*241 The opinion of the court was delivered by

Abbott, J.:

The plaintiff challenged the tax rate schedule found in the Kansas Income Tax Act (KITA), K.S.A. 79-3201 et seq., to the extent that tax rates imposed on single taxpayers are higher than the highest tax rate imposed on married taxpayers fifing jointly. The Shawnee County District Court found that such tax rates under KITA were an unconstitutional violation of equal protection. The trial court also ruled that it had subject matter jurisdiction to hear the plaintiff’s claim for a tax refund, even though the plaintiff had not exhausted all of his administrative remedies to receive a refund. Finally, the trial court certified a plaintiff class of single taxpayers who were subject to tax rates higher than the highest tax rate imposed on married taxpayers filing jointly. The State appeals all three of these rulings to this court.

Plaintiff’s narrow challenge is directed only to the Kansas tax rates applied to unmarried taxpayers which exceed the highest income tax rate charged to married individuals fifing under the status of “married fifing joint.” Stated in terms more understandable to the unmarried taxpayers of this state, this appeal does not involve any taxpayer having a taxable income of $20,000 or less, nor does it involve the first $20,000 of taxable income of any unmarried taxpayer having a taxable income in excess of $20,000 under the 1992-1996 tax rate schedules. (Prior to 1992, the tax rate schedules had different rates and tax brackets and had an overall smaller disparity between married and unmarried taxpayers.) Further, the largest current rate disparity at issue in this appeal between unmarried taxpayers and married taxpayers fifing jointly is a 1.3% difference in tax rates. The trial judge only held part of KITA unconstitutional — that part which taxes an unmarried person’s taxable income at a tax rate in excess of the highest income tax rate applicable to married taxpayers fifing a joint return. Currently, the highest rate for a married person fifing a joint return is 6.45%, while an unmarried taxpayer is subject to a 7.5% tax rate if the taxpayer earns more than $20,000 in taxable income and is subject to a 7.75% tax rate if the taxpayer earns more than $30,000 in taxable income. If we were to affirm the trial court’s holding that KITA *242 tax rates áre unconstitutional in part, KITA’s savings clauses (K.S.A. 79-3239 and K.S.A. 79-32,108) would leave the remainder of the act intact. Thus, a single taxpayer earning $20,000 or less in taxable income would continue to be taxed at a 4.4% rate, while a married taxpayer filing jointly who earns $30,000 or less would continue to be taxed at a lower rate of 3.5%.

A summary of the Kansas income tax system is helpful in analyzing this case. Under KITA, each taxpayer in Kansas must pay a certain percentage of his or her taxable income in Kansas income tax. A taxpayer’s taxable income is determined by calculating the taxpayer’s Kansas adjusted gross income (using the federal adjusted gross income as a base) and subtracting Kansas deductions and personal exemptions from the gross income. The percentage of the taxpayer’s taxable income that should be paid as income tax is equivalent to the tax rate applicable to the taxpayer. In Kansas, the tax rate applicable to each taxpayer is dependent upon the amount of taxable income the taxpayer earns and whether the taxpayer is married and filing jointly or filing a separate return as a single person.

The Kansas Legislature first created the tax rate disparity subject to challenge in this case in 1988. At that time, the Kansas Legislature made extensive changes to KITA by replacing the then existing eight-bracket tax rate schedule with a two-bracket tax rate schedule. This new 1988 tax rate schedule imposed higher rates on individual, single tax filers than the tax rates imposed on married taxpayers filing jointly. The 1988 tax rates (L. 1988, ch. 381, § 2) provided:

(1) Married individuals filing joint returns.

If the taxable income is:

Not over $35,000

Over $35,000

The tax is:

4.5% of Kansas taxable income

$1,418 plus 5.3% of excess over $35,000

(2) All other individuals.

If the taxable income is:

Not over $27,000

Over $27,500

The tax is:

4.8% of Kansas taxable income

$1,320 plus 6.1% of excess over $27,500

For the tax years 1989,1990, and 1991, approximately the same *243 disparity in tax rates applied, but lower tax rates were imposed overall (see K.S.A. 79-32,110[a]):

(1) Married individuals filing joint returns.

If the taxable income is:

Not over $35,000

Over $35,000

The tax is:

3.65% of Kansas taxable income

$1,278 plus 5.15% of excess over $35,000

(2) All other individuals.

If the taxable income is:

Not over $27,000

Over $27,500

The tax is:

4.5% of Kansas taxable income

$1,238 plus 5.95% of excess over $27,500

In 1992, the Kansas Legislature revised the tax rate schedule to create three different tax rates, rather than two, thereby widening the tax rate disparity between married taxpayers filing jointly and single taxpayers. The 1992 rate schedule remains unchanged and is used today. See K.S.A. 1995 Supp. 79-32,110(a). It provides:

(1) Married individuals filing joint returns.

If the taxable income is:

Not over $30,000

Over $30,000

but not over $60,000

Over $60,000

The tax is:

3.5% of Kansas taxable income

$1,050 plus 6.25% of excess over $30,000

$2,925 plus 6.45% of excess over $60,000

(2) All other individuals.

If the taxable income is:

Not over $20,000

Over $20,000

but not over $30,000

Over $30,000

The tax is:

4.4% of Kansas taxable income

$880 plus 7.5% of excess over $20,000

$1,630 plus 7.75% of excess over $30,000

On April 14,1993, Peden filed a petition in the Shawnee County District Court against the State of Kansas, the Kansas Department of Revenue, the Secretary of Revenue of the State of Kansas, Nancy Parrish, and the Director of Taxation of the State of Kansas, Alisa Dotson. (All defendants are hereinafter referred to as the State.) In this action, Peden sought four things. First, Peden requested that the trial court certify a plaintiff class consisting of all single taxpayers in Kansas who paid taxes under the single tax rates which were higher than the highest tax rate imposed on married taxpayers filing jointly. Second, Peden requested that the trial court *244 declare the single tax rates found in K.S.A. 79-32,110(a) unconstitutional as a violation of equal protection under Article 11, § 2 of the Kansas Constitution and under the 14th Amendment to the United States Constitution to the extent that the single tax rates were higher than the highest tax rate imposed on married taxpayers filing jointly. Third, Peden asked the trial court to enjoin the State from collecting taxes from single taxpayers at an unconstitutional rate in excess of the highest rate imposed on married taxpayers filing jointly. Fourth and finally, Peden sought an accounting of all improperly collected income taxes based on the discriminatory tax rates from 1988 to the present and a refund of such taxes paid with interest.

