VICE CHIEF JUSTICE TIMMER, opinion of the Court

Brett W. Johnson, Eric H. Spencer, Colin P. Ahler, Snell & Wilmer L.L.P., Phoenix; Dominic E. Draye, Greenberg Traurig, LLP, Phoenix, Attorneys for Jaime A. Molera, et al.

Roopali H. Desai, D. Andrew Gaona, Marvin C. Ruth, Kristen Yost, Coppersmith Brockelman PLC, Phoenix, Attorneys for Invest in Education (Sponsored by AEA and Stand for Children)

Grant Woods, Michael Riikola, Gallagher & Kennedy, P.A., Phoenix, Attorneys for Amicus Curiae Wes Oswald and Kelley Fisher, et al.

Rhonda L. Barnes, Jane Ahern, Arizona House of Representatives, Phoenix; Lisette Flores, Arizona Senate, Phoenix, Attorneys for Amici Legislative Democrats

Oscar S. Lizardi, Rebecca K. O'Brien, Rusing Lopez & Lizardi, P.L.L.C., Tucson; Isaac S. Crum, Rusing Lopez & Lizardi, P.L.L.C., Scottsdale, Attorneys for Amicus Curiae The Arizona Business Community

Erin Adele Scharff, Sandra Day O'Connor College of Law, Phoenix, Attorney for Amici Curiae Tax Scholars

Roy Herrera, Daniel A. Arellano, Jillian L. Andrews, Ballard Spahr LLP, Phoenix, Attorneys for Amici Curiae Ballot Initiative Strategy Center, et al.

Shawn K. Aiken, Shawn Aiken, PLLC, Phoenix, Attorney for Amici Curiae Changing Hands Bookstore, Inc., et al.

Paul F. Eckstein, Daniel C. Barr, Austin C. Yost, Margo R. Casselman, Perkins Coie LLP, Phoenix, Attorneys for Amici Curiae Kathy Hoffman, et al.

Daniel J. Adelman, Arizona Center for Law in the Public Interest, Phoenix, Attorneys for Amici Curiae Save Our Schools Arizona, et al.

Timothy Sandefur, Christina Sandefur, Scharf-Norton Center for Constitutional Litigation at the Goldwater Institute, Attorneys for Amici Curiae Goldwater Institute, et al.

Mary R. O'Grady, Joshua D. Bendor, Emma J. Cone-Roddy, Osborn Maledon, P.A., Phoenix, Attorneys for Amici Curiae Arizona Non-Profit Organizations, et al.

Lisa T. Hauser, Bonnett, Fairbourn, Friedman & Balint, PC, Phoenix, Attorneys for Amicus Curiae Lisa T. Hauser

Kory Langhofer, Statecraft, Phoenix, Attorney for Amicus Curiae Rich Crandall

Kraig J. Marton, Jeffrey A. Silence, Jaburg & Wilk, P.C., Phoenix; Patricia Ronan, Patricia E. Ronan Law, LLC, Phoenix; Christopher Houk, Houk Law Firm, PLLC, Tempe, Attorneys for Amicus Curiae Arizona Employment Lawyers Association

VICE CHIEF JUSTICE TIMMER authored the opinion of the Court, in which CHIEF JUSTICE BRUTINEL, and JUSTICES BOLICK, GOULD, LOPEZ, BEENE, and MONTGOMERY joined.

VICE CHIEF JUSTICE TIMMER, opinion of the Court: ¶1 In Molera v. Reagan , 245 Ariz. 291, 293 ¶ 1, 428 P.3d 490, 492 (2018), we disqualified an "Invest in Education Act" initiative from the 2018 ballot because proponents failed to comply with A.R.S. § 19-102(A), which requires that petition sheets used to gather signatures contain a short description of the initiative's principal provisions (the "100-word description"). Now, we are asked to decide whether the 100-word description for the currently proposed "Invest in Education Act" initiative complied with § 19-102(A). We also address whether petition circulators were paid in accordance with A.R.S. § 19-118.01(A), which prohibits payments "based on the number of signatures collected." We hold that the initiative proponents complied with § 19-102(A) and gathered enough signatures under § 19-118.01(A) to qualify for the November 3, 2020 General Election ballot.

BACKGROUND

¶2 Defendant is a political action committee (the "Committee") seeking to place the "Invest in Education Act" initiative ("Initiative") on the 2020 ballot. To qualify, the Committee was required to obtain 237,645 valid petition signatures demonstrating support for the measure. See Ariz. Const. art. IV, pt. 1, § 1 (2); 2020 Initiative & Referendum Quick Reference Guide, Ariz. Sec'y of State , https://azsos.gov/sites/default/files/2020_Initiative%20_Referendum_Guide.pdf (last visited Oct. 22, 2020). On July 2, 2020, the Committee filed petition sheets containing 435,669 signatures with the Secretary of State. The Secretary reviewed the sheets for statutory compliance pursuant to A.R.S. § 19-121.01(A) and determined that 377,456 signatures were eligible for verification by county recorders. See § 19-121.01(B).

