ORTEGA, P.J.
*801This case involves Article IV, section 1b, of the Oregon Constitution, which makes it "unlawful to pay or receive money or other thing of value based on the number of signatures obtained on an initiative or referendum petition." Wolfe, the chief petitioner of Initiative Petition 24 in 2012, petitions for judicial review of a final order of the Secretary of State (secretary) that imposed a civil penalty of $65,000 against Wolfe for 26 violations of Article IV, section 1b 's prohibition against paying circulators per signature collected. Although Wolfe raises several assignments of error, we write only to address his challenge to evidentiary rulings of the administrative law judge (ALJ) concerning Wolfe's argument that the pay-per-signature ban imposed by Article IV, section 1b, and the vicarious liability and civil penalty provisions promulgated by the Elections Division of the secretary (division) to enforce the ban burden his free speech rights under the First Amendment to the United States Constitution.1 Because the evidence he sought to introduce was relevant and its exclusion substantially prejudiced Wolfe, we reverse and remand the secretary's order.
We begin by explaining the background of Article IV, section 1b, and the statutes and rules implementing it. Article IV, section 1b, was added to the Oregon Constitution by the passage of Ballot Measure 26 (2002), the purpose of which was to "protect the integrity of initiative and referendum petitions."2 Official Voters' Pamphlet 106, General *802Election, Nov. 5, 2002; see Ecumenical Ministries v. Oregon State Lottery Comm. , 318 Or. 551, 560 n. 8, 871 P.2d 106 (1994) (examining voters' pamphlet materials as part of "legislative facts" in determining meaning of constitutional provision adopted through initiative process). That constitutional provision states:
"It shall be unlawful to pay or receive money or other thing of value based on the number of signatures obtained on an initiative or referendum petition. Nothing herein prohibits payment for signature gathering which is not based, either directly or indirectly, on the number of signatures obtained."
Those who argued in favor of the measure asserted that the initiative process had been dominated by an unregulated signature-gathering industry, resulting in fraudulent and *1123forged signatures. See Voters' Pamphlet at 107 (proponents of the measure asserting that the "payment-per-signature business has gotten out of control," the signature-gathering industry has "no accountability [and] is ripe for fraud and abuse," and "[i]f [the measure] passes, there will be no reason to cheat. A person can work 8, 10, 12 hours a day gathering signatures and be paid accordingly-without the incentive to copy signatures from one petition to another.").
The division promulgated rules to effectuate Article IV, section 1b. Included among those rules is OAR 165-014-0260(2), which provides:
"Section 1b and ORS 260.569 _bans the practice of paying circulators or others involved in an initiative, referendum, candidate nominating petition or voter registration card collection effort if the basis for payment is the number of signatures obtained. This means that payment cannot be made on a per signature basis. Employment relationships that do not base payment on the number of signatures collected are allowed. Allowable practices include: paying an hourly wage or salary, using express minimum signature requirements (quota), terminating those who do not meet the productivity requirements, adjusting salaries prospectively relative to productivity, and paying discretionary bonuses based on reliability, longevity and productivity, provided no payments are made on a per signature basis. The use of express minimum signature requirements (quota) for an initiative or referendum petition is allowable *803so long as that requirement is disclosed to the Elections Division on the SEL 320 as part of accounts."
A chief petitioner need not have actual knowledge that circulators for the petition campaign are being paid per signature to be deemed responsible for violating Article IV, section 1b. Under ORS 260.561(1)(b), if a chief petitioner
"has knowledge or should have had knowledge of a violation of * * * section 1b, Article IV of the Oregon Constitution, or any rule adopted by the Secretary of State related to section 1b, Article IV of the Oregon Constitution, petition sheets or circulator training, registration or certification, committed by a person obtaining signatures on the chief petitioner's petition or prospective petition or a contractor or subcontractor, as defined in ORS 260.563, the violation by the person obtaining signatures or the contractor or subcontractor is conclusively considered a violation by the chief petitioner."
