S. § 43 — 1183(J). No similar tax credit is offered for students attending public schools to offset fees or expenses at those schools.
¶ 54 The statutory scheme does not limit either the STOs or the recipient private schools from discriminating on the basis of religion on who will receive tuition scholarships from the tax credits. Indeed, the complaint alleges both the STOs and the recipient schools do and will discriminate on the basis of religion. Thus, except for attending a qualified private school and family income, no standards are set in the statute for determining who will receive scholarship funds. The statute does not prohibit STOs from discriminating on the basis of race, sex, religion or any other basis.15 A.R.S. § 43-1183(Q)(2). The statute permits schools receiving the scholarship money to discriminate on the basis of religion and sex, but not race, color, handicap, familial status or national origin. A.R.S. § 43-1183(Q)(l)(a).
¶ 55 The Appellants analogize to the similar individual tax credit scheme which was the subject of Kotterman to allege the effect of the corporate tax credit scheme. They do so because they allege the same STOs eligible under the individual credit scheme are the STOs for the corporate scheme. Accordingly, there are approximately fifty-four STOs eligible for the funds under the scheme, twenty-four of which are religious organizations or are affiliated with religious organizations. Those religiously-based STOs annually distribute more than seventy percent of all the tax credited funds under the similar individual tax credit scheme. In 2005, $42 million was contributed to STOs under: the individual tax credit scheme of which more than $30 million was contributed to religiously affiliated STOs.16 All or almost all scholarships awarded by religiously affiliated STOs under the individual tax credit act allegedly are awarded on a religiously discriminatory basis and on the express condition that the recipient attend a religious school for a particular religious denomination. Appellants allege that will be the same under the corporate tax credit scheme. Like the STOs, religious schools receiving these funds are free to discriminate against students on the basis of religion and sex and allegedly do discriminate by limiting acceptance to students based on their religious beliefs while requiring students to attend and participate in religious ceremonies and observances. See A.R.S. § 43 — 1183(Q)(1 )(a). *419Aside from the scholarship cap, the only limitation on STOs is that they must not limit scholarships to just one school. A.R.S. § 43-1183(Q)(2)(b).
¶ 56 While the record reflects fifty-six STOs currently receiving donations and making scholarships to attend approximately 357 schools, that may be very different under the corporate tax scheme. This is because unlike the individual scheme, § 43-1183 has an aggregate annual limit of tax credits. Thus, it is possible under the corporate scheme to have several corporations use the entire aggregate tax credit in any year (with carryovers to future years) to fund scholarships to one STO, which will only fund schools of one religious denomination; and those schools will not accept students unless they are of that denomination and/or agree to participate in the religious observances of that denomination.
III. Analysis
¶ 57 The current test for school vouchers (or in this case tax credits) for Establishment Clause purposes is found in Zelman. The Court in Zelman held that in eases involving indirect governmental aid to religious schools, four criteria must be met to comport with the Establishment Clause. First, the aid program’s predominate purpose must be secular and not to advance or inhibit religion. 536 U.S. at 648-49, 122 S.Ct. 2460. In addition, to ensure that the program’s effect does not advance or inhibit religion, the indirect aid program must: (1) be “[ejntirely neutral with respect to religion,” (2) provide “benefits directly to a wide spectrum of individuals,” defined without reference to religion, and (3) permit “such individuals to exercise genuine choice among options public and private, secular and religious.” Zelman, 536 U.S. at 662, 122 S.Ct. 2460; Ira C. Lupu & Robert W. Tuttle, Zelman’s Future: Vouchers, Sectarian Providers, and the Next Round of Constitutional Battles, 78 Notre Dame L.Rev. 917, 928-29 (2003) (“Lupu”). If any one of these criteria is not met, “the program should be struck down under the Establishment Clause.” Zelman, 536 U.S. at 669, 122 S.Ct. 2460 (O’Connor, J., concurring).
1158 Applying these three effects criteria, Zelman considered other relevant factors. Specifically, the Court found persuasive that the Cleveland voucher program at issue: (1) was part of a multifaceted attempt by the State to improve educational opportunities, (2) allowed all schools within the district to participate, (3) only gave preference to low-income families, allowing no other preferences, and (4) did not provide financial incentives to skew the program toward religious schools. Zelman, 536 U.S. at 653, 122 S.Ct. 2460. The Court reached its holding based on the aggregate of these factors. It held that by looking at the Cleveland program as a whole, the voucher system was one of true private choice and it did not have the effect of advancing religion.