Peden filed a motion specifically asking the trial court to certify this case as a class action and find that Peden was a proper class representative. The trial court ruled that there was no compelling need for class certification in this case because there was “nothing that [could] be accomplished by class certification that [could not] be handled by ordinary litigation procedures.” The court denied Peden’s motion and refused to certify a plaintiff class.

On June 21, 1993, the State filed a motion to dismiss Peden’s tax refund claim for lack of subject matter jurisdiction because Peden had not yet exhausted all of his administrative remedies. On January 5, 1994, the trial court ruled on this motion. The court found that Peden had not yet exhausted his administrative remedies in regard to the refund claim; thus, the court held that it did not have subject matter jurisdiction to hear the refund claim. The court granted the State’s motion and dismissed the refund claim for lack of subject matter jurisdiction. As such, only two claims remained in Peden’s petition — (1) an individual request that the court declare the disparate tax rates an unconstitutional violation of equal protection and (2) an individual request that the court enjoin the State from collecting future income taxes from Peden at a rate in excess of the highest rate imposed on married taxpayers filing jointly.

Based on these two claims, Peden filed a motion for summary judgment on March 15, 1994. The State also filed a motion for summary judgment on April 25, 1994. In analyzing the constitu *245 tionality of the tax rates, the trial court applied the rational basis test — asking whether there was a valid legislative purpose for KITA’s different treatment of similarly situated taxpayers and whether the disparate tax rates were rationally related to this valid legislative purpose. In applying this test, the trial court stated:

“There has been no rational basis for the [tax rate] distinction articulated by the Kansas Legislature, the Kansas Courts, or the executive agencies administering the Kansas tax law. In the Court’s view the weakness of the Defendant’s attempt to rationalize this disparity is exhibited by the lengths to which the Defendant goes to postulate some reasonable explanation for the obvious discrimination. Even recognizing that the Defendant State is allowed a great deal of latitude, the Court finds no rational basis for the discrimination. The fact that supposedly married couples have a greater economic burden is highly speculative and subject to a great deal of question and uncertainty. That the public policy of the State of Kansas promotes marriage through its tax structure is almost fanciful when analytically examined. The same can be said of the proposition that married individuals are less likely to relocate and will remain a stable economic unit for the State for longer periods of time. This leaves ease of administration’ and again this Court finds that to be speculative and conjectural. None of the reasons, therefore, advanced by the State would seem to justify the gross disparity and discrimination that exists in the tax rate structure between single individuals and married individuals.”

Based on this analysis, the trial court found the tax rates imposed on singles which were higher than the highest tax rate imposed on married taxpayers filing jointly were an unconstitutional violation of equal protection under the United States and Kansas Constitutions. The trial court granted Peden’s motion for summary judgment for declaratory relief. The court held that injunctive relief would be unnecessary and therefore inappropriate.

On November 7, 1994, the State timely filed a notice of appeal to the Kansas Supreme Court, challenging the trial court’s ruling that the tax rate schedule in KITA was unconstitutional in part. However, after the trial court filed its journal entry, Peden timely filed post-judgment motions on November 14, 1994, asking the trial court, inter alia, to reconsider its prior ruling which dismissed the tax refund claim for lack of subject matter jurisdiction and to reconsider its prior ruling which refused to certify a plaintiff class of single taxpayers.

*246 The trial court declined to rule on Peden’s post-judgment motions until after the State’s appeal on the constitutionality of the tax rates had been decided. Peden filed a motion to this court, asking it to dismiss the State’s appeal. Finding that the trial court had not yet ruled on Peden’s post-judgment motions, this court held that State’s appeal was premature and dismissed the State’s appeal.

On August 22, 1995, the trial court entered a memorandum decision and order regarding Peden’s post-judgment motions. In regard to the refund issue, the trial court found that Peden sought relief from unconstitutional tax rates. Since an administrative agency cannot determine the constitutionality of a statute, the court held that no purpose would have been served by requiring Peden to exhaust his administrative remedies. The court ruled that the exhaustion doctrine should not apply in this case. As such, the court found that Peden’s apparent lack of exhaustion of administrative remedies should not prevent the court from exercising subject matter jurisdiction over the refund claim. Thus, the court reconsidered its prior ruling, which dismissed the refund claim for lack of subject matter jurisdiction, and held that it could and would exercise jurisdiction over Peden’s refund claim. Next, the court reconsidered its ruling in which it refused to certify a plaintiff class of single taxpayers. The court found that all the requirements needed to certify a class under K.S.A. 60-223(a) and (b)(3) were met. The court certified a plaintiff class and found Peden to be a proper class representative. The court defined the class as follows:

“All unmarried persons subject to Kansas state income taxation, at a rate in excess of the highest rate charged to married individuals filing joint returns, for one or more of the tax years 1988 through the present.”

Thus, on August 22, 1995, the court granted Peden’s motions for reconsideration of subject matter jurisdiction over the refund claim and for reconsideration of class certification. Upon certification of the class, Peden filed a motion to certify Sharon L. Smith as a plaintiff and as another class representative. Apparently, the trial court granted this motion, although the record before us does not so show.

*247 On September 20,1995, the State filed a timely notice of appeal with this court, challenging these three rulings — the trial court’s holding that the tax rates under KITA are an unconstitutional violation of equal protection to the extent that single taxpayers are subject to tax rates which are higher than the highest tax rate imposed on married taxpayers filing jointly; the trial court’s ruling that it had subject matter jurisdiction to decide the tax refund claim; and the trial court’s certification of the plaintiff class.

Since the trial court ruled that part of the Kansas tax rate schedule was unconstitutional, approximately 5,000 single taxpayers have filed claims with the Director of Taxation requesting tax refunds. All such requests for hearings are being held in abeyance by an administrative law judge pending resolution of this case.

The legislative history behind the rate disparity is slim to none. In a few of the committee meetings in 1988, when the rate disparity was first promulgated, and in 1992, when the rate disparity was widened, a few' legislators expressed concern about the fact that single taxpayers were being taxed at higher rates than married taxpayers filing jointly. However, no rationale or explanation for such disparity was ever provided. The legislature passed KITA without any further discussion about the tax rate disparity being recorded. See Minutes of the Senate Committee on Assessment and Taxation, February 3, 1988; February 19, 1988; March 31, 1992; Minutes of the House Committee on Taxation, March 6, 1992.