Additionally, the division has promulgated OAR 165-014-0260(4), which provides that "the chief petitioners are responsible for insuring that agents of the chief petitioner (anyone who is delegated the task of obtaining signatures on the initiative or referendum petition) do not violate Section 1b."3
Under OAR 165-014-0260(5), each individual signature sheet violating Article IV, section 1b, which is turned in counts as a single violation.4 The minimum penalty for not *1124*804complying with Article IV, section 1b, began as a $100 fine per signature sheet, which was increased to $250 per signature sheet. Prete v. Bradbury , 438 F.3d 949, 952 n. 1 (9th Cir. 2006) ("A violation of Measure 26 will result in civil penalties of a minimum of $100 for each signature sheet containing signatures collected in violation of Measure 26."); Day v. Elections Division , 246 Or. App. 140, 142, 265 P.3d 16 (2011) (a Measure 26 violation results in a $250 penalty per violation). In 2009, however, the division increased the minimum civil penalty to $2,500 per signature sheet.5
Pay-per-signature restrictions are not without controversy, however. First Amendment challenges to restrictions placed on payment to petition circulators have been litigated in many states, including Oregon. The First Amendment, which is made applicable to the states by the Fourteenth Amendment to the United States Constitution, prohibits the enactment of laws "abridging the freedom of speech." In the seminal decision on this topic, Meyer v. Grant , 486 U.S. 414, 108 S.Ct. 1886, 100 L.Ed.2d 425 (1988), the United States Supreme Court considered the constitutionality of a Colorado statute that completely barred payment for initiative-petition circulators. The Court explained that the circulation of initiative petitions involves "both the expression of a desire for political change and a discussion of the merits of the proposed change" and, because "the circulation of a petition involves the type of interactive communication concerning political change, [it] is appropriately described as 'core political speech.' " Id. at 421-22, 108 S.Ct. 1886. Further, the Court concluded that a bar on payment "involves a limitation on political expression subject to exacting scrutiny." Id . at 420, 108 S.Ct. 1886 (citing Buckley v. Valeo , 424 U.S. 1, 45, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976) ). Because the payment bar made it "less likely that [proponents of an initiative measure to *805amend the Colorado Constitution regarding trucking regulation would] garner the number of signatures necessary to place the matter on the ballot, thus limiting their ability to make the matter the focus of statewide discussion," the Court struck down the prohibition against paying circulators. Id . at 422-23, 108 S.Ct. 1886.
A decade later, in Buckley v. American Constitutional Law Foundation, Inc. , 525 U.S. 182, 186, 119 S.Ct. 636, 142 L.Ed.2d 599 (1999), the Court addressed the constitutionality of another attempt by the Colorado legislature to curb payment of petition circulators striking down that state's statute requiring that (1) petition circulators be registered Colorado voters, (2) petition circulators wear an identification badge bearing the circulator's name and status as a volunteer or paid circulator, and (3) initiative proponents disclose publicly the names, addresses, and amounts paid to circulators. The Court observed that the First Amendment does not categorically prohibit a state's efforts to regulate election processes. Id . at 191, 119 S.Ct. 636 ("States allowing ballot initiatives have considerable leeway to protect the integrity and reliability of the initiative process, as they have with respect to election processes generally.").
As to the registered voter requirement, the Court explained that it burdened speech rights because it
"decreases the pool of potential circulators as certainly as that pool is decreased by the prohibition of payment to circulators. Both provisions limit the number of voices who will convey the initiative proponents' message and, consequently, cut down the size of the audience proponents can reach."
Id. at 194-95, 119 S.Ct. 636 (internal quotation marks and alterations omitted). The Court also concluded that the name-badge requirement and the disclosure provisions burdened speech rights, explaining that both provisions discouraged participation in the circulation process. Id. at 198-204, 119 S.Ct. 636.
