YES ON TERM LIMITS, INC., Robert Murphy; Sherri Ferrell; Eric Dondero Rittberg, Plaintiffs-Appellants, v. M. Susan SAVAGE, individually and in her official capacity as Oklahoma Secretary of State; W.A. Drew Edmondson, individually and in his official capacity as the Oklahoma Attorney General, Defendants-Appellees. Institute For Justice; American Civil Rights Coalition, Amici Curiae, 550 F.3d 1023

MURPHY, Circuit Judge. I.INTRODUCTION Plaintiffs brought this action pursuant to 42 U.S.C. § 1983, challenging the validity of Oklahoma’s ban on non-resident petition circulators under the First Amendment, Privileges and Immunities Clause, and Commerce Clause of the United States Constitution. Plaintiffs sought declaratory and injunctive relief. The United States District Court for the Western District of Oklahoma upheld the ban and denied Plaintiffs’ request for injunctive relief.

The district court concluded the ban survived strict scrutiny analysis under the First Amendment because it was narrowly tailored to further Oklahoma’s compelling interest in protecting the integrity of its initiative process. Plaintiffs appeal. Exercising jurisdiction pursuant to 28 U.S.C. § 1291, we hold Oklahoma’s ban on nonresident circulators does not survive strict scrutiny analysis because it is not sufficiently tailored to further Oklahoma’s compelling interest. We therefore reverse the decision of the district court and remand for proceedings consistent with this opinion. II.

BACKGROUND The Oklahoma Constitution grants its citizens the right to “propose laws and amendments to the Constitution and to enact or reject the same at the polls independent of the Legislature” through the initiative and referendum processes. Okla. Const, art. V, §§ 1-2. To place an initiative to amend the Oklahoma Constitution on the ballot, a proponent must gather signatures totaling fifteen percent of the total number of votes cast at the last general election for the state office receiving the highest number of votes. Okla. Const, art. V, § 2.

The signatures must be gathered within 90 days of the filing of the petition. Okla. Stat. Ann. tit. 34, §§ 4, 8. The proponent delivers the signatures to the Secretary of State’s office for counting. Id. §§ 4, 6. After the counting, the Secretary of State certifies to the Oklahoma Supreme Court the number of signatures collected by the proponent and the number of votes cast in the last election for the state office receiving the highest number of votes. Id. § 8.

Oklahoma citizens have the right to challenge the Secretary of State’s signature count or protest a petition by filing written notice with the Oklahoma Supreme Court within ten days of the Secretary’s publication of apparent sufficiency. Id. The Oklahoma Supreme Court ascertains whether there are enough signatures for the petition to reach the ballot. Id. Under Oklahoma law, the Secretary of State does not count signatures gathered by non-resident circulators. Okla. Stat.

Ann. tit. 34, §§ 6, 6.1 (requiring the petition circulator to swear by affidavit to be an elector in order for the gathered signatures to be counted); Okla. Const, art. III, § 1 (defining elector as “all citizens of the United States, over the age of eighteen (18) years, who are bona fide residents of this state”); see also In re Initiative Petition No. 379, 155 P.3d 32, 48 (Okla.2006) (striking all signatures gathered by nonresident circulators); In re Initiative Petition No. 365, 55 P.3d 1048, 1050 (Okla.2002) (disqualifying signatures gathered by an individual who was not a qualified elector).

In addition, non-residents who circulate petitions face criminal penalties including fines and/or imprisonment. Okla. Stat. Ann. tit. 34, § 3.1. Plaintiff Yes on Term Limits, Inc. (“YOTL”) is an Oklahoma organization seeking to place on the ballot a proposed amendment to the Oklahoma Constitution imposing term limits for various state offices. Plaintiff Robert Murphy is the vice president of YOTL and an Oklahoma resident. Plaintiffs Sherri Ferrell and. Eric Rittberg are professional petition circula-tors. Neither Ferrell nor Rittberg is a resident of Oklahoma.

YOTL and Murphy wish to hire professional circulators, including Ferrell and Rittberg, to aid in the signature gathering process. Ferrell and Rittberg claim they would work for YOTL if not for the ban on non-resident circula-tors. YOTL and Murphy contend there are not enough professional circulators who are Oklahoma residents to gather the required signatures. In addition, they contend that hiring professional, non-resident circulators is significantly more cost-efficient and effective than hiring and training resident circulators.