In its brief, the State argues that the federal income tax system taxes single taxpayers and married taxpayers filing jointly differently, as Kansas does. The State points out that this unequal treatment of taxpayers has not been found to be a constitutional violation of equal protection. Druker v. C.I.R., 697 F.2d 46 (2d Cir. 1982), cert. denied 461 U.S. 957 (1983) (upholding marriage penalty); Ma pes v. United States, 576 F.2d 896 (Cl. Ct.), cert. denied 439 U.S. 1046 (1978) (upholding marriage penalty); Kellems v. Commissioner of Internal Revenue, 58 T.C. 556 (1972), aff’d 474 F.2d 1399 (2d Cir.), cert. denied 414 U.S. 831 (1973) (upholding single penalty). Thus, the State asserts that the similar tax rate disparity in Kansas does not violate equal protection, either.

*248 Peden acknowledges that the federal income tax system does treat single taxpayers differently from married taxpayers filing jointly and that such differential treatment has been found to be constitutional. However, Peden asserts that the manner in which the federal tax code treats single and married taxpayers differently is not the same manner in which Kansas tax code treats single and married taxpayers differently. Thus, Peden reasons that just because the federal tax system is constitutional, this does not mean that the state tax system is constitutional. Peden contends that the two systems discriminate between single and married taxpayers in different ways and for different reasons. In fact, Peden argues that the federal tax code has never charged single taxpayers a higher income tax rate than the highest rate applicable to married taxpayers filing jointly. According to Peden and our research on the issue, Kansas is the only state in the nation that has ever imposed higher tax rates on single taxpayers than the highest rate imposed on married taxpayers filing jointly. See Minutes of Senate Committee on Assessment and Taxation, February 19, 1988.

Peden then goes on to explain how the federal income tax system treats single and married taxpayers differently and how this system differs from the Kansas income tax system. As Peden explains, the federal income tax system treats married and single taxpayers differently by setting up different tax brackets, not rates, for married taxpayers and single taxpayers. Peden gives the following example of the 1995 federal income tax brackets.

Rate Single Individuals Bracket Married Filing Joint Bracket

15% Under $23,350 Under $39,000

28% $23,350 — 56,550 $39,000 — 94,250

31% $56,550 — 117,950 $94,250 — 143,600

36% $117,950 — 256,500 $143,600 — 256,500

39.6% Over $256,500 Over $256,500

1995 Form 1040 Instructions (Department of the Treasury), p. 53. (The 1995 federal income tax schedule also provided brackets for married taxpayers filing separately and heads of household. These are not applicable here.)

Under these brackets, different rates may apply to taxpayers who earn the same amount of taxable income, depending upon whether *249 the taxpayer files a separate return as a single taxpayer or files a joint return as a married taxpayer. A single taxpayer with $25,000 in taxable income is taxed at the 28% rate, while married taxpayers filing jointly with $25,000 in taxable income are taxed at a lower rate of 15%. However, as Peden points out, once married taxpayers filing jointly and single taxpayers earn a certain amount of taxable income, the tax rates even out. For instance, once a single taxpayer earns $117,950 in taxable income, the taxpayer will be taxed at a 36% rate, and once married taxpayers filing jointly earn $143,600 in taxable income, they also are taxed at a 36% rate. Once single taxpayers or married taxpayers filing jointly earn $256,500 in taxable income, both types of taxpayers enter a new bracket at this same income level and are taxed at the same rate — 39.6%. Both types of taxpayers will continue to be taxed at this same rate no matter how much more money they earn. However, in 1993, only about .9% of all federal taxpayers fell into the 39.6% tax bracket and were taxed equally regardless of their marital status. Internal Revenue Service, Statistics of Income-1993; Individual Income Tax Returns.

Thus, under the federal law, tax rates for both married taxpayers filing jointly and single taxpayers are the same, and have always been the same, with respect to the lowest and highest rates. 26 U.S.C. § 1 (1994). (The current rates for both types of taxpayers start at 15% and end at 39.6%.) While there are different brackets of taxable income to which the same rates apply, 39.6% is the highest rate to which any individual may be subject under the federal tax system — whether married or unmarried. This is not the case under the Kansas income tax system, which applies different rates to approximately the same amount of income. Under the Kansas system, the top rates never even out, no matter how much money the married taxpayers filing jointly earn. The single taxpayer wifi always be taxed at a higher rate. In 1995, if a single taxpayer earned $30,001 in taxable income, he or she was taxed at a 7.75% rate, while if married taxpayers filing jointly earned $120,000 in taxable income (four times the single taxpayer), they were still taxed at the lower rate of 6.45% and always would be.

*250 Thus, just because the federal tax system, which treats single and married taxpayers differently, is constitutional, Peden contends that this is not the case with the Kansas tax system and its treatment of single and married taxpayers. According to Peden, the Kansas tax system distinguishes between single and married taxpayers in a manner which is different from how the federal tax system distinguishes between single and married taxpayers. Further, Peden points out that the Kansas tax system does not have the same legislative history and rationale as the federal tax system. The federal tax system has a clear rationale, enumerated in the legislative history, which clarifies that the federal tax brackets treat single and married taxpayers differently in order to create geographic equality among married taxpayers in community property states and non-community property states. The Kansas tax system has no such rationale or legislative history. Thus, Peden asks this court to evaluate the constitutionality of the Kansas tax rates without regard to the constitutionality of the federal tax system.

Determining whether a statute unconstitutionally violates equal protection is a question of law. When determining a question of law, this court may exercise an unlimited de novo standard of review. See Dickerson v. Kansas Dept. of Revenue, 253 Kan. 843, 844, 863 P.2d 364 (1993); State v. Nelson, 249 Kan. 689, 692, 882 P.2d 53 (1991). This court is not bound by the decision of the trial court. Memorial Hospital Ass’n, Inc. v. Knutson, 239 Kan. 663, 668, 722 P.2d 1093 (1986).