¶ 59 In reaching its conclusion, the Court in Zelman also distinguished Committee for Public Education & Religious Liberty v. Ny-quist, 413 U.S. 756, 93 S.Ct. 2993, 37 L.Ed.2d 948 (1973). In Nyquist, the Court held in part that providing tuition reimbursement and tax benefits to parents of children attending nonpublic schools was unconstitutional under the Establishment Clause because its effect was to advance religion. Id. at 780-94, 93 S.Ct. 2993. The Court in Zelman held that the Cleveland voucher system was unlike the tax exemption and tuition reimbursement system in Nyquist for two general reasons. First, in Nyquist, the function was unmistakably to provide financial support for sectarian institutions, the tax benefits were unrelated to the amount of any money actually expended by any parent, the tuition reimbursement was designed explicitly to offer an incentive to parents to send their children to sectarian schools, and the program flatly prohibited the participation of any public school or parent of any public school enrollee. Zelman, 536 U.S. at 661, 122 S.Ct. 2460. Second, the Court noted that in Nyquist it had reserved judgment as to eases involving “some form of public assistance ... made available generally without regard to the sectarian-nonsectarian, or public-nonpublic nature of the institution benefit-ted,” which was the question presented in Zelman. Id. (quoting Nyquist, 413 U.S. at 782-83 n. 38, 93 S.Ct. 2955).
*420¶ 60 Here, the majority applies the same analytical strategy as Zelman and reaches the same conclusion. To reach its holding, the majority looks at the Arizona corporate tax scheme as a whole by addressing each of the above items. It then concludes that the corporate tax credit program does not have the stated purpose or the effect of advancing religion and is a program of true private choice. Supra ¶¶ 13-28. The majority’s analysis, however, overlooks several aspects of the Arizona program that unconstitutionally have the effect of advancing religion and that negate genuine choice. To understand that, one must compare the program in Zel-man and the tax scheme in Arizona.
¶ 61 In Zelman, Ohio enacted a school voucher program because Cleveland schools had failed to such an extent that a federal court order required that the state superintendent take over the district’s management and operation. Zelman, 536 U.S. at 644-45, 122 S.Ct. 2460. The government provided tuition vouchers directly to parents based on financial need, giving preference to families with incomes two hundred pei'eent below the poverty line. Id. at 646, 122 S.Ct. 2460. Those parents could use the vouchers for private schools in Cleveland or for public schools in adjacent districts. Id. at 645, 122 S.Ct. 2460. If parents chose to keep their child in a Cleveland public school, they could use voucher money to hire a private tutor for their child. Id. at 646, 122 S.Ct. 2460. The participating private schools were precluded from discriminating on the basis of religion. Id. The state superintendent determined the number of vouchers available on an annual basis. Id. at 646, n. 2, 122 S.Ct. 2460.
¶ 62 The Arizona tax credit program is different from Cleveland’s voucher program in a number of ways. As alleged by the Appellants, Arizona gives tax credit money to STOs which then distribute that money in the form of scholarships to qualifying students to use only in private schools which the STOs have pre-selected. Religious STOs control and distribute approximately seventy percent of the tax credit scholarships. Those STOs can require students to attend a religious school of a particular denomination as a condition of their scholarship award. In addition, religious STOs can refuse to grant scholarships to students who do not subscribe to the same denomination as that of the STO. The schools receiving STO scholarship students can and allegedly do discriminate on the basis of religion and may require students to participate in religious observances as a condition of their scholarship. Finally, the corporate tax credit program does not directly limit the amount that corporations can donate, but it does impose a limit of $10 million in the aggregate amount donated by all corporations and credits are only available on a first-come, first-served basis.
¶ 63 For the reasons stated below, I conclude that the record raises questions as to the tax program’s secular purpose, that Arizona tax scheme is not neutral with respect to religion and does not provide benefits directly to a wide spectrum of individuals defined without reference to religion. Either of these factors is sufficient to reverse and remand this matter to the superior court. While I conclude that the tax program meets the third prong of the effects test in Zelman, I reach that conclusion in a different manner than the majority.17
A. Secular Intent
¶ 64 As Zelman points out, the first test for Establishment Clause analysis is whether the purpose of the program is secular in nature or the government acted with the purpose of advancing or inhibiting religion. Id. at 648-49, 122 S.Ct. 2460. In Zelman, the Court dispatched that issue in a single sentence noting there was “no dispute that the [voucher] program challenged [in Zel-man] was enacted for the valid secular purpose of providing educational assistance to *421poor children in a demonstrably failing public school system.” Id. at 649, 122 S.Ct. 2460.