¶3 Plaintiffs are a qualified elector and a political action committee ("Challengers") who oppose the Initiative. On July 10, before completion of the signature verification process, they filed a verified complaint asking the superior court to enjoin the Secretary from placing the Initiative on the ballot because (1) the 100-word description on petition sheets violated § 19-102(A), and (2) after removing signatures gathered by petition circulators who were paid in violation of § 19-118.01(A), the measure lacked enough signatures.

¶4 After conducting a bench trial, the superior court rejected the Challengers’ signature-based objection. But it found that the 100-word description on the petition signature sheets failed to comply with § 19-102(A). The court therefore enjoined the Secretary from certifying and placing the Initiative on the 2020 ballot. ¶5 This expedited appeal and cross-appeal followed. After considering the briefs and authorities filed by the parties and amici, this Court issued an order reversing in part and affirming in part the superior court's judgment and directing the Secretary to include the Initiative in the general election publicity pamphlet and to place it on the general election ballot. This opinion explains our reasoning.

DISCUSSION

I. The 100-word description

A. General principles

¶6 Section 19-102(A) requires initiative sponsors to insert on petition signature sheets "a description of no more than one hundred words of the principal provisions of the proposed measure." The description must be followed by this language:

Notice: This is only a description of the proposed measure (or constitutional amendment) prepared by the sponsor of the measure. It may not include every provision contained in the measure. Before signing, make sure the title and text of the measure are attached. You have the right to read or examine the title and text before signing.

§ 19-102(A).

¶7 Increasingly, sponsors, opponents, and courts have struggled both to identify "principal provisions" in measures and to determine whether their descriptions satisfy § 19-102(A), particularly when the proposed measures are lengthy or complex. There is also confusion about whether omission of a principal provision alone disqualifies a measure from the ballot or whether the omission also must make the description misleading or confusing. Further guidance on what § 19-102(A) does and does not require is warranted before addressing the superior court's ruling here.

¶8 There are two grounds on which to challenge a measure's 100-word description under § 19-102(A). The first is that the sponsor omitted a "principal provision" of the measure from the description. If the court agrees, it should disqualify the measure from the ballot without further inquiry. Whether the omission made the description misleading, confusing, or unfair is irrelevant. We did not intend to suggest otherwise in Molera , where we observed that omitting a principal provision concerning the elimination of tax indexing also "create[d] a significant risk of confusion or unfairness and could certainly materially impact whether a person would sign the petition." 245 Ariz. at 297 ¶ 25, 428 P.3d at 496. Section 19-102(A) requires a sponsor to describe all principal provisions of a measure.

¶9 What are "principal provisions"? They are the "most important," "consequential," and "primary" features of the initiative. Id. ¶ 24 (quoting Sklar v. Town of Fountain Hills , 220 Ariz. 449, 453 ¶ 13, 207 P.3d 702, 706 (App. 2008) ). They are not all provisions. The 100-word description serves as the "elevator pitch" that alerts prospective signatories to the measure's key operative provisions, enabling them to decide in short order whether to sign the petition, refuse to do so, or make further inquiry about the measure. See id. at 297 ¶27, 428 P.3d at 496 (stating that the purpose of the description is to enable prospective signatories to determine whether to endorse the measure for the ballot); see also Kromko v. Superior Court , 168 Ariz. 51, 60, 811 P.2d 12, 21 (1991) (observing that "rare is the elector who stops long enough in the summer heat to read or listen to a complete description of the entire nature and scope of a proposed measure").

¶10 The second ground on which to challenge a 100-word description focuses on the manner of describing the measure's principal provisions. Section 19-102(A) does not require the description to be impartial. Save Our Vote, Opposing C-03-2012 v. Bennett , 231 Ariz. 145, 152 ¶ 28, 291 P.3d 342, 349 (2013). But to comply, the description must describe the principal provisions to accurately communicate their general objectives. In light of the 100-word limit, and the constitutional right to propose initiatives of any length, the sponsor may cast the description in broad terms, and more complex or multi-faceted initiatives may be unavoidably less detailed than shorter measures. See Quality Educ. & Jobs Supporting I-16-2012 v. Bennett , 231 Ariz. 206, 208 ¶ 9, 292 P.3d 192, 194 (2013) (providing that "[t]he length and complexity of the initiative" and the fifty-word limit for the Secretary of State's ballot summary are "factors in assessing compliance" with the statute requiring that summary). If necessary, a sponsor may refer potential signatories to the measure's text for more detail when explaining technical terms or difficult-to-grasp concepts.