“[A] statute is presumed constitutional and all doubts must be resolved in favor of its validity. If there is any reasonable way to construe a statute as constitutionally valid, the court must do so. A statute must clearly violate the constitution before it may be struck down.” Boatright v. Kansas Racing Comm’n, 251 Kan. 240, 243, 834 P.2d 368 (1992); see State v. Scherzer, 254 Kan. 926, Syl. ¶ 6, 869 P.2d 729 (1994). “This court not only has the authority, but also the duty, to construe a statute in such a manner that it is constitutional if the same can be done within the apparent intent of the legislature in passing the statute.” State v. Durrant, 244 Kan. 522, 534, 769 P.2d 1174, cert. denied 492 U.S. 923 (1989).

*251 A legislative act may not always be viewed by this court as fair or what we think should have been done, but that does not make the act unconstitutional. In Morton Salt Co. v. City of South Hutchinson, 159 F.2d 897, 900 (10th Cir. 1947), the 10th Circuit Court of Appeals commented:

“It is not the judicial province to correct every abuse of legislative taxing power. In the words of Mr. Chief Justice Marshall, The interest, wisdom, and justice of the representative body, and its relations with its constituents, furnish the only security, where there is no express contract, against unjust and excessive taxation; as well as against unwise legislation generally.’ Providence Bank v. Billings, 29 U.S. 514, 561, 7 L. Ed. 939; Dane v. Jackson, [256 U.S. 589, 65 L. Ed. 1107, 41 S. Ct. 566]; French v. Barber Asphalt Paving Co., 181 U.S. 324, 21 S. Ct. 625, 45 L. Ed. 879; Cooley, Sec. 67.”

For this appeal, the relevant principle of equal protection is embodied in § 1 of the Kansas Constitution Bill of Rights. This section provides:

“§ 1. Equal Rights. All men are possessed of equal and inalienable natural rights, among which are life, liberty, and the pursuit of happiness.”

The equal protection provision of the United States Constitution is found in the 14th Amendment, which provides:

“All persons bom or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States and of the State wherein they reside. No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws.”

The trial court found that KITA classifies single taxpayers differently than it classifies married taxpayers filing jointly and applies a higher tax rate on singles than the highest tax rate imposed on married taxpayers filing jointly. According to the trial court, KITA creates a disparity between two classes of citizens, single taxpayers and married taxpayers filing jointly, thereby violating the equal protection rights of single taxpayers.

Equal protection is implicated when a statute treats “arguably indistinguishable” classes of people differently. Smith v. Printup, 254 Kan. 315, 321, 866 P.2d 985 (1993), Single taxpayers and married taxpayers are arguably indistinguishable in a legal sense. Nei *252 ther party argues otherwise. Since KITA treats these two arguably indistinguishable groups differently in regard to the tax rates applicable to each group, KITA implicates equal protection.

Since KITA tax rates implicate equal protection, the court must determine which standard to use in evaluating the statute. The rational basis standard (sometimes referred to as the reasonable basis test) applies to laws which result in some economic inequality. The disparate tax rates applicable to single taxpayers and married taxpayers filing jointly create economic inequality. Thus, the tax rates in KITA are analyzed below under the rational basis standard. Under this standard, a law is constitutional, despite some unequal classification of citizens, if the “classification bears some reasonable relationship to a valid legislative objective.” Farley v. Engelken, 241 Kan. 663, Syl. ¶ 3, 740 P.2d 1058 (1987). Leiker v. Gafford, 245 Kan. 325, 363-64, 778 P.2d 823 (1989), explains the “reasonable basis” test as follows:

“Tbe ‘reasonable basis’ test is violated only if the statutory classification rests on grounds wholly irrelevant to the achievement of the State’s legitimate objective. The state legislature is presumed to have acted within its constitutional power, even if the statute results in some inequality. Under the reasonable basis test, a statutory discrimination will not be set aside if any state of facts reasonably may be conceived to justify it.”

Thus, under the rational basis standard, if the State can show that the different treatment between single taxpayers and married taxpayers filing jointly is rationally related to a valid legislative interest, then the single tax rates under KITA, which are higher than the highest tax rate imposed on married taxpayers filing jointly, do not violate equal protection. See Leiker, 245 Kan. at 364; State ex rel. Schneider v. Liggett, 223 Kan. 610, 620, 576 P.2d 221 (1978). While both parties agree that the rational basis standard should be utilized in analyzing KITA, the parties disagree over the manner in which the standard should be used.

Relying on the presumption that statutes are constitutional, the State contends that KITA should be upheld unless the tax rate disparity is so arbitrary and irrational that it lacks any conceivable rational basis. Thus, the State points out that it only needs to offer “any state of facts [which] reasonably may be conceived to justify” *253 the tax rate disparity in order to prevail. McGowan v. Maryland, 366 U.S. 420, 426, 6 L. Ed. 2d 393, 81 S. Ct. 1101 (1961). On the other hand, Peden can only prevail under the rational basis standard, according to the State, if he can prove that no possible state of facts supports a rational basis for the classification. As the State points out, a plaintiff asserting the unconstitutionality of a statute under the rational basis standard “ha[s] the burden To negative every conceivable basis which might support [the classification]/ ” FCC v. Beach Communications, Inc., 508 U.S. 307, 315, 124 L. Ed. 2d 211, 113 S. Ct. 2096 (1993); Madden v. Kentucky, 309 U.S. 83, 88, 84 L. Ed. 590, 60 S. Ct. 406 (1940). Further, the State asserts, and we agree, that “in taxation, even more than in other fields, legislatures possess the greatest freedom in classification.” Madden, 309 U.S. at 88.

The State acknowledges the absence of legislative history concerning the disparate tax rates. However, the State contends that this is irrelevant. According to the State, the legislature is not required to provide legislative history explaining the reasons behind, a statute, and even if the legislature does provide such legislative history, the rational basis inquiry is not limited to those stated reasons. The State points to Bair v. Peck, 248 Kan. 824, 834, 811 P.2d 1176 (1991), in which this court stated: “ ‘Under the reasonable basis test, it is unnecessary to ascertain the specific purpose the Kansas Legislature espoused, if any, in establishing the challenged classification. Rather, if any state of facts reasonably may be conceived to justify the alleged statutory discrimination, the statute will not be set aside as a violation of equal protection/ ” (Quoting McGowan, 366 U.S. at 426.) (Emphasis added.)