¶ 65 The majority takes the same tact here, concluding that the stated purpose of the tax credit scheme is to encourage businesses to direct a portion of their taxes to provide scholarships “‘in order to improve education.’ ” Supra ¶ 13. As the majority explains, by providing such scholarships through tax credits, competition in education is increased and competition is bound to improve the quality of both private and public schools. Supra ¶¶ 16-18.
¶ 66 I agree with the majority that a system with both private and public schools, under certain circumstances, may foster competition resulting in improving education for most students. However, in conducting Establishment Clause analysis to determine purpose, while we give deference to the stated purpose out of a matter of judicial restraint, we are required to determine that the stated purpose must be sincere, legitimate and not a sham. Bowen v. Kendrick, 487 U.S. 589, 602-04, 108 S.Ct. 2562, 101 L.Ed.2d 520 (1988); Edwards v. Aguillard, 482 U.S. 578, 586-87, 107 S.Ct. 2573, 96 L.Ed.2d 510 (1987). Rather, as the United States Supreme Court most recently explained, we approach the intent test in Establishment Clause cases differently than we do in reviewing economic legislation. McCreary County v. ACLU of Ky., 545 U.S. 844, 865 n. 13, 125 S.Ct. 2722, 162 L.Ed.2d 729 (2005). Unlike the deference given to stated purposes in review of economic legislation, we review stated purposes in Establishment Clause cases to ensure the alleged secular purpose is genuine and is not secondary to a religious objective. Id. at 846, 125 S.Ct. 2722. The secular purpose test is not meant to be a “pushover” for any claim of secular objectives. Id. at 864, 125 S.Ct. 2722. As the Court has advised us, we must not be blind to predominating religious purpose. Edwards, 482 U.S. at 590, 107 S.Ct. 2573 (citing to Epperson v. Arkansas, 393 U.S. 97, 89 S.Ct. 266, 21 L.Ed.2d 228 (1968)).
¶ 67 There is evidence that the State is not seeking to increase competition between the public and private school sector, but actually diverting funds from the public to the private sector and decreasing spending for public schools. This would inhibit the ability of the public schools to truly compete and improve the services they provide.
¶ 68 The public schools rely on government funding for their sole or main source of general revenue and the State now offers a supplementary source of revenue for the private schools through tax credits. The State completed a cost analysis on the individual cost savings to the State’s general fund when tax credit moneys were spent for private school scholarships. Those documents reflect that the State will expend less general fund money for the public schools than the funds diverted through the tax credits. Vicki Murray, Fiscal Analysis of Arizona’s Corporate Tax Credit Scholarship Program at 1 (Oet.2006). That study is consistent with Arizona legislative studies stating that the loss of tax revenue, at least from the individual tax credit scheme for private school scholarships, would be offset by savings in having to teach fewer students in public schools.18 This reduction was not ignored by the Legislature when it considered the similar individual tax credit program for private school scholarship contributions.19
¶ 69 In sum, this implies that the State’s intent is not to continue to spend the same amounts of general fund moneys on fewer public school students, but to divert tax revenues from the general fund money to private schools. The public schools are thus provided with fewer resources as the State decreases public school funding in light of the diversion of such funds to private school *422scholarships. True competition arises when two independent entities compete for market share, not when the primary source of funding for one of the competitors diverts such funds to the other competitor. This lack of competitiveness is underscored by the fact that Arizona is currently near the bottom of all states in public school spending per student and the legislature is proposing massive decreases in public education spending due to the current fiscal crisis.20
¶ 70 The fact that the State may be seeking to divert funds from public schools to private schools, however, is insufficient for Establishment Clause analysis unless the predominate purpose is to advance religion. Zelman, 536 U.S. at 648-49, 122 S.Ct. 2460; Edwards, 482 U.S. at 599, 107 S.Ct. 2573 (Powell, J., concurring). Cf. McCreary County, 545 U.S. at 864, 125 S.Ct. 2722 (court must make sure secular purpose is not secondary to religious purpose). Here, there is evidence in the record and in the complaint that the State was aware that over seventy percent of the scholarship aid under the individual tax credit scheme was being used for sectarian private schools, many of which allegedly discriminated on the basis of religion in admitting students.