¶11 Reasonable people can differ about the best way to describe a principal provision, but a court should not enmesh itself in such quarrels. See id. at 209 ¶13, 292 P.3d at 195. If the chosen language would alert a reasonable person to the principal provisions’ general objectives, that is sufficient. Applying this reasonable person standard, the trial judge should ordinarily decide the sufficiency of a description without expert witness evidence. See Ariz. R. Evid. 702.

¶12 What constitutes a deficient description under § 19-102(A) ? In recent years, we have described such deficiencies as those that are "fraudulent or create[ ] a significant danger of confusion or unfairness," Molera , 245 Ariz. at 295 ¶ 13, 428 P.3d at 494 (quoting Save Our Vote , 231 Ariz. at 152 ¶ 26, 291 P.3d at 349 ), contain "untrue representations designed to defraud potential signatories," or "obscure[ ]" a measure's principal provisions or "thrust," Wilhelm v. Brewer , 219 Ariz. 45, 48 ¶¶ 13–15, 192 P.3d 404, 407 (2008) (quoting Kromko , 168 Ariz. at 59, 811 P.2d at 20 ). We now rephrase these standards more precisely to assist both initiative sponsors in complying with § 19-102(A) and courts in deciding whether sponsors did so.

¶13 The court should disqualify an initiative from the ballot whenever the 100-word description either communicates objectively false or misleading information or obscures the principal provisions’ basic thrust. The latter can occur through the language used to describe the principal provisions or by failing to refer to key features of those provisions. In other words, although sponsors are free to describe the measure in a positive way and emphasize its most popular features, they may not engage in a "bait and switch" in which the summary attracts signers but misrepresents or omits key provisions. In addressing challenges, a court should "consider the meaning a reasonable person would ascribe to the description." Ariz. Chapter of the Associated Gen. Contractors of Am. v. City of Phoenix , 247 Ariz. 45, 48 ¶ 15, 445 P.3d 2, 5 (2019).

B. Application to this case

¶14 The Initiative is nine pages long and proposes to amend both A.R.S. Title 15 (Education) and Title 43 (Taxation of Income). The 100-word description in the petition signature sheets for the Initiative provides:

The Invest in Education Act provides additional funding for public education by establishing a 3.5% surcharge on taxable income above $250,000 annually for single persons or married persons filing separately, and on taxable income above $500,000 annually for married persons filing jointly or head of household filers; dedicates additional revenue to (a) hire and increase salaries for teachers, classroom support personnel and student support services personnel, (b) mentoring and retention programs for new classroom teachers, (c) career training and post-secondary preparation programs, (d) Arizona Teachers Academy; amends the Arizona Teachers Academy statute; requires annual accounting of additional revenue.

The notice required by § 19-102(A) followed the description. See supra ¶ 6.

¶15 The superior court found that this summary omitted five principal provisions, which also made the description misleading and "created a significant danger of confusion or unfairness to a reasonable Arizona voter." It further ruled that use of the word "surcharge" masked the Initiative's proposal to substantially increase the tax rate for wealthier taxpayers. We review the court's ruling de novo. See Molera , 245 Ariz. at 294 ¶ 8, 428 P.3d at 493.

¶16 (1) Percentage distribution of new revenues. The 100-word description listed who would receive the new tax revenues but not in what percentages. The superior court reasoned that the percentage distribution is a principal provision because a reasonable voter's decision whether to support or reject the measure might turn on that knowledge. It pointed out that "[t]o some reasonable voters, devoting 50% of the money generated by the Initiative directly to teacher salaries may have sounded too rich; to other reasonable voters, devoting 50% of the money raised directly to teacher salaries may have sounded too modest." The court also found that this omission risked confusing reasonable voters "who may believe that more or less than 50% of the funds raised would be used to increase teacher salaries."

¶17 The percentage distribution is not a principal provision because it is not the Initiative's "most important," "consequential," or "primary" provision that must be described to alert prospective petition signatories to the measure's key operative provisions. See Molera , 245 Ariz. at 297 ¶¶ 24, 27, 428 P.3d at 496. The Initiative's principal provisions impose a substantially increased tax rate on individuals’ taxable income exceeding $250,000 and $500,000, depending on filing status (the "marginal tax rate"), and dedicate the resulting revenues to public education needs. The 100-word description sufficiently describes these provisions by setting out the tax rate increase and identifying the recipients who will divide the resulting revenues. It was not necessary to describe the percentage distribution for the reader to understand these provisions. Similarly, omission of this detail did not make any part of the description false or misleading or obscure the Initiative's key operative provisions. Pursuant to the required notice beneath the 100-word description, prospective signatories interested in learning the percentage distribution could readily discover it by reading the Initiative text appended to the petition.