Thus, under the rational basis standard, the State asserts that all it is required to do to withstand an equal protection challenge is show “that a [valid] purpose may conceivably or ‘may reasonably have been the purpose and policy’ of the relevant governmental decisionmaker” in promulgating the tax classifications. Nordlinger v. Hahn, 505 U.S. 1, 15, 120 L. Ed. 2d 1, 112 S. Ct. 2326 (1992). (Emphasis added.) The State points out that because a legislature is not required to articulate reasons for enacting a statute, “it is entirely irrelevant for constitutional purposes whether the con *254 ceived reason for the challenged distinction actually motivated the legislature.” Beach Communications, Inc., 508 U.S. at 315.

Under this standard, the State argues it can conceive of four valid state interests which rationally justify the disparate tax rate classifications. These state interests are: the encouragement of marriage, an alleviation of the increased financial burdens associated with marriage, an alleviation of the “marriage penalty” under the federal income tax law, and the encouragement of taxpayers who can file one joint return to do so.

Peden takes issue with the State’s lenient interpretation of the rational basis standard and with the State’s conceived justifications for the tax rate disparity. First, Peden asserts that the rational basis standard “ 'is not a toothless one’ ” as stated in Stephenson v. Sugar Creek Packing, 250 Kan. 768, 775, 830 P.2d 41 (1992) (quoting Mathews v. De Castro, 429 U.S. 181, 185, 50 L. Ed. 2d 308, 97 S. Ct. 431 (1976). “[E]ven the standard of rationality . . . must find some footing in the realities of the subject addressed by the legislation.” Heller v. Doe, 509 U.S. 312, 321, 125 L. Ed. 2d 257, 113 S. Ct. 2637 (1993). Peden then cites to several tax schemes which have classified taxpayers on baseless and irrational grounds and have been struck down under Kansas and federal constitutional law. Allegheny Pittsburgh Coal v. Webster County, 488 U.S. 336, 102 L. Ed. 2d 688, 109 S. Ct. 633 (1989) (real property tax assessment rates, based on recent purchase prices which were 8 to 35 times those applied to comparable neighboring properties, were held to violate equal protection); Metropolitan Life Ins. Co. v. Ward, 470 U.S. 869, 84 L. Ed. 2d 751, 105 S. Ct. 1676, reh. denied 471 U.S. 1120 (1985) (state statute taxing out-of-state insurance companies at a higher rate than domestic insurance companies held not to have legitimate state purpose, thereby violating the equal protection clause); Zobel v. Williams, 457 U.S. 55, 72 L. Ed. 2d 672, 102 S. Ct. 2309 (1982) (Alaska statute distributing income derived from state’s natural resources to state’s citizens in varying amounts, based on the length of each citizen’s residency, held to violate equal protection clause); and Hartman v. State Commission of Revenue and Taxation, 164 Kan. 67, 187 P.2d 939 (1947) (An income tax statute which imposed a different tax rate upon gains *255 from the disposal of stock, depending upon whether the stockholder sold the stock or had his other share of the corporate assets distributed to him or her upon liquidation of the corporation, was found to violate equal protection because the different tax rates were arbitrary classifications that had no reasonable relation to the subject matter.).

Peden also takes issue with the purported justifications the State contends rationally support the different tax rates. According to Peden, just because the State has articulated some possible basis for the Kansas tax rate disparities, this does not mean that KITA automatically becomes constitutional. Peden contends that KITA is unconstitutional and does not survive an equal protection challenge because there is not an adequate nexus or “fit” between the disparate tax rates and any valid legislative objective the State is trying to achieve through the rates. As an example of this nexus problem, Peden cites to Romer v. Evans, 517 U.S. 620, 134 L. Ed. 2d 855, 116 S. Ct. 1620 (1996).

In Romer, the plaintiff challenged a Colorado state constitutional amendment, which effectively repealed the state and local provisions that barred discrimination on the basis of sexual orientation, as a violation of equal protection under a rational basis standard. The State offered two legislative interests as rationally justifying the constitutional amendment — (1) respecting other citizens’ freedom to associate, such as employers or landlords who have personal or religious objections to homosexuality, and (2) conserving resources to fight discrimination against other groups. The Court found that there was not a close enough nexus between the amendment and the purported valid state interests justifying the amendment. Thus, the Court held that the amendment violated equal protection under the rational basis standard. In so holding, the court stated:

“The breadth of the Amendment is so far removed from these particular justifications that we find it impossible to credit them. We cannot say that Amendment 2 is directed to any identifiable legitimate purpose or discrete objective. It is a status-based enactment divorced from any factual context from which we could discern a relationship to legitimate state interests; it is a classification of persons undertaken for its own sake, something the Equal Protection Clause does not *256 permit. ‘[C]lass legislation . . . [is] obnoxious to the prohibitions of the Fourteenth Amendment . . . .’ Civil Rights Cases, 109 US, at 24, 27 L Ed 835, 3 S Ct 18.” 134 L. Ed. 2d at 868.

Peden argues that this same lack of nexus exists between KITA and the purported valid state interests which the State contend rationally justify KITA.

According to Peden, this lack of nexus between the tax rates and the purported state interests exists because the tax rate disparity applies to all levels of taxable income. For instance, married taxpayers with $100,000, $1,000,000 or $10,000,000 of taxable income will never be subject to a tax rate higher than 6.45%, while single taxpayers are subject to a tax rate of 7.5% for all amounts of taxable income in excess of $20,000 and a tax rate of 7.75% for all amounts of taxable income in excess of $30,000. Peden asserts that at some level of taxable income for married taxpayers, the differential in rates makes no sense and cannot be said to rationally further any legislative objective.

Peden contends that the tax rate disparity is not rationally related to and does not have a nexus with any of the four purported state interests because the tax rate disparity is not drawn with any precision. For instance, the disparate tax rate schedule does not provide lower rates to taxpayers who actually need the benefit of the state interest the rates are meant to foster or impose a higher rate on those taxpayers who do not need such benefits. According to Peden, “There must be some difference in character, condition, or situation, to justify distinction . . . ; otherwise, the classification is forced and unreal, and greater burdens are, in fact, imposed on some than on others of the same desert.” Henry v. Bauder, 213 Kan. 751, 753, 518 P.2d 362 (1974). As such, Peden asserts that the nexus or rational relation between the disparate tax rates and any purported legislative interests meant to be achieved through the rates is nonexistent and cannot support the tax rates under an equal protection challenge.