Among lower court decisions that followed Meyer and Buckley , two federal decisions, *1125central to the division's reasoning and the parties' arguments, as we explain below, bear mention. Both cases, Prete , 438 F.3d at 961, and Independence Institute v. Gessler , 936 F.Supp.2d 1256, 1274 (D. Col. 2013), used the framework set out in Timmons v. Twin Cities Area New Party , 520 U.S. 351, 358-59, 117 S.Ct. 1364, 137 L.Ed.2d 589 (1997), for determining whether a state election restriction violates the First Amendment:
"[The Court weighs] character and magnitude of the burden the State's rule imposes on those rights against the interests the State contends justify that burden, and consider[s] the extent to which the State's concerns make the burden necessary. Regulations imposing severe burdens on plaintiffs' rights must be narrowly tailored and advance a compelling state interest. Lesser burdens, however, trigger less exacting review, and a State's important regulatory interests will usually be enough to justify reasonable, non-discriminatory restrictions."
(Internal citations and quotation marks omitted.)6
In Prete , the Ninth Circuit held that an Oregon federal district court did not clearly err when it concluded that, based on the evidence presented in that case, Measure 26 ( Article IV, section 1b ) did not unconstitutionally burden petitioners' core political speech rights under the First Amendment. 438 F.3d at 971. Upon reviewing Meyer and Buckley , the Ninth Circuit distinguished Oregon's pay-per-signature ban from the total ban on payments held unconstitutional in Meyer because,
"Measure 26 does not completely prohibit the payment of initiative-petition circulators. Instead it prohibits one method of payment. Plaintiffs claim that Measure 26 in practice limits the available pool of people willing to circulate petitions. To the extent Meyer may be read to indicate that any resulting decrease in the pool of available circulators is sufficient to constitute a 'severe burden' under the First Amendment, in Buckley the Court refined its analysis and made clear that the degree of the decrease resulting from the measure is properly considered in determining the severity of the burden."
Prete , 438 F.3d at 962-63 (emphasis in original). The court concluded that the district court's findings that the plaintiffs'
*807evidence asserting that circulators were unwilling to work on an hourly basis was "unsupported speculation" and that an estimate that asserted that signature gathering costs had increased because of the ban was unreliable were not clearly erroneous. Id . at 964-66.
Although the court recognized that, "from an economic perspective, eliminating one method of payment (but not every method, a la Meyer ) for petition circulators could result in some barriers to entry in the signature procurement market" and paying circulators by the signature "can be more productive of signatures than paying an hourly wage," the court observed that it was "a question of historical fact" whether the measure actually created any barriers to entry and the district court's finding that no such barriers existed was not clear error. Id. at 967. Further, to the extent that the petitioners had made a showing of a First Amendment burden, the evidence was only sufficient to show a lesser, permissible burden that was justified by the state's regulatory interest in preventing fraud. Id . at 968. The court reasoned that Oregon's pay-per-signature ban is "quite limited in its proscription, barring only payment of petition circulators on the basis of the number of signatures gathered. It does not prohibit adjusting salaries or paying bonuses according to validity rates or productivity * * * which could likely counter any barriers to entry." Id . The court upheld the ban. Id.
More recently, in Independence Institute , a Colorado district court considered the constitutionality of further attempts by the Colorado legislature to regulate paid petitioners. 936 F.Supp.2d at 1258. The Colorado legislature had enacted a hybrid compensation scheme that made it unlawful to pay a circulator more than 20 percent of the circulator's *1126compensation on a per-signature basis, which, "as a practical matter, limit[ed] per signature compensation to bonuses or incentive payments," requiring "that circulators receive the majority of their compensation in the form of hourly payments." Id . at 1259. The plaintiffs-petition circulators, nonprofit organizations, and petition entities involved in Colorado's initiative and referendum process-challenged the statute. Id . at 1258. The evidence in that case established that the effect of the payment limitation was to raise the per-signature cost to a petition entity and, therefore, *808the "cost of running signature-gathering campaigns will increase because the hybrid scheme excludes some professional circulators from working in Colorado and makes the signature-gathering process significantly less efficient." Id . at 1260.