¶ 71 The majority attempts to limit the effect of this factual background by noting the Court in two Establishment Clause cases was not concerned with the statistics of the percentage of students who went to secular schools because that percentage was a result of private parental choice. Supra ¶ 25. The distinguishing fact here, however, is that these percentages show that a vast majority of the tax credit money is controlled by STOs, which allegedly discriminate on the basis of religion; thus precluding or at least inhibiting parents from sending their children to schools of their choice. Given that the same STOs were allegedly implementing both tax credit schemes and that the same results would incur under the corporate tax credit, there is at least a question of fact whether the real purpose of the programs was to advance religious education.21
¶ 72 I cannot determine from this record that the predominate purpose of the tax scheme is religious. Edwards, 482 U.S. at 599, 107 S.Ct. 2573 (Powell, J., concurring). This, however, is not from the failing of the parties, but rather from the fact the superior court assumed that this case was controlled by Kotterman when it was controlled by Zelman. I would, therefore, reverse and remand on this issue to allow the lower court to conduct the necessary inquiry on whether the predominate purpose violates the Establishment Clause. Accordingly, this matter should be remanded to the superior court to explore.
B. Neutrality and Benefits Awarded Without Reference to Religion
¶ 73 The majority concludes that the Arizona corporate tax credit program is neutral with respect to religion because the statute does not distinguish between sectarian and non-sectarian schools, and the credit is available to corporations without respect toward religion. According to the majority, those two factors combined with what it describes as layers of choice, result in a neutral program. Supra ¶¶ 24-25.
*423¶ 74 I disagree. In light of the differences between the Cleveland voucher program and Arizona’s tax program noted above, the Arizona program is unconstitutional under the Establishment Clause because it is not neutral with respect to religion and the aid is not provided without reference to religion — the first two prongs of the Zelman effect analysis. Indeed, the program is not neutral toward the ultimate beneficiaries, parents and students, even though those are precisely the beneficiaries that Zelman and its progeny seek to protect. Based on these differences, the tax credit program here is more like the tax relief struck down in Nyquist than the voucher program upheld in Zelman.
¶ 75 Zelman states that where a government aid program benefits a broad class of people and is neutral with respect toward religion, it is likely to be constitutional. 536 U.S. at 652, 122 S.Ct. 2460. While these first two requirements of the effects test overlap, they serve slightly different purposes. The first requirement, that the program must be “neutral in all respects toward religion[,]” 536 U.S. at 652, 122 S.Ct. 2460, means that there must be formal neutrality: “the classes of both the participating schools and the eligible students must be defined in nonreligious terms.” Lupu, 78 Notre Dame L.Rev. at 928. The second effects factor looks at how the program is effectuated: it must provide aid “[d]ireetly to a broad class of individuals defined without reference to religion.” Zelman, 536 U.S. at 653, 122 S.Ct. 2460. As explained by one commentator, this test:
[E'Jnsures that the formal neutrality required by the first criterion does not in fact represent a gerrymander in favor of a particular religious group; the more dispersed the benefits, the less likely any one religious group would be considered the intended beneficiary of government largesse.
Lupu, 78 Notre Dame L.Rev. at 928.
¶ 76 Thus, the first two criterion of the effects test mean that the program distributes aid to parents and students without reference to religion. Zelman, 536 U.S. at 651, 669, 122 S.Ct. 2460 (a program distributes aid in a neutral fashion when there is no “[differentiation based on the religious status of beneficiaries....”) (O’Connor, J., concurring). The Supreme Court has repeatedly emphasized this point.
¶ 77 In Mitchell v. Helms, 530 U.S. 793, 120 S.Ct. 2530, 147 L.Ed.2d 660 (2000), the federal government gave money directly to state agencies — local educational agencies (“LEAs”) and state educational agencies (“SEAs”) — which then distributed money to schools. Id. at 801-02, 120 S.Ct. 2530. The federal government required that the LEAs and SEAs distribute funds to both public and private schools, including religious schools, based on enrollment numbers. Id. at 802, 120 S.Ct. 2530. The Court upheld the program because aid was “[allocated on the basis of neutral, secular criteria ... and [was] made available to both religious and secular beneficiaries on a nondiseriminatory basis.” Id. at 795, 120 S.Ct. 2530. Although the programs in Mitchell and the present case are different in some ways, in both cases there is an intermediary organization which distributes funds. The LEAs and SEAs are similar to the STOs in that they channel government money for educational purposes to beneficiaries. In Mitchell, funds were distributed regardless of whether the school beneficiary was public, private, or religious, only that it met certain enrollment requirements. Here, some funds are distributed only if the parent or student subscribes to a particular religious sect.