The State freely admits that not all married couples may suffer increased economic burdens as a married couple, or suffer a marriage penalty under the federal tax law, or be encouraged to get married and file a joint tax return due to lower tax rates for married *257 taxpayers filing jointly. As such, the State concedes that the lower tax rates for all married couples are overinclusive because not all married couples need the benefit of the legislative interests which the tax rates are intended to attain. Further, the State acknowledges that the tax rates may be underinclusive because single taxpayers might experience more economic burdens than some married taxpayers do, but single taxpayers do not receive a lower tax rate to relieve those burdens as married taxpayers do. Thus, the State admits that there is not a perfect nexus between lower tax rates for married taxpayers and the interests the tax rates are intended to achieve. Nevertheless, the State argues that under the rational basis standard, a close fit or mathematical perfection between a legislative goal and the means chosen to further that goal is not required. See New Orleans v. Dukes, 427 U.S. 297, 303, 49 L. Ed. 2d 511, 96 S. Ct. 2513 (1976). The State points to Metropolis Theatre Co. v. Chicago, 228 U.S. 61, 69-70, 57 L. Ed 730, 33 S. Ct. 441 (1913), which provides:

“The problems of government are practical ones and may justify, if they do not require, rough accommodations- — illogical, it may be, and unscientific. But even such criticism should not be hastily expressed. What is best is not always discernible; the wisdom of any choice may be disputed or condemned. Mere errors of government are not subject to our judicial review. It is only its palpably arbitrary exercises which can be declared void under the Fourteenth Amendment . . . .”

See also Heller, 509 U.S. at 321, which states:

“[Cjourts are compelled under rational-basis review to accept a legislature’s generalizations even when there is an imperfect fit between means and ends. A classification does not fail rational-basis review because it ‘is not made with mathematical nicety or because in practice it results in some inequality.’ [Citations omitted.]”

Thus, despite the inexact nexus or fit between lower tax rates for married taxpayers and the legislative objectives the rates are supposed to attain, the State contends that the tax rates do not violate equal protection. The State asserts that it has done what is necessary to defend against an equal protection challenge — it has raised a state of facts reasonably conceived to justify the tax rate disparity. In fact, the State claims it has raised four conceivable state of facts or valid state interests which rationally (not perfectly) *258 justify the tax rates. Thus, the State asserts that it has done enough and Peden’s extreme critique of the nexus or relationship between the rates and the interests they are intended to achieve is unnecessary under a rational basis standard, especially because the legislature possesses the greatest freedom of classification in the area of taxation. See Madden, 309 U.S. at 87-88.

We agree with the State. The rational basis standard is a very lenient standard. All the court must do to uphold a legislative classification under the rational basis standard is perceive any state of facts which rationally justifies the classification. Kellems v. Commissioner of Internal Revenue, 58 T.C. 556, 558 (1972), aff’d 474 F.2d 1399 (2d Cir.), cert. denied 414 U.S. 831 (1973). “Relevance is the only relationship required between the classification and the objective.” Stephenson, 250 Kan. at 774. See also Stephens v. Snyder Clinic Ass’n, 230 Kan. 115, 129, 631 P.2d 222 (1981) (stating that a classification which may result in some inequality only violates equal protection if the classification is “irrelevant” to the goals the State intended to achieve through passage of the statute). A classification is “relevant” to its intended goal if it is rationally related to the legitimate legislative purpose behind the statute. Thompson v. KFB Ins. Co., 252 Kan. 1010, 1018, 850 P.2d 773 (1993). However, a statute cannot classify persons into groups based on a criteria which is “wholly unrelated” to the goal of the statute. Henry, 213 Kan. at 753-54. A classification “ ‘must be reasonable, not arbitrary, and must rest upon some ground of difference having a fair and substantial relation to the object of the legislation, so that all persons similarly circumstanced shall be treated alike.’ ” Thompson, 252 Kan. at 1018 (quoting Royster Guano Co. v. Virginia, 253 U.S. 412, 415, 64 L. Ed. 989, 40 S. Ct. 560 [1920]).

Although the rational basis standard requires that the discriminatory classification (the tax rate disparity between single and married taxpayers) be rationally related to valid state interests or goals, the standard does not require that the classification be the perfect solution to achieve such goals. See Thompson, 252 Kan. at 1021 (When the legislature must draw a line and “ ‘there is no mathematical or logical way of fixing it precisely, the decision of the legislature must be accepted unless [the court] can say that it was *259 very wide of any reasonable mark.’ ”); Liggett, 223 Kan. at 619 (“Establishment of classifications with mathematical precision is not required.”). In trying to achieve certain goals, a legislature is affected by several factors which may prevent it from reaching the perfect solution. For instance, the tax rate classifications which may have been the perfect tax rate schedule to exactly meet all of the State’s interests without any overinclusion or underinclusion may also have been impossible for the legislature to formulate and too difficult for the taxpayers to understand. Thus, a tax rate schedule which generally imposes lower tax' rates on all marriéd taxpayers filing jointly may achieve the State’s interests in a more realistic way even though it results in some overinclusion or underinclusion. Druker v. C.I.R., 697 F.2d 46, 50 (2d Cir. 1982) (“In the area of family taxation every legislative disposition is 'virtually fated to be both overinclusive and underinclusive when judged from one perspective or another.’ ”), cert. denied 461 U.S. 957 (1983). Compromise legislation, which is not made with mathematical perfection and results in some inequality, does not violate equal protection as long as the tax rate is rationally (not perfectly) related to the State’s valid interests.

The State has conceived four interests which it believes are legitimate and which it believes the disparate tax rates are rationally related to, even though a perfect nexus does not exist between the tax rates and the goals they are intended to achieve. As previously mentioned, these interests are: the encouragement of marriage, an alleviation of the increased financial burden placed on married taxpayers, an alleviation of the “marriage penalty” under the federal income tax law, and the encouragement of tax filers who can file one joint return to do so.

Under a rational basis standard, it is irrelevant whether these interests were enunciated in the legislative history of KITA or whether any of these interests actually motivated the legislature to pass the disparate tax rates. All that is necessary is that the court “perceives” one or more of the interests could reasonably and rationally justify the disparate tax rates. See Beach Communications, Inc., 508 U.S. at 315; see also Stephenson, 250 Kan. at 774 (“[T]he *260 legislature’s purpose in creating the classification need not be established.”). Kellems, 58 T.C. at 558.