The evidence showed that itinerant professional signature gatherers were reluctant or unwilling to work on an hourly basis because they would be paid less and, thus, the payment-limitation statute would reduce the pool of professional circulators available to work on signature-gathering campaigns. Id . at 1261-64. Thus, petition entities would have to expend more resources training new circulators and, because new circulators were less efficient and had lower validity rates, the loss of itinerant professional circulators (and the need to use low-volume professionals) would raise petition campaign costs and make it less likely that a petition would qualify for the ballot. Id . at 1264. The district court also found that, although "an increase in signature gathering costs is unlikely to have an effect on individuals or entities who can raise large sums to qualify a measure for the statewide ballot[,] the cost increase is likely to disproportionally impact individuals with limited resources by making measure qualification unaffordable." Id . at 1268.
Relying on Buckley , the court sought to determine the degree to which the payment-limitation statute burdened the expressive activities of the plaintiffs. Unlike in Prete (and other cases), the court concluded that the plaintiffs had shown that the statute at issue imposed a severe burden on their expressive activities and, thus, applied strict scrutiny. Id. at 1277. Although Colorado had a "compelling interest in ensuring the reliability and honesty of the referendum and initiative process," the court found that no "evidence at trial established a connection between the type of compensation and the validity rates of petition sections." Id. at 1278. Therefore, the court concluded, "[g]iven the availability of other effective and less burdensome statutory tools to safeguard [Colorado's] interest," the statute posed an "undue restriction on plaintiffs' First Amendment rights." Id . at 1279. Thus, the payment-limitation statute was unconstitutional. Id . at 1280.
*809With that background in mind, we state the facts consistently with the ALJ's findings, which are supported by substantial evidence in the record.7 See Vector Marketing Corp. v. Employment Dept. , 275 Or. App. 999, 1001, 365 P.3d 686 (2015), rev. den. , 359 Or. 667, 379 P.3d 528 (2016). Wolfe was the chief petitioner of Initiative Petition 24 (IP 24), which was a 2012 initiative that sought to amend the Oregon Constitution to legalize personal and private possession and cultivation of marijuana. In November 2011, Wolfe hired two managers to supervise the circulation of petition sheets to obtain the requisite number of signatures to place IP 24 on the ballot. Those management duties, which were delegated by Wolfe, included the primary responsibility for hiring, firing, training, and supervising paid circulators.
The campaign advertised on Craigslist to recruit circulators, offering $20 an hour to gather signatures. Several circulators were hired in December 2011, including Bennett and Colbert (the two paid circulators whose per-signature payments formed the basis for the fine imposed on Wolfe). At the end of December, the campaign began to have new hires sign an employment agreement, which described the nature of the work and how wages would be paid. Namely, the agreement specified that circulators would be paid $20 *1127an hour-noting that payment by the signature was illegal in Oregon-and setting out a quota of 19-21 signatures for each hour of work paid and that employment would be terminated if the quota was not met, although failure to meet quota was not an automatic ground for termination; the campaign's managers had discretion to retain circulators who had not met quota if they could be redeemed with more training. Neither Bennett nor Colbert signed the agreement.
Every week, IP 24 circulators turned in their signature sheets at mandatory weekly meetings. Although these meetings were primarily handled by the circulation managers, Wolfe remained actively involved in the campaign: He counted signature sheets turned in by the circulators, signed and submitted paperwork required by the division, *810and reviewed payroll summary sheets compiled by a manager before they were submitted to the payroll company. Wolfe did not run the meetings but attended several of them during the first few months of 2012. During January and February 2012, the campaign did not use time sheets and did not require the circulators to document the number and timing of the hours they had worked. What the circulators did submit was a cover sheet for the signature sheets on which they wrote their name, the number of signatures collected per day, and the total number of signature sheets collected for the week. There was no place on the cover sheet for the circulators to document the number of hours worked for the week. During January and February 2012, there was a close correlation between the number of signatures circulators gathered and the circulators' pay.