¶18 (2) The percentage increase in the marginal tax rate. The superior court found that the Committee omitted a principal provision by failing to describe the percentage increase in the marginal tax rate. It construed Molera , 245 Ariz. at 298 ¶ 29, 428 P.3d at 497, which disqualified that initiative for falsely describing the proposed percentage increase of a marginal tax rate, as meaning that the percentage increase of the marginal tax rate is a principal provision. Likewise, the court ruled that omission of that information concealed "how profoundly taxes are being increased" for wealthier individuals.

¶19 The superior court misconstrued Molera and therefore erred in its ruling. The 100-word description there stated that the measure would raise the tax rate "by 3.46%" on individual incomes over $250,000 and household incomes over $500,000, and "by 4.46%" for individual incomes over $500,000 and household incomes over $1,000,000. Id. at 293 ¶2, 428 P.3d at 492. In fact, the affected tax rates would have increased by 76% and 98%, respectively, making the description false. Id . at 298 ¶29, 428 P.3d at 497. But in finding that the description of the tax rate increase was false, we did not require that any percentage increase in a given tax rate must be included in an initiative description to comply with § 19-102(A), as the superior court here surmised. See id. ¶ 30 (acknowledging that the sponsor could have stated that the marginal tax rate would "increase[ ] by 3.46 and 4.46 percentage points " without providing the rate's percentage increase). Rather, we specified that any description of the increase must be truthful. See id. (stating that the sponsor was not required to describe the change in the tax rate in great detail but could not do so "in a confusing way").

¶20 The description here accurately stated that a 3.5% surcharge would be imposed on individuals’ taxable incomes over specified amounts. This sufficiently communicated a principal provision—raising the marginal tax rate on wealthier taxpayers—to alert a prospective signatory to the Initiative's substance. See Ariz. Chapter of the Associated Gen. Contractors of Am. , 247 Ariz. at 49 ¶ 18, 445 P.3d at 6 (stating sponsors are not required to describe "all potential effects of a measure"). Although the Committee could have provided the percentage increase, it was entitled to cast its description in the best light to garner support, as long as the wording was accurate and did not obscure the tax increase. See Molera , 245 Ariz. at 295 ¶ 13, 428 P.3d at 494 (noting that a sponsor is not required to use neutral wording); Save Our Vote , 231 Ariz. at 152–53 ¶¶ 27–28, 291 P.3d at 349-50 (to same effect). ¶21 Relatedly, the court found the word "surcharge" confusing because prospective signatories could have understood it "to mean a temporary tax, or to mean a modest 3.5% increase of the existing tax rate." But neither the definition of "surcharge" nor other language in the description supports this finding. The term is commonly understood to mean an additional charge, not a temporary one. See Surcharge , Webster's Third New International Dictionary (3d ed. 2002) (defining "surcharge" as meaning "a charge in excess of the usual or normal amount: an additional tax, cost, or impost"); see also Saban Rent-a-Car LLC v. Ariz. Dep't of Revenue , 246 Ariz. 89, 98 ¶ 34, 434 P.3d 1168, 1177 (2019) (recognizing that a surcharge can be a "special tax" not levied generally on all businesses). And because the summary describes the Initiative as providing "additional funding" for K-12 education that is subject to an "annual accounting" requirement, it is doubtful a reasonable voter would have believed the surcharge to be temporary. Also, unlike the situation in Molera , the description here did not assign a percentage increase to the existing tax rate. Instead, it accurately described an additional 3.5% tax on taxable incomes exceeding $250,000.

¶22 (3) The applicability to pass-through business income. Under existing tax law, income generated by some businesses, for example, Subchapter S corporations and partnerships, is "passed through" to owners and taxed as individual taxable income. See, e.g. , Watts v. Ariz. Dep't of Revenue , 221 Ariz. 97, 100 ¶ 8, 210 P.3d 1268, 1271 (App. 2009). The superior court found that the description omitted a principal provision by failing to alert prospective signatories that the proposed marginal tax rate would apply to business income. It further found that this omission created a substantial danger that a signatory would fail to appreciate this fact. We disagree.

¶23 What is taxable as income is dictated by state and federal law. The Initiative does not alter or affect the type of income Arizona has chosen to tax as individual taxable income, including business income. The fact the increased marginal tax rate on individual taxable income would encompass business income taxable to individuals is therefore not a principal provision, but rather is an effect of a principal provision that was not required to be included in the description. See Ariz. Chapter of the Associated Gen. Contractors of Am. , 247 Ariz. at 49 ¶ 18, 445 P.3d at 6. Also, omission of language describing this application does not make any language in the description false or misleading, nor does it obscure a basic thrust of the Initiative. Challengers and amici argue that the Initiative's applicability to pass-through taxable income is material to reasonable voters due to the "impact on economic growth, employment, and the uneven profits that a business generates from year to year." That may be so, but "[t]he proper forum to argue the consequences of passing the Initiative is in statements of support and opposition, editorials, and the like." See id.