¶ 78 Similarly to Mitchell, in Zobrest v. Catalina Foothills Sch. Dist., 509 U.S. 1, 113 S.Ct. 2462, 125 L.Ed.2d 1 (1993), government benefits were distributed “neutrally to any child qualifying as disabled.” Id. at 10, 113 S.Ct. 2462. In Witters v. Wash. Dep’t of Serv. for the Blind, 474 U.S. 481, 106 S.Ct. 748, 88 L.Ed.2d 846 (1986), government benefits were given to any visually handicapped student seeking vocational assistance. Id. at 483, 106 S.Ct. 748. In Mueller v. Allen, 463 U.S. 388, 103 S.Ct. 3062, 77 L.Ed.2d 721 (1983), government benefits were given to any parent who paid for school-related expenses. Id. at 390, 103 S.Ct. 3062. Finally, in Zelman, government benefits were distributed to any parent qualifying as low-income. 536 U.S. at 646, 122 S.Ct. 2460. These cases *424show that a government aid program is neutral if parents receive aid without first being asked if they subscribe to a certain religious sect or belief. In essence, the questions posed above before individuals received government benefits were: (1) Is your child disabled?, (2) Are you blind or visually handicapped?, (3) Do you pay for school-related expenses?, or (4) Are you poor? If the answer was “yes” to any of these questions, then government aid was distributed. Nowhere in these cases is the threshold question for receipt of government benefits: Are you Catholic, Jewish, Lutheran, Baptist, Mormon or Muslim.
¶ 79 By stark contrast, the Arizona corporate tax credit program authorizes STOs and qualified schools to first ask parents and students what religious sect they belong to before benefits are distributed. STOs are permitted to and allegedly do refuse to give a parent scholarship money if they do not subscribe to a certain religious belief. This is not neutral with respect to religion and does not disseminate the benefit without regard to religion.
¶ 80 The majority downplays this fact by stating that it is the STOs which discriminate, not the State of Arizona.22 Supra ¶ 27. However, while that might work in some instances for the private choice criteria, see infra III(C), that is insufficient for the first two criteria of the effects test. The State cannot do by indirection what it cannot do directly. Norwood v. Harrison, 413 U.S. 455, 462-65, 470, 93 S.Ct. 2804, 37 L.Ed.2d 723 (1973); Airport Prop. v. Maricopa County, 195 Ariz. 89, 98, 985 P.2d 574, 583 (quoting State v. Yuma Irrigation Dist., 55 Ariz. 178, 184, 99 P.2d 704, 706 (1940)). The sole puipose of STOs is to distribute tax credit money, the government permits the creation of STOs, and the government funnels funds directly to the STOs. E.g. http://www.acsto. org/ (“The Arizona Christian School Tuition Organization, Inc. (ACSTO) was incorporated in 1998 for the sole purpose of implementing the private school tuition tax credit law.”). If the State performed the same functions as the STOs, it would be unconstitutional. For example, if the State directed seventy percent of the tax credit funds to religious schools and if those schools only admitted students of one religion, then its effect would be the unconstitutional government advancement of religion. However, Arizona permits STOs to do just that. Thus, the government permits STOs by proxy to do indirectly what it is precluded from doing directly. STOs, as an integral part of the government program, convey a message of exclusion based on religious beliefs, and have the imprimatur of governmental approval. See Zobrest, 509 U.S. at 23, 113 S.Ct. 2462 (Blackmun, J., dissenting). This violates the Establishment Clause because it has the effect of advancing or inhibiting religion.
¶ 81 Nor, as argued by the majority, does the fact that private entities are permitted to discriminate mean that such discrimination is the result of parental choice in choosing the STOs to which they apply for funds. Supra ¶ 27. When such funds are limited and controlled by a small number of STOs that discriminate on the basis of religion, it is not the parental choice which furthers the discrimination, but the State by permitting the STOs to control the funds and discriminate by awarding funds on the basis of religion.23
*425¶ 82 Finally, it is the lack of any tax benefit to parents sending their children to public schools which further distinguishes this case from Zelman and makes it more comparable to Nyquist. In Nyquist, the Court struck down under the Establishment Clause direct aid for repair and maintenance of private schools (413 U.S. at 774-80, 93 S.Ct. 2993), tuition reimbursements (413 U.S. at 780-89, 93 S.Ct. 2993) and tax credits (413 U.S. at 789-94, 93 S.Ct. 2993) for parents who sent their children to private, mainly sectarian schools. In Zelman, the Court upheld the voucher program when it was used to allow parents to choose to send their children to private schools, to send their children to nonfailing public schools or to hire tutors for their children if they continued to attend Cleveland’s failing public schools. Zelman, 536 U.S. at 644-46, 122 S.Ct. 2460. The Court in Zelman explained that what distinguished Zelman from Nyquist was a number of factors, primarily that the benefits in Nyquist were “exclusively to private schools and the parents of private school enrollees ... [whose] ‘function’ was ‘unmistakably to provide desired financial support for nonpublic, sectarian institutions’ ” and “flatly prohibited the participation of any public school, or parent of any public school enrollee.” 536 U.S. at 661, 122 S.Ct. 2460 (quoting Nyquist, 413 U.S. at 783, 786, 93 S.Ct. 2955). The voucher program, the Court in Zelman held, shared none of these features. Id.24 Additionally, the Court in Zelman emphasized that Nyquist did not control the Cleveland voucher program because Nyquist was expressly limited to programs in which parents sent their children to private schools. Id. Nyquist reserved judgment for eases involving aid to persons sending their children to both public and private school, the question presented in Zelman. Id. (quoting Nyquist, 413 U.S. at 782-83 n. 38, 93 S.Ct. 2993).