After having received complaints about the payment methods of the campaign, the division notified Wolfe in April 2012 of its proposed penalty of $65,000-$2,500 for each of the 26 signature sheets gathered by Bennett and Colbert-for violating Article IV, section 1b. Wolfe filed a hearing request to contest the penalty, and a hearing was held in the early months of 2013. At various points in the contested hearing, the constitutionality of the pay-per-signature ban and its implementation were discussed. During the proceeding, Wolfe attempted to testify about what he believed the campaign would have needed to comply with the pay-per-signature ban, in support of his position that doing so was prohibitively expensive and therefore unconstitutional. The division objected on the grounds of relevance. Wolfe replied that "facts that pertain to the burden on collecting signatures is 100 percent relevant to the constitutionality of the pay per signature ban," citing Independence Institute , and how that decision's discussion of Oregon's pay-per-signature ban was relevant to the burden imposed on his speech rights. At the hearing, the ALJ refused to hear testimony and entertain other evidence that Article IV, section 1b, burdened Wolfe's speech rights on the bases that such a challenge was "above [her] pay grade" and that, because the challenge was brought in a contested case hearing, the "efficacy of the constitutional provision [was] not squarely before" her. Nor did the ALJ view the hearing *811as "the proper forum to relitigate the Prete case." The ALJ, however, did permit Wolfe to submit an offer of proof of testimony that he had intended to elicit from Ross Day, an attorney who had been very involved in the petitioning process as a circulator and owner of a circulation company, and by representing other petitioners (including the plaintiffs in Prete ), as well as other evidence.
The ALJ reiterated, in her proposed order, that the exhibits offered as proof by Wolfe were irrelevant. Also in the proposed order, the ALJ addressed Wolfe's First Amendment challenge, remarking that Independence Institute was neither instructive nor binding. "Rather, the governing federal court opinion for purposes of Oregon's pay-per-signature ban is Prete [.]" The ALJ "adopt[ed]" the Prete court's reasoning and "decline[d] to revisit the issue of whether Article IV, section 1b violates the First Amendment." In the ALJ's view, the only relevant change since Prete was decided was the "significantly larger fine" of $2,500 for each violation of Article IV, section 1b, in this case. The ALJ reasoned, however, that there was no evidence that a civil penalty of $65,000 acted as a prior restraint on Wolfe's speech. Wolfe, the ALJ observed, continued to gather signatures after the April 2012 *1128notice until the July 2012 deadline for turning in signatures.
Wolfe took exceptions to the ALJ's rulings excluding evidence concerning his argument that Article IV, section 1b, and its vicarious liability and civil penalty provisions unduly burdened his First Amendment speech rights. The division rejected those exceptions and adopted the ALJ's rulings on the following bases:
"The constitutional issue at the hearing was whether, if Wolfe violated Article IV, section 1b, the imposition of a civil penalty on Wolfe for those violations violates his rights under the United States Constitution * * *. The Division notes that the Independence Institute exhibits * * * are not relevant because they concern a Colorado statute not at issue in this proceeding, nor do they pertain to Wolfe's conduct. The Division notes that the Prete v. Bradbury exhibits * * * are not relevant because they do not concern Wolfe's conduct or IP 24. Moreover, those exhibits do not aid Wolfe's argument because, in Prete , the Ninth Circuit upheld *812the constitutionality of Article IV, Section 1b, against an as-applied challenge. The Division notes that the Walker v. Oregon [, No 08-06135-HO, 2010 WL 1224235 (D Or. Mar. 23, 2010) ] exhibits * * * are not relevant because they do not concern Wolfe's conduct or IP 24. Further, like the excluded exhibits from Prete , the Walker exhibits were submitted in a case that upheld the constitutionality of certain Oregon election laws. Therefore, they would not have advanced Wolfe's argument. Thus, even if they had been relevant in this as-applied challenge to the constitutional provision, the exclusion of those exhibits did not prejudice Wolfe."