¶24 (4) The proscription on decreasing other education funding. The Initiative proposes adding § 15-1284(E), which provides that monies received by school districts, charter schools, and career technical education districts under the Initiative "are in addition to any other appropriation, transfer or allocation of public or private monies from any other source and may not supplant, replace or cause a reduction in other funding sources." The superior court found that this is a principal provision that should have been described because it "[c]urtail[s] the discretion, authority, and operations of the Legislature as it relates to funding public education," which is a constitutionally bestowed function. It additionally found that the omission was confusing because a prospective signatory would not be put on notice of this curtailment.

¶25 The Committee did not violate § 19-102(A) by omitting mention of § 15-1284(E) or its impact on legislative funding decisions in the 100-word description. Curtailing legislative discretion is not the Initiative's most important, consequential, or primary feature, so the proposed statute is not a principal provision. And omitting it from the 100-word description is not fatal. The description states that the new tax revenue would be "additional funding for public education," and proposed § 15-1284(E) furthers that purpose. See Wilhelm , 219 Ariz. at 48 ¶ 15, 192 P.3d at 407 (rejecting challenge to 100-word summary because proposed extension of a statute of repose not mentioned in the description was nevertheless "consistent with the ten-year warranty" highlighted in the description). A reasonable person would not expect that the legislature could effectively nullify that "additional funding" by using the fact of new funding to decrease education funding from other sources. Whether § 15-1284(E) unconstitutionally curtails legislative authority, as the superior court implies, cannot be decided until after its adoption. See Tilson v. Mofford , 153 Ariz. 468, 473, 737 P.2d 1367, 1372 (1987) (concluding that the proper time to consider the constitutionality of a proposed initiative is after its adoption when affected litigants can present the issue).

¶26 (5) Expenditure limitation. Article 9, section 21 of the Arizona Constitution imposes an expenditure limitation on school districts. Proposed § 15-1285(1) declares that this limitation does not apply to revenues generated by the new tax. The court found that the failure to describe this provision's application to districts’ expenditures of Initiative-generated revenues constituted an omission of a principal provision, and that prospective signatories would have been confused by not appreciating that "the Initiative was an attempt to change and/or circumvent Constitutional spending limits." But whether article 9, section 21 limits district expenditures despite § 15-1285(1) is undecided and will remain so unless the Initiative is adopted and later challenged. See Tilson , 153 Ariz. at 473, 737 P.2d at 1372. Section 19-102(A) does not require mention that article 9, section 21 may limit district expenditures or that proposed § 15-1285(1) may be unconstitutional. See Iman v. Bolin , 98 Ariz. 358, 364–65, 404 P.2d 705 (1965) ("[E]ven were the measure in conflict with the Constitution, this has no bearing on the right of the people to enact it. The same is true of an act of the legislature." (internal citation omitted)).

¶27 In sum, the 100-word description here complies with § 19-102(A). The description sufficed to alert prospective signatories to the Initiative's principal provisions to enable them to endorse the measure for the ballot, reject it, or seek more information. It did not contain any objectively false or misleading information or conceal a basic thrust of the Initiative. The superior court erred by ruling otherwise. In light of our holding, we do not address the Committee's additional arguments concerning the 100-word description.

II. Circulator compensation

¶28 The Committee hired AZ Petition Partners, LLC ("Petition Partners"), a signature-gathering business, to obtain enough signatures to qualify the Initiative for the ballot. Petition Partners employed registered circulators and paid them an hourly rate on a weekly basis to collect signatures on petitions for all Petition Partners’ clients, including the Committee.

¶29 Although three pay scales applied while the Initiative petitions circulated, they operated similarly. The scales had multiple levels, each of which set an hourly rate and an expected average number-range of signatures to be gathered each hour. The higher the productivity expectation the higher the rate, and vice versa. Circulators were eligible to move up or down the pay scale from week to week depending on whether they exceeded or fell below their level's productivity expectation the week before and the existence of other factors, such as the number of hours worked and how the circulators conducted themselves. All adjustments to hourly rates were prospective only.

¶30 Circulators could also earn additional money each week through incentive programs. One such program offered in June 2020 permitted circulators to spin a wheel for prizes each Monday when they turned in collected signatures. The prizes ranged from $10 to $100, an extra hour of pay, or double these "spins."