¶ 83 The tax program here is closer to that of Nyquist than Zelman. The benefits are not available to any parent, but only to parents who want to send their children to private schools. Parents who desire to send their children to public schools are deprived of any assistance to meet school expenses not paid for by public schools. Moreover, some, if not many, parents who desire to send their children to private schools may and will be excluded from assistance because allegedly the vast majority of corporate-funded scholarships will only be available to persons holding the same religious beliefs as the STOs or schools participating in the program. Since the tax credit program is not equally available to parents with children attending public as well as private schools, in contrast to Zelman’s voucher program, it violates the Establishment Clause under Nyquist and is not protected by Zelman. While, unlike Nyquist, the private schools may not yet be failing, the benefits here are “exclusively to ... the parents of private school enrollees ... [so the program’s] ‘function’ [is] ‘[t]o provide desired financial support for nonpublic, sectarian institutions.’ ” Zelman, 536 U.S. at 661, 122 S.Ct. 2460 (quoting Nyquist, 413 U.S. at 783, 786, 93 S.Ct. 2955).25
C. Private Choice
¶ 84 The majority also holds that the Arizona program is one of true private choice because there are two layers of individual choices involved in which schools receive government aid. Supra ¶¶ 21-22. Specifically, the majority relies on the corporate taxpayer’s choice to contribute its credits to STOs and the parent’s choice of where to use the *426scholarship money. While I agree that the program meets the third prong of the Zel-man test, private choice, it does so not because of these separate layers, but rather because it must be viewed in the context of all other choices available to paz-ezzts and children.
¶ 85 Zelman requires that courts determine true piivate choice by evaluating the universe of school options available to parents in the State and then, based on those options, couz’ts must determine whether parents are being coerced into choosing religious schools.26 536 U.S. at 655-56, 122 S.Ct. 2460 (“The Establishment Clause question is whether Ohio is coercing parents into sending their children to religious schools, and that question must be answered by evaluating all options Ohio provides Cleveland schoolchildren, only one of which is to obtain a program scholarship and then choose a i’eligious school”). In addition, Zelman looked at vai'ious other factors to determine true private choice, including whether the program: (1) was part of the State’s multifaceted attempt to improve educational opportunities, (2) allowed all schools within the district to participate, (3) only gave preference to low-income families without allowing other preferences, and (4) did not provide financial incentives to skew the program toward religious schools. 536 U.S. at 653, 122 S.Ct. 2460.
¶ 86 In contrast, the majority’s analysis measures choice based on the options created by the corporate taxpayer program, not the totality of school options. Under that analysis, the majority fails to recognize that the corporate tax credit program negates or reduces private choice for a number of reasons. First, Zelman and its progeny are primarily concerned with parental choice, not the choice of some other entities like corporations.27 Second, parental choice is severely limited because the STOs are in complete control over which schools are able to participate. Third, the corporate tax credit program primarily incitases the availability of religious schools to parents who subscribe to certain religious beliefs and prefer to send their children to schools of that religion. It does not increase options for all parents. Fourth, in reviewing the other Zelman factors, there is at least some evidence here that the tax credit program was not part of a multifaceted attempt by the State to improve educational opportunities. Supra ¶¶ 64-71.