The division also considered the excluded exhibits and, without any explanation, concluded that Wolfe was not prejudiced by the exclusion because he had not made a showing that his speech rights were unduly burdened.8
On judicial review, Wolfe assigns as error the exclusion of the evidence concerning his First Amendment challenge. Before we proceed to discuss whether the ALJ erred in excluding the evidence as irrelevant, we must properly frame Wolfe's First Amendment argument. Like the Colorado pay-per-signature restrictions that were determined by a federal district court to violate the First Amendment rights of initiative petitioners in Independence Institute , in Wolfe's view, the pay-per-signature restrictions in Oregon burden his and other petitioners' First Amendment speech rights. Stated differently, he argues, as he did below, that he does not challenge Article IV, section 1b, alone; rather, Wolfe challenges Article IV, section 1b, as implemented by the division's rules and its application, namely the civil penalty of $2,500 per violation of the pay-per-signature ban on a theory of vicarious liability. Thus, his challenge is not restricted to the *813$65,000 penalty, nor is his challenge restricted to Article IV, section 1b.9
To the extent that the order excluded Wolfe's evidence as irrelevant on the basis that Wolfe was solely challenging *1129Article IV, section 1b, and that that argument was foreclosed by the Ninth Circuit's Prete decision, we are persuaded that Wolfe's argument was broader than that. Furthermore, the secretary's reliance on Prete is problematic. To begin with, we "are not bound by the Ninth Circuit's decisions on federal questions that are unresolved by the United States Supreme Court." Mears v. Marshall , 138 Or. App. 476, 478, 909 P.2d 212 (1996) (citing Beason v. Harcleroad , 105 Or. App. 376, 382, 805 P.2d 700 (1991) ). That is, "we consider decisions by all lower federal courts, including the Ninth Circuit, as persuasive authority but employ an independent analysis to reach our own conclusion." Id .
Further, the Ninth Circuit in Prete did not hold that Article IV, section 1b, is constitutional. 438 F.3d at 953 n. 5. Rather, its decision was constrained by its standard of review, which is that, in First Amendment cases, questions of historical fact are reviewed for clear error, "while constitutional questions of fact (such as whether certain restrictions create a 'severe burden' on an individual's First Amendment rights) are reviewed de novo [for legal error]." Id . at 960-61 ; Vasquez-Lopez v. Beneficial Oregon, Inc. , 210 Or. App. 553, 582, 152 P.3d 940 (2007) ("The latter question is a legal issue which we review for errors of law-what the federal courts call 'de novo ' review."). That is, in Prete , because the district court did not clearly err in finding that the plaintiffs in that case failed to establish that Article IV, section 1b, "significantly diminishes the pool of potential circulators, increases the cost of signature gathering, or increases the validity rate of signature gathered," the Ninth Circuit could not conclude *814that Measure 26 imposes a severe burden under the First Amendment. 438 F.3d at 953 n. 5 ("We express no opinion, however, regarding whether Measure 26 could withstand strict scrutiny had plaintiffs proven the measure imposed a 'severe burden' under the First Amendment."). The question of the burden that the restrictions placed on the payment of circulators is fact intensive, and the Ninth Circuit's conclusion that the trial court did not err when it found that the evidence was insufficient does not preclude creating a sufficient record to challenge such a conclusion in a different case.