¶31 The superior court rejected Challengers’ arguments that Petition Partners’ hourly rate structure and the "spin-the-wheel" program violated § 19-118.01(A), which prohibits paying circulators "based on the number of signatures collected." It agreed with Challengers that four other incentive programs violated § 19-118.01(A). But after approximating the number of disqualified signatures, the court concluded that more than 300,000 valid signatures remained, which was "well in excess of the legal requirement."

A. Hourly rate structure and spin-the-wheel program

¶32 In 2017, the legislature enacted § 19-118.01(A), which provides:

A person shall not pay or receive money or any other thing of value based on the number of signatures collected on a statewide initiative or referendum petition. Signatures that are obtained by a paid circulator who violates this section are void and shall not be counted in determining the legal sufficiency of the petition.

Paying or receiving compensation in violation of this restriction is punishable as a class one misdemeanor. § 19-118.01(B).

¶33 Challengers argue that Petition Partners’ "forward-looking rates" and the spin-the-wheel program violated § 19-118.01(A) because they were "based on" the number of signatures collected the prior week. The Committee counters, and the superior court agreed, that § 19-118.01(A) imposes a narrower restriction and does not preclude consideration of productivity in compensating circulators.

¶34 Resolving this dispute depends on the meaning of "based on" in § 19-118.01(A), an issue we review de novo. See Ariz. Chapter of the Associated Gen. Contractors of Am. , 247 Ariz. at 47 ¶ 7, 445 P.3d at 4. "In doing so, ‘[w]e interpret statutory language in view of the entire text, considering the context and related statutes on the same subject.’ " Id. (quoting Nicaise v. Sundaram , 245 Ariz. 566, 568 ¶ 11, 432 P.3d 925, 927 (2019) ). If the language is clear and has only one reasonable meaning, we will apply that meaning. See State v. Francis , 243 Ariz. 434, 435 ¶ 6, 410 P.3d 416, 417 (2018). But if the language is susceptible to more than one reasonable meaning, we apply secondary interpretive principles, such as considering "the statute's subject matter, historical background, effect and consequences, and spirit and purpose." See Rosas v. Ariz. Dep't of Econ. Sec. , 249 Ariz. 26, 28 ¶ 13, 465 P.3d 516, 518 (2020).

¶35 The legislature did not define "based on," so we give the term its common meaning. See Chaparro v. Shinn , 248 Ariz. 138, 141 ¶ 14, 459 P.3d 50, 53 (2020). Dictionaries define the verb "base," in relevant part, as "to use as a base or basis for: establish, found," see Base , Webster's Third New International Dictionary (3d ed. 2002), and "to ground," see Base , Black's Law Dictionary (11th ed. 2019). Applying these definitions, § 19-118.01(A) is violated if the compensation paid to a circulator for collecting signatures is dependent on or calculated by, in whole or in part, the number of signatures collected during the compensation period. Thus, for example, § 19-118.01(A) prohibits a circulator from being paid per signature, per completed signature sheet, or by an hourly, daily, or weekly rate that is contingent on collecting a specified number of signatures.

¶36 We reject Challengers’ assertion that "based on" "indicates a broad connection between two things," and Petition Partners was therefore prohibited from basing prospective hourly rates on, or otherwise rewarding circulators for, past productivity. The plain meaning of "base," which uses descriptors like "ground[ing]," "establish[ing]," and "foundation[al]" rather than words like "relating to" does not support this view. But even if "based on" could reasonably mean "a broad connection between two things," we would reject this interpretation in favor of our narrower one after applying secondary interpretive principles. See Rosas , 249 Ariz. at 28 ¶ 13, 465 P.3d at 518.

¶37 First, adopting a narrow interpretation of "based on" avoids possible constitutional conflict. See Slayton v. Shumway , 166 Ariz. 87, 92, 800 P.2d 590, 595 (1990) ("[W]here alternate constructions are available, we should choose that which avoids constitutional difficulty."). By limiting how circulators may be paid, § 19-118.01(A) restricts "core political speech," and the constitutionality of the provision depends on the severity of the burden imposed and its justification. See Meyer v. Grant , 486 U.S. 414, 421–22, 428, 108 S.Ct. 1886, 100 L.Ed.2d 425 (1988) (holding that a Colorado statute prohibiting payment of initiative petition circulators imposed an unjustified burden on core political speech and therefore violated the First and Fourteenth Amendments). Courts in other jurisdictions have upheld bans on per-signature compensation for circulators. See, e.g. , Prete v. Bradbury , 438 F.3d 949, 962 (9th Cir. 2006) (distinguishing Meyer because the per-signature ban simply "prohibit[ed] one method of payment"); Person v. N.Y. State Bd. of Elections , 467 F.3d 141, 143 (2d Cir. 2006) (to similar effect); Initiative & Referendum Inst. v. Jaeger , 241 F.3d 614, 618 (8th Cir. 2001) (to similar effect). The Sixth Circuit, however, struck an Ohio law that limited circulator compensation to pay "on the basis of time worked." Citizens for Tax Reform v. Deters , 518 F.3d 375, 377 (6th Cir. 2008). The court reasoned that "[w]hile petitioners are not constitutionally guaranteed an endless variety of means, when their means are limited to volunteers and to paid hourly workers who cannot be rewarded for being productive and arguably cannot be punished for being unproductive, they carry a significant burden in exercising their right to core political speech." Id. at 386–87.