Courts must defer to the legislative branch in the area of taxation because all taxation schemes carry some form of discriminatory impact. Johnson v. United States, 422 F. Supp. 958, 971 (N.D. Ind. 1976). “The process of evaluating specific tax proposals and of weighing their relative discriminatory impacts is primarily a legislative, not a judicial, function.” Johnson, 422 F. Supp. at 974. The burden is on Peden to overcome KITA’s presumption of constitutionality and to prove that the disparate tax rates are not rationally related to any of the four valid state interests the State puts forth. Stephen, 230 Kan. at 129.

First, the State argues that encouraging marriage is a legitimate State interest. Kansas recognizes a public policy in fostering, protecting, and encouraging marriage. See Hill v. Hill, 228 Kan. 680, 688, 620 P.2d 1114 (1980) (Schroeder, C.J., concurring.) See also Fincham v. Fincham, 160 Kan. 683, Syl. ¶ 1 and 2, 165 P.2d 209 (1946) (“Public policy relating to marriage is to foster and protect it, to make it a permanent and public institution, to encourage the parties to live together and to prevent separation.”). The State points to several statutes which promote Kansas’ public policy favoring marriage; K.S.A. 23-101, K.S.A. 23-104a, K.S.A. 23-116 (sanctioning marriage ceremonies); K.S.A. 60-428 (recognizing marital privilege to protect confidential marriage communications); K.S.A. 23-205 (allowing an injured person to pursue a loss of consortium claim on behalf of his or her spouse); K.S.A. 59-504 (providing for intestate succession of surviving spouses). None of these benefits are available to single individuals, even when they are living with a significant other whether of the opposite or same sex.

Finally, in an effort to prove that the encouragement of marriage is a legitimate state interest, the State cites to the following studies which indicate that children are better off being in a family with two parents who are married to each other and that adults are better off, both emotionally and physically, when they are married:

Couples who merely cohabit, rather than marry, are far less likely to remain together. See Stets, The Link Between Past and Present Intimate Relationships, 14 J. of Fam. Issues 236, 251 (1993). Marriage provides a stable, coherent, reg *261 ulated environment. Ross, Mirowsky & Goldsteen, The Impact of Family on Health: The Decade In Review, 52 J. of Marriage & Fam. 1059,1063 (1990). See Hughes & Gove, Living Alone, Social Integration, and Mental Health, 87 Am. J. of Soc. 48 (1981); Umberson, Family Status and Health Behaviors: Social Control as a Dimension of Social Integration, 28 J. of Health & Soc. Behav. 306 (1987). A stable environment is crucial to the developmental needs of children. Hafen, The Constitutional Status of Marriage, Kinship, and Sexual Privacy — Balancing the Individual and Social Interests, 81 Mich. L. Rev. 463,472 (1983). See Hughes & Gove, 87 Am. J. of Soc. at 50. Finally, scientific studies show that marriage is good for the physical, mental, and emotional well-being of the participants. As compared to unmarried persons, married persons exhibit significantly lower rates of alcoholism, suicide, schizophrenia, and other mental disorders. Coombs, Marital Status and Personal Well-Being: A Literature Review, 40 Fam. Rel. 97 (1991). Marriage is associated with lower rates of coronary heart disease, stroke, pneumonia, cirrhosis of the liver, automobile accidents, homicide, suicide, anxiety, and depression. Ross, Mirowsky & Goldsteen, 52 J. of Marriage & Fam. at 1061-62. Marriage also has positive and salutary effects on individuals’ recovery rates from cancer and other serious illnesses. See Coombs, 40 Fam. Rel. 97. The evidence is overwhelming: marriage is beneficial to individuals and society, on balance. Lichter, The Retreat from Marriage and the Rise in Nonmarital Fertility, Report to Congress on Out-of-Wedlock Childbearing 137,143 (Dep’t of Health & Hum. Serv. 1995). Children not in two-parent homes have two to three times more behavioral and psychological problems than do children from intact homes. Zill & Schoenbom, Developmental, Learning and Emotional Problems: Health of Our Nation’s Children, United States, 1988, Advance Data (Nat’l Center for Health Stat., Publication #120, Nov. 1990); McLanahan & S ande fur, Growing Up With a Single Parent: What Hurts, What Helps (Harv. U. Press 1994). Seventy percent of juveniles and young adults serving in long-term correctional facilities came from broken homes. Bureau of Justice Statistics, Surveys of Youth in Custody, 1987, p. 1 (U.S. Dep’t of Just. 1988). Having two married parents at home also substantially increases the chances that a child will finish his or her education: If a child comes from a single-parent home, that factor alone increases the risk of dropping out of school by a factor of two. McLanahan, The Consequences of Nonmarital Childbearing for Women, Children and Society, Report to Congress on Out-of-Wedlock Childbearing 229, 231 (Dep’t of Health & Hum. Serv. 1995). Children not living with two biological parents are at greater risk for maltreatment and abuse. Margolin, Child Abuse by Mother’s Boyfriends: Why the Overrepresentation?, 16 Child Abuse and Neglect 541, 545-46 (1992). The rate of child poverty is five times higher for children living with single mothers than for children in intact families. U.S. Bureau of the Census, Statistical Abstract of the United States: 1992, Table 719, p. 457 (112th ed. Washington, D.C. 1992).

As a result, the State contends that society is better off when people are married. Since marriage conveys benefits to children, *262 marriage participants, and society as a whole, the State asserts that the encouragement of marriage is a valid state interest and that it is proper for a state to favor marriage.

Peden takes issue with the State’s cite to the studies mentioned above. Peden contends that the State did not come forward with these studies when it rebutted Pederis motion for summary judgment. Thus, according to Peden, these studies are not a part of the record on appeal and cannot be cited by the State or relied upon by the court. This court has chosen to take judicial notice of these studies. Thus, based on Kansas case law, Kansas statutes, and the studies mentioned above, it is clear that the encouragement of marriage is a valid state interest.

Next, the State argues that the tax rates imposed on single taxpayers which are higher than the highest tax rate imposed on married taxpayers filing jointly are rationally related to the valid state interest of encouraging marriage. The State argues that the financial incentives created by the lower tax rates for married taxpayers will encourage couples to get married and assist couples financially during the marriage.