With that said, we turn to our own standard of review regarding the secretary's exclusion of Wolfe's evidence on the ground that it was irrelevant. ORS 183.450(1) governs generally the correctness of agency relevance rulings: "Irrelevant, immaterial or unduly repetitious evidence shall be excluded but erroneous rulings on evidence shall not preclude agency action on the record unless shown to have substantially prejudiced the rights of a party." In considering whether an agency erred in upholding an ALJ's decision to exclude evidence on the basis of irrelevance, we review for legal error. See Adams v. Board of Clinical Social Workers , 201 Or. App. 500, 506, 119 P.3d 260, rev. den. , 339 Or. 700, 127 P.3d 1202 (2005) ("The relevance of evidence is a legal question that we review for errors of law.").
Further, although "irrelevant" is not defined by the Oregon Administrative Procedures Act, we have concluded that the legal inquiry is the same as it is for OEC 401, in which evidence is relevant if it has any "tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence." Gaylord v. DMV , 283 Or. App. 811, 819, 391 P.3d 900 (2017). " OEC 401 establishes a very low threshold for the admission of evidence; evidence is relevant so long as it increases or decreases, even slightly, the probability of the existence of a fact that is of consequence to the determination of the action." State v. Davis , 269 Or. App. 532, 541, 345 P.3d 499, rev. den. , 358 Or. 69, 363 P.3d 500 (2015) (internal quotation marks omitted). In addition to our inquiry as to whether the evidence was relevant, if the evidence was erroneously excluded, we must answer the question whether *815the exclusion "substantially prejudiced" Wolfe's rights. ORS 183.450(1). Thus, our task here is to determine if the excluded evidence has any tendency to make it more likely that the Article IV, section 1b, pay-per-signature ban, as implemented here by means of the vicarious liability and civil penalty rules promulgated and enforced by the division, unduly *1130burdened Wolfe's First Amendment speech rights, and if so, whether Wolfe was substantially prejudiced by the exclusion.
Turning to the excluded evidence, we categorize it as follows: (1) exhibits including affidavits, declarations, and a portion of a transcript from Prete or Independence Institute ; (2) the declaration of Ross Day, who testified at the contested case hearing and was prevented from discussing the burden of Article IV, section 1b, as implemented by the division and the state's interest in preventing fraud; (3) affidavits used in Walker to challenge the passage of H.B. 2082 (2007), later codified mainly as ORS 250.048, ORS 260.262, and ORS 260.563, which imposed training and registration requirements, among other restrictions, for paid signature-gatherers, and; (4) assorted newspaper articles, a memorandum, and a document produced by a political advocacy group.
With respect to the first category, affidavits, declarations, and trial testimony from professional petition circulators in the Independence Institute case were offered as evidence to show that professional signature gatherers would avoid working in states with a ban against pay-per-signature compensation. In one such affidavit, a professional circulator who had experience working hourly in Oregon as a circulator averred that "in the future it is very unlikely that I will participate in petition drives as a signature gatherer that is paid with an hourly rate. Oregon is one state I avoid for that very reason." Another professional signature gatherer had also managed other signature gatherers and averred that instituting a fixed hourly system would deter professional signature gatherers and that petition campaign managers would likely be left with "largely unskilled workers" because professionals will travel to other states, thus forcing campaigns to pay unproven circulators and hire managers to train and supervise the unskilled circulators.
*816In the Prete affidavits, experienced petition supervisors and petition-company operators averred that paying circulators by the signature is more cost-effective than paying circulators hourly, which significantly increases the costs of petition campaigns, and would make it more difficult for initiatives to qualify for the ballot. Moreover, the Prete affidavits included information about the signature-gathering industry, including that the industry relies on professional signature gatherers who work in multiple states, some of whom have refused to work in Oregon because of the pay-per-signature ban.