¶ 87 Despite that failure in analysis, if we apply the Zelman test for the pi’ivate choice criterion, the scheme here meets that factor. In Zelman, true private choice is present when pai’ents have almost total control over where to send their child to school among a broad range of options. 536 U.S. at 664, 122 S.Ct. 2460. In that program, “[wjhere tuition aid is spent depends solely upon where parents who receive tuition aid choose to enroll their child.” Id. at 646, 122 S.Ct. 2460. Further, the Court describes a true private choice program as one “[i]n which government aid reaches religious schools only as a result of the genuine and independent choices of private individuals.” Id. at 649, 122 S.Ct. 2460 (citation omitted). The phrase “independent choices of private individuals” means choices made by parents or students. For example, in Zobrest, the Court stated, “By according parents freedom to select a school of their choice, the statute ensures that a government-paid interpreter will be present in a sectarian school only as a result of the pi’ivate decision of individual parents.” 509 U.S. at 10, 113 S.Ct. 2462. In Witters, “Lajid ... that ultimately flows to religious institutions does so.only as a result of the *427genuinely independent and private choices of aid recipients.” 474 U.S. at 488, 106 S.Ct. 748. The “aid recipients” in Witters were students. Finally, in Mueller, the Court found a valid program because of the “numerous, private choices of individual parents of school-age children.” 463 U.S. at 398, 103 S.Ct. 3062.
¶ 88 That requirement is met here because of the range of schools available to parents. Parents can choose to continue to send their children to public schools, including charter schools. Alternatively, parents can send them to any one of many private schools at their own expense or using tax credit funds to offset that expense.
IV. Conclusion
¶ 89 Zelmcm established that four requirements must be met for an indirect aid program to pass muster under the Establishment Clause. Here, there is a fact question whether the predominate intent was secular or was aimed at promoting religion. Additionally, based on the record presented, the program is neither entirely neutral with respect to religion nor does it provide aid to a broad class of individuals defined without reference to religion. Accordingly, I would reverse the judgment of the superior court and remand for further proceedings consistent with my conclusions.
. As to the general and uniform clause of the Arizona Constitution, the superior court relied on Arizona Supreme Court decisions other than Kottemían.
. To the extent that a factual reference in the amended complaint is supported by the language of § 43-1183, I reference the statutory provision in my dissent. To the extent the statute does not support the allegation, I rely solely on the statutory provision.
. The STOs must be charitable organizations exempt from federal taxation under 501(c)(3) of the United States Internal Revenue Code.
. According to a DOR document revising implementation of the scheme (“Revised Implementation"), the DOR will disapprove a proposed contribution to an STO only if it would exceed the aggregate statutory cap for annual tax credits under the scheme. Appellants have not moved to strike the Revised Implementation. Accordingly, we will take judicial notice of that fact. Jarvis v. State Land Dep't, 104 Ariz. 527, 530, 456 P.2d 385, 388 (1969) (citation omitted), modified by, 106 Ariz. 506, 479 P.2d 169 (1970) and by 113 Ariz. 230, 550 P.2d 227 (1976).
. However, since the statute requires STOs to be qualified under § 501(c)(3) of the Internal Revenue Code, STOs risk their tax exempt status under that code if they discriminate based on race, but they may still discriminate on the basis . of religion if they are affiliated with a religious denomination.
. Appellants allege that of those tax credits, in 2005 almost $22 million was donated to three STOs, two of them associated with the Catholic Church and one with a Christian academy. These allegations are supported by documents filed with the superior court in response to the motions to dismiss. According to those documents from the DOR, in fiscal year 2006, approximately $51 million in donations were made to 56 STOs and over $40 million in scholarship funds were distributed. Slightly over $28 million in donations were made to four STOs, three of which were connected with Catholic or Christian denominations. Those documents do not show, however, which STOs gave scholarship funds for students attending particular schools.
. The majority posits that Zelman upheld the Cleveland voucher program based on a number of inter-related factors and that the bar to private discrimination in the Cleveland voucher program was not dispositive. Supra n. 6. Since the Ohio program did not permit private discrimination by schools, it is impossible to tell what the Court would have done with a situation such as this, in which vast percentages of the tax credit funds are allegedly controlled by a few STOs which allegedly discriminate on the basis of religion and in fact limit true parental choice. Failure to meet any of the Establishment Clause factors would invalidate a program.
. Income Tax Credit Review: Joint Legis. Income Tax Credit Review Comm, at 18 (Dec. 7, 2006) available at http:Avww.azleg.gov/jlbc/ jlitcrcpptl70706 (last visited Jan. 23, 2009); Joint Legis. Budget Comm. Staff Memorandum at 9, 11 (Nov. 30, 2006) available at http:/www. azleg.gov/jlbc/jlitcrcpptl 70706 (last visited Jan. 23, 2009). Those savings are substantial. As indicated in the latter report, as of November 2006, the Joint Legislative Budget Committee estimated the tax revenues diverted to private schools exceeded $42 million annually.