¶38 The constitutionality of § 19-118.01(A) is not before us. Regardless, we are further persuaded to reject Challengers’ broad interpretation of the statutory restriction to minimize any First Amendment infringement on core political speech. See Slayton , 166 Ariz. at 92, 800 P.2d at 595.

¶39 Second, the legislative history for § 19-118.01(A) supports a narrow interpretation of "based on." See Rosas , 249 Ariz. at 28 ¶ 13, 465 P.3d at 518. The legislature found that other states had "enacted prohibitions on payment per signature" to "reduce fraud in the signature collecting process," and that evidence "suggest[ed] that circulators paid by the hour [ ] have a higher validity rate than those paid by the signature." 2017 Ariz. Sess. Laws. ch. 52, § 5 (1st Reg. Sess.). The bill's sponsor similarly testified at a House committee hearing that the provision would end "the practice of paying circulators per signature." The House rules attorney, apparently aware that the legislature was swimming in constitutionally choppy waters, explained at another hearing that "after lots of discussion" among lawyers, the proposed "restriction against petition circulators being paid on a per signature basis" was likely constitutional because it was "similar enough" to the ban upheld in Prete that it "could be upheld." The legislature's focus on eradicating the practice of per-signature payments, and its awareness of the constitutional implications of restricting other circulator compensation methods, demonstrate it intended a narrow application of the term "based on."

¶40 Finally, we are persuaded to apply a narrow interpretation of "based on" by considering the practical application of Challengers’ broader view. See Rosas , 249 Ariz. at 28 ¶ 13, 465 P.3d at 518. Interpreting "based on" as a "broad connection between two things" would mean that circulators’ compensation could not relate in any way to their only assigned task—collecting signatures. Taken to its logical conclusion, Challengers’ interpretation would prohibit a sponsor from requiring circulators to gather any number of signatures as doing so would constitute payment "based on the number of signatures collected." Such a result would be peculiar considering the legislature permits the practice of paying circulators to collect signatures, see A.R.S. § 19-118, and it would likely violate the First and Fourteenth Amendments as described in Meyer .

¶41 Turning to the record here, we agree with the superior court that Petition Partners’ graduated hourly wage scales did not violate § 19-118.01(A). Petition Partners paid circulators "based on" the number of hours worked each week at fixed hourly rates, regardless of the number of signatures collected. Its prospective adjustment to hourly rates in light of a circulator's past productivity did not make subsequent payments at the new rate "based on the number of signatures collected." Payment at the new rate remained payment for the hours worked rather than the number of signatures collected.

¶42 The superior court likewise did not err by rejecting the challenge to the spin-the-wheel program as violating § 19-118.01(A). The court ruled that Challengers had failed to prove a "correlation to actual number of signatures turned in and a circulator's ability to ‘spin the wheel.’ " It further found that "[t]his program appeared to be used to enhance morale among petition circulators" rather than to compensate for signatures. We defer to the court's factual findings if supported by sufficient evidence, but we review whether Challengers proved a violation of § 19-118.01(A) de novo. See Helvetica Servicing, Inc. v. Pasquan , 249 Ariz. 349, 352 ¶ 10, 470 P.3d 155, 158 (2020).

¶43 The record supports the court's factual findings. Petition Partners advertised that a circulator could spin the wheel once or twice, depending, in part, on the number of self-reported signatures collected. If the program operated that way, whether it violated § 19-118.01(A) would be a closer question. But the evidence at trial showed that a circulator did not have to collect any number of signatures to spin the wheel.

¶44 Circulator Colby Jensen testified he "[didn't] know of any prerequisites" to spin the wheel, including "the number of signatures [a circulator] collected." He characterized the program as "a fun thing at the end of turn-in to keep everyone from getting depressed having to stand in a parking lot around a ton of people for hours," and said he never saw anyone turned away from spinning the wheel who wanted to.

¶45 Petition Partners manager Tom Bilsten similarly described the program as a "show" and stated that the number of self-reported signatures did not disqualify a circulator from spinning the wheel. He said Petition Partners "did not deny anyone an opportunity to spin the wheel as long as they were an employee in good standing and they were nice to people." Bilsten emphasized that Petition Partners never conditioned a spin-the-wheel opportunity on the number of signatures collected by a circulator. Owner Andrew Chavez echoed this testimony, stating that "people who actually didn't get their signatures ... spun the wheel too."