As for the second category of excluded evidence, in his declaration, Day, an experienced petition circulator, chief petitioner, and attorney for initiative cases, averred that, "because of the regulations and restrictions imposed on Oregon's initiative process by the Secretary of State," he had ceased being a chief petitioner since 2010 and would not do so in the future. That was so because the "risk is too high" because of substantial civil penalties that can be imposed on a chief petitioner without the chief petitioner "having any actual knowledge of a violation." Further, Day stated that he had personal knowledge of the effects of Article IV, section 1b, and "the other 'implementation' restrictions on paying per signature and how they affect the cost of circulating an initiative petition." According to Day, the added restrictions, which included the potential risk of being subject to "massively increased monetary penalties," substantially increased the market price for companies hired to collect the signatures required to place a statutory or amendment measure on the ballot. Many persons interested in qualifying a measure for the Oregon ballot had told Day that the cost was too prohibitive. Further, in Day's view, several well-known petition companies and petition circulators had refused to ply their trade in Oregon because of "onerous additional requirements imposed" to enforce the ban, the consequence of which is that chief petitioners must rely on less experienced signature gatherers, resulting in lower efficiency and higher invalidity rates of signatures gathered.
The secretary upheld the ALJ's exclusion of the Day declaration and the Prete and Independence Institute exhibits mainly because they did not specifically pertain *817to Wolfe's conduct or IP 24. Yet neither Meyer , Buckley , nor the cases that decided First Amendment challenges to pay-per-signature restrictions relied on by the parties required *1131evidence that the restrictions burdened the actual conduct of chief petitioners or the success of particular petition campaigns. Rather, those cases found relevant evidence that the state restrictions burdened the right to exercise core political speech by participating in the initiative process. See, e.g. , Buckley , 525 U.S. at 193, 119 S.Ct. 636 (statistical evidence, rather than historical evidence regarding the particular campaigns of the plaintiffs, was relevant to show that Colorado's in-state voter registration requirement reduced the number of petitioners available to gather signatures); Prete , 438 F.3d at 963-64 (the issue was whether Article IV, section 1b, imposed a severe burden on the circulation of initiative petitions generally, not a particular petition campaign of the plaintiffs, and evidence that petition circulators were less likely to work in Oregon was relevant (although not persuasive to the district court) ).
In this case, the Independence Institute , Prete , and Day exhibits are relevant to show that Article IV, section 1b, and the civil penalty and the vicarious liability provisions burden the First Amendment speech rights of chief petitioners. The excluded evidence is consistent with much of the evidence that has been considered by courts addressing whether a pay-per-signature restriction violated the First Amendment. That is, the excluded evidence is relevant to show a reduction in the number of professional circulators available to gather signatures and that circulators paid hourly are less efficient, which in turn is relevant to whether the cost of petition campaigns has increased because of the restrictions. Moreover, the evidence provides context for Wolfe's contention that, since the increase in the minimum civil penalties imposed based on vicarious liability, costs of established petition companies have increased even further.10 Further, the exclusion of the evidence substantially *818prejudiced Wolfe because the evidence went to the heart of his First Amendment argument.
As to the third category, because Wolfe has failed to properly develop an argument on appeal that explains how the Walker affidavits, which assert that H.B. 2082 (2007) affected the ability of petition circulation companies and chief petitioners to operate in the state and engage in petition campaigns, fit within the framework of his argument that Article IV, section 1b, and the rules imposing vicarious liability and the minimum $2,500 civil penalty per violation burdened his speech rights, we decline to consider the relevance of those exhibits. See Beall Transport Equipment v. Southern Pacific , 186 Or. App. 696, 700 n. 2, 64 P.3d 1193, adh'd to on recons. , 187 Or. App. 472, 68 P.3d 259 (2003) ("[I]t is not this court's function to speculate as to what a party's argument might be. Nor is it our proper function to make or develop a party's argument when that party has not endeavored to do so itself."). As to the fourth category, we conclude without further explication, that the ALJ did not err in excluding that evidence because it is either irrelevant or did not substantially prejudice Wolfe.
Because the first and second categories of evidence were relevant and their exclusion substantially prejudiced Wolfe, we reverse and remand the secretary's order.
Reversed and remanded.