. See Minutes: H.B.2074, S. Comm, on Educ. at 10 (March 24, 1997); Minutes: H.B.2074, H. Comm. on Educ. at 4 (Jan. 29, 1997).
. Arizona has one of the lowest levels of public school spending per student in the United States. Other sources indicate that Arizona is failing in its mission to educate its youth in public schools. Quality Counts 2009: Portrait of a Population, Education Week Jan. 9, 2009. Given the current fiscal crisis, the State recently proposed cutting over $1 billion from the education budget for fiscal years 2009 and 2010. FY 2009 and FY 2010 Budget Options: Appropriations Comm. 24 (Jan. 15, 2009) available at http:/azleg.gov/jlbc. him (report made by Comm. Chairman) (last visited Jan. 23, 2009).
. In Kotterman, the Arizona Supreme Court applied a more deferential test to determine the secular purpose of the individual tax credit program. 193 Ariz. at 278-79, ¶¶ 6-8, 972 P.2d at 611-12. We are not bound by that holding because it preceded McCreary County, which explained that we are not to apply the very deferential test for economic legislation in this context. 545 U.S. at 865 n. 13, 125 S.Ct. 2722. Additionally, when it decided Kotterman, the court did not have the additional facts related to the legislature's knowledge of the extent of diversion of funds under the two tax credit programs, which are available to us now, nor the current fiscal crisis with resulting proposed reduction of state funding to public education.
. The majority also argues that the individual school tax credit program has the same features as the corporate lax credit program and was upheld in Kotterman. Supra ¶ 27. However, it is the later Zelman analysis which controls here. Supra ¶ 49.
. The majority points to no case on which Zel-man relied for its conclusions that permitted schools or other indirect beneficiaries of state largess to discriminate on the basis of religious belief. The only such case on which Zelman relied as a direct predecessor in which schools might discriminate on the basis of religion is Mueller. In Mueller, the Court upheld tax deductions for educational expenses of parents who chose to send their children to public or private, secular or nonsecular, schools even though apparently the nonsecular schools could discriminate in admission of students on the basis of religious belief without violating stale law. 463 U.S. at 390 and n. 1, 103 S.Ct. 3062. The Court did not address that factor in Mueller. As Mueller explained, however, the tax program upheld there applied to all students, public and private, not to just private school students, thus distinguishing Mueller from Nyquist. Id. at 397-98, 103 S.Ct. 3062. Moreover, Mueller also pointed out that there the tax deductions were granted directly to parents who chose the schools their children attended. Id. at 399-400, 103 S.Ct. *4253062. Here, the discrimination occurs not by the parents seeking the scholarships, but by the STOs which limit the choices of parents, the presumed ultimate beneficiaries of the funds, based solely on their religious beliefs.
. Included in those features were that the tax benefits were unrelated to any amount of money actually expended by any parent on tuition and that tuition reimbursements offered an incentive to send children to sectarian schools. Zelman, 536 U.S. at 661, 122 S.Ct. 2460.
. Interestingly, the Court in Nyquist found that despite the program’s effects, its purpose was still secular because it allegedly sought to increase school choice and avoid overburdening public schools with students who would otherwise go to private schools. 413 U.S. at 763-67, 773-74, 783, 791, 93 S.Ct. 2993. While the alleged purpose here is secular, although subject to some doubt, supra 111(A), the effect is to encourage parents to send their children to private schools by providing scholarships, mainly to sectarian schools.
. In dissent, Justice Souter described the majorily’s definition of "choice" in Zelman as “diluted.” 536 U.S. at 702, 122 S.Ct. 2460 (Souter, J., dissenting). He wrote that “if the choice of relevant alternatives is an open one, proponents of voucher aid will always win, because they will always be able to find a ‘choice’ somewhere that will show the bulk of public spending to be secular." Id. at 701, 122 S.Ct. 2460. The Arizona system underscores the accuracy of that analysis. Nevertheless, we are bound by the majority’s opinion in Zelman.
. In Mueller, the program allowed any parent to take a tax deduction for school-related expenses at any elementary or secondary school in Minnesota. 463 U.S. at 391, 103 S.Ct. 3062. In Witters, the voucher program allowed visually-impaired students to choose any school which offered vocational training in Washington. 474 U.S. at 483, 106 S.Ct. 748. In Zobrest, the program allowed hearing-impaired students to use a government funded sign-language interpreter at the school of their choice. 509 U.S. at 10, 113 S.Ct. 2462.