Peden argues that KITA is not rationally related to the valid state interest of encouraging marriage because there is no proof that lower tax rates for married taxpayers will actually entice any singles to get married. Peden cites to a case which acknowledges a tax code’s incentive potential, but contends that tax codes do not affect marriage in the real world. Mapes v. United States, 576 F.2d 896 (Cl. Ct.), cert. denied 439 U.S. 1046 (1978). This case provides:

“For the tax-mindéd young man, or woman, with a substantial income, the Code adds to the attractiveness of a prospective spouse without taxable income and detracts from one with it. Thus the provisions may have an income-levelling effect. But we have no data showing that as yet they operate in that manner. Love and marriage defy economic analysis.” 576 F.2d at 898.

Further, Peden asserts that even if income tax rates were an actual incentive to marry, federal income tax laws would act as the real incentive, not the Kansas income tax rates, because the federal income tax rates are approximately five times greater than the Kansas income tax rates. Thus, Peden contends that the Kansas income tax rates at issue have not been proven by any data to actually entice *263 single persons to get married and thus are not rationally related to the valid state interest of encouraging marriage.

This argument is unpersuasive because the State is not required to come forward with data that the tax rates are rationally related to the valid state interest of encouraging marriage. As the State points out, “a legislative choice is not subject to courtroom fact-finding and may be based on rational speculation unsupported by evidence or empirical data.” FCC v. Beach Communications, Inc., 508 U.S. 307, 315, 124 L. Ed. 2d 211, 113 S. Ct. 2096 (1993). (Emphasis added.) It is certainly a rational speculation to assume that imposing a lower tax rate on married taxpayers will entice singles to get married and encourage married couples to stay together. Many tax benefits are based on the rationale that they will entice a taxpayer to behave in a certain way. For instance, Congress provides tax benefits to charitable organizations to “encourage the development of private institutions that serve a useful public purpose.” Bob Jones University v. United States, 461 U.S. 574, 587-88, 76 L. Ed. 2d 157, 103 S. Ct. 2017 (1983). To encourage home ownership, Congress allows deductions for mortgage interest and related costs regardless of the taxpayer’s income level or marital status. 26 U.S.C. § 163(h)(2)(D), (h)(3) (1994). To encourage saving, Congress allows for tax-deferred IRAs and 401(k) plans. 26 U.S.C. § 401(k) (1994); 26 U.S.C. § 408 (1994). Thus, it is rational for the State to speculate that favorable tax treatment for married taxpayers in the form of lower rates will encourage single taxpayers to get married and stay married, thereby fulfilling a valid state objective, even though the State has not put forth any hard data to prove this assumption.

Under the rational basis standard, “those attacking the rationality of the legislative classification have the burden ‘to negative every conceivable basis which might support it.’ Beach Communications, Inc., 508 U.S. at 315. Peden has not come forward with any data to negate this conceivable basis and prove that lower tax rates for married taxpayers do not in fact encourage single taxpayers to marry. All Peden has done is argue that if income tax rates encourage taxpayers to marry, it is the federal income tax rates which would act as the actual incentive, not the Kansas income tax rates, *264 because the federal income tax rates are approximately five times greater than the Kansas income tax rates. It may be true that federal income tax rates provide an incentive for couples to marry. However, this does not mean that the Kansas income tax rates would not also provide an additional incentive for taxpayers to get married. Further, if two taxpayers earn approximately the same amount of taxable income, they will be discouraged from getting married under the federal income tax laws due to the marriage penalty. Thus, the lower Kansas tax rates for married couples may counteract the federal marriage penalty to some degree and prevent couples from being discouraged to many. Thus, it is not clear that the federal income tax rates are the only tax rates which affect a taxpayer’s decision to marry. It is reasonable to speculate that the Kansas income tax rates will provide some additional, if not great, incentive for taxpayers to marry and stay married. See Druker, 697 F.2d at 50 (“In the area of family taxation every legislative disposition is ‘virtually fated to be both overinclusive and underinclusive when judged from one perspective to another.’ ”).

Further, even though the State is not required to come forward with data supporting its rational basis, there is some evidence indicating that lower tax rates for married taxpayers do in fact encourage singles to marry. Druker, 697 F.2d at 50 (“marriage penalty” caused couple to divorce). In Mopes, 576 F.2d at 901, the court stated:

“[T]he elevated tax burden might in fact dissuade some couples from entering into matrimony, but does not present an insuperable barrier to marriage. More often it changes the relative attraction of different prospective spouses for the tax-minded individual wishing to marry. For many it provides an incentive to marry. The additional tax liability suffered by two-income couples who cannot avail themselves of the rates for single persons is an indirect burden on the exercise of the right to marry. It is suffered not for marrying but for marrying one in a particular income group.” (Emphasis added.)

Thus, the Mopes court acknowledged that favorable or disfavorable tax treatment can affect a couple’s decision to marry. In the end, the Mopes court ruled that the federal marriage penalty was not a constitutional violation of equal protection because it was rationally related to a valid state interest.

*265 It is rational to speculate that lower tax rates for married taxpayers will entice singles to marry and stay married in order to receive the benefit of the lower rates, even if there is no data to prove this speculation in reality. Importantly, there is no data put forth by Peden to disprove this speculation. As such, the discriminatory classifications of single taxpayers, who are subjected to high tax rates, and married taxpayers, who receive lower tax rates, are rationally related to the valid state interest of encouraging marriage. Thus, the fact that KITA imposes tax rates on single taxpayers which are higher than the highest tax rate imposed on married taxpayers filing jointly is not an unconstitutional violation of equal protection. It is true that the higher single tax rates never even out with the married rates, no matter how much money the married taxpayers earn. However, the benefits of marriage to society never cease, no matter how wealthy the married taxpayers become. Thus, the encouragement of taxpayers to marry and stay married, such as through lower tax rates for married taxpayers, should never cease either, no matter how wealthy the married taxpayers become. Further, all taxation schemes cany some form of discriminatory impact. Johnson, 422 F. Supp. at 971. “The process of evaluating specific tax proposals and weighing their relative discriminatory impacts is primarily a legislative, not a judicial, function.” 422 F. Supp. 974. Thus, we defer to the legislature and find that KITA is not an unconstitutional violation of equal protection.

Reversed.

McFarland, C.J., and Davis, J. not participating. Robert J. Lewis, Jr., J., and Richard W. Wahl, Senior Judge, assigned.