¶46 In light of this evidence, the superior court did not err in ruling that the spin-the-wheel program did not violate § 19-118.01(A). Because circulators did not spin the wheel in exchange for collecting a specific number of signatures but did so as a consequence of employment, the prizes were not payments "based on the number of signatures collected."

B. Identifying disqualified signatures

¶47 Challengers next argue the superior court incorrectly disqualified only signatures collected by circulators during the weeks they participated in one or more of the bonus programs that violated § 19-118.01(A). They assert that § 19-118.01(A) disqualifies all signatures collected by those circulators even when they were paid in compliance with that statute. We disagree.

¶48 Section 19-118.01(A) could reasonably mean that all signatures ever gathered by a circulator are disqualified or that only signatures gathered when the circulator was paid in violation of the statute are disqualified. We adopt the latter interpretation because it serves the legislative purpose in reducing fraud in the signature collecting process, see supra ¶ 39, while imposing a lesser burden on core political speech, see supra ¶¶ 37–38.

C. Preliminary injunction

¶49 Challengers finally argue the superior court erred by refusing to preliminarily enjoin the Initiative from the ballot to permit them to conduct discovery to identify the number of signatures collected during the weeks circulators were paid bonuses in violation of § 19-118.01(A). We review the denial of a preliminary injunction for an abuse of discretion. Valley Med. Specialists v. Farber , 194 Ariz. 363, 370 ¶ 21, 982 P.2d 1277, 1280 (1999).

¶50 The superior court ruled that Challengers did not have a strong likelihood of successfully voiding enough signatures to keep the Initiative from the ballot. See Apache Produce Imps., LLC v. Malena Produce, Inc. , 247 Ariz. 160, 164 ¶ 10, 447 P.3d 341, 345 (App. 2019) (listing "a strong likelihood of success" on the merits as a consideration in ruling on a preliminary injunction request). Although the court found that 146 circulators received weekly compensation that included improper payments, the record did not reflect exactly how many signatures were collected during those weeks. Regardless, the court reasoned that even if those circulators worked forty hours per week and collected twelve signatures per hour—the maximum collected by the most productive circulators—only 70,080 signatures would be voided. Because the Secretary had approved 377,456 signatures, subtracting the voided signatures would leave more than the 237,645 signatures required to place the measure on the ballot.

¶51 Challengers argue that their inability to demonstrate a likelihood of success on the merits was "tied to [the Committee's] failure to maintain sufficient records," and so the Committee "should bear the burden of proving that the circulators who were improperly paid collected too few ballots to enjoin the Initiative." Even if we agreed, we fail to see how enjoining the Initiative to permit discovery would have remedied insufficient record-keeping or a misplaced burden of proof. Regardless, we reject Challengers’ implicit premise that the Committee was obligated to maintain any records linking circulator compensation to the number of signatures collected. Neither § 19-118.01(A) nor any other authority cited by Challengers required such record-keeping.

¶52 We also disagree that Challengers’ burden to prove the invalidity of petition signatures by clear and convincing evidence, see Leach v. Reagan , 245 Ariz. 430, 437 ¶ 30, 430 P.3d 1241, 1248 (2018), shifted to the Committee to prove that the number of signatures collected under the improper bonus programs were too few to disqualify the Initiative. The caselaw relied on by Challengers shifts the burden of coming forward with evidence, not the burden of proof, "[w]hen proof of a negative assertion lies ‘peculiarly within the knowledge of the adverse party.’ " See Woerth v. City of Flagstaff , 167 Ariz. 412, 419, 808 P.2d 297, 304 (App. 1990) (quoting Sw. Cotton Co. v. Ryan , 22 Ariz. 520, 533, 199 P. 124 (1921) ); see also Parker v. City of Tucson , 233 Ariz. 422, 432 ¶ 24 n.9, 314 P.3d 100, 110 (App. 2013). Here, Challengers were not tasked with proving a negative by establishing the number of signatures that were collected by 146 circulators during the weeks in which they received payment in violation of § 19-118.01(A).

¶53 In sum, the superior court did not err by ruling that Petition Partners’ graduated hourly rate scale and spin-the-wheel program complied with § 19-118.01(A). The court correctly disqualified only signatures collected by circulators during the pay periods in which they were paid bonuses that violated § 19-118.01(A), and properly denied Challengers’ request for a preliminary injunction. In light of our holding, we need not address the Committee's arguments that the court erred by invalidating any signatures under § 19-118.01(A).

CONCLUSION

¶54 For the foregoing reasons, we affirm the superior court's judgment in part and reverse in part. The Initiative will be placed on the November 3, 2020 ballot.