This is so, they argue, because professional circulators do not have to go through the training process. In addition, they claim professional circulators have greater productivity due to prior experience with the difficulties of signature gathering and strong incentives to collect valid signatures in order to remain marketable in their field. Plaintiffs filed suit in the United States District Court for the Western District of Oklahoma against M. Susan Savage, individually and in her official capacity as Oklahoma Secretary of State, and W.A.

Drew Edmondson, individually and in his official capacity as Oklahoma Attorney General. Plaintiffs challenged the constitutionality of the civil and criminal enforcement provisions of Oklahoma’s ban on nonresident circulators under the First Amendment, Privileges and Immunities Clause, and dormant Commerce Clause of the United States Constitution. The district court concluded Plaintiffs had standing to challenge the civil provisions of the ban, but lacked standing as to the criminal provisions because they could not establish injury in fact.

Thus, Plaintiffs could move forward only with their claims against the Secretary of State. The district court first considered whether the ban violates the First Amendment. It applied a strict scrutiny analysis to the ban, concluding Oklahoma had a compelling interest “in protecting and policing both the integrity and reliability of its initiative process” and the ban was narrowly tailored to meet this compelling interest. In reaching this conclusion, the district court relied heavily on Oklahoma’s evidence calling into question the integrity of certain non-resident circulators, including Rittberg.

Oklahoma presented evidence that during his career as a professional circulator, Rittberg: (1) falsely claimed to be a resident of Colorado; (2) failed to register as required in Missouri before circulating petitions in that state; and (3) was part of a four-person team of nonresident circulators in Montana who unlawfully attested to signatures gathered outside of their presence and engaged in “bait and switch” tactics. The district court also relied on evidence regarding Oklahoma’s experience with non-resident circulators in the Taxpayer Bill of Rights (“TABOR”) petition drive in 2005.

Oklahoma presented evidence that during the TABOR drive, non-resident cir-culators unlawfully participated in signature gathering. In addition, some nonresident circulators listed motel addresses as their permanent residences. The evidence demonstrated that the motels did not have residence information for a number of these non-resident circulators. Thus, they were extremely difficult for the TABOR petition protestants to locate.

Due to the motel addresses and lack of cooperation from the petition proponents and non-resident circulators, the protestants could not question many of the non-resident circulators within the ten-day protest period. The Oklahoma Supreme Court eventually invalidated the TABOR petition, citing “criminal wrongdoing and fraud” in the initiative process. In re Initiative Petition No. 379, 155 P.3d at 50. The district court found this evidence demonstrative of the questionable integrity of non-resident circulators and the difficulties of policing the petition process when non-resident circulators participate.

Thus, the court concluded the ban was necessary to protect the integrity and reliability of the petition process. Moreover, the court concluded the ban was narrowly tailored to protect the integrity of the process. The court reasoned that non-resident circula-tors have already “demonstrated a propensity to flout state laws regarding the petition process” and Oklahoma has no way to compel the non-resident circulators to return to the state for questioning.

In addition, the district court found the ban allows Oklahoma to more effectively police the petition process, since resident circulators are easily located and subject to the state’s subpoena power. Finally, the district court rejected Plaintiffs’ proposal that nonresident circulators agree to return to Oklahoma in the event of a dispute and have their gathered signatures stricken if they fail to return. According to the court, the agreement to return would be “unenforceable” and the proposed penalty of striking petition signatures would disenfranchise Oklahoma voters.

The district court then rejected Plaintiffs’ Privileges and Immunities Clause and dormant Commerce Clause claims because the ban survived the more stringent First Amendment analysis. Plaintiffs appeal to this court. III. DISCUSSION A. First Amendment Claims This court reviews a challenge to the constitutionality of a statute de novo. Powers v. Harris, 379 F.3d 1208, 1214 (10th Cir.2004). “Additionally, First Amendment cases demand our rigorous review of the record.” Chandler v. City of Arvada, 292 F.3d 1236, 1240 (10th Cir.2002). Thus, this court also reviews constitutional facts de novo. Z.J.

Gifts D-2, L.L.C. v. City of Aurora, 136 F.3d 683, 685 (10th Cir.1998). The First Amendment, made applicable to the states by the Fourteenth Amendment, provides “Congress shall make no law ... abridging the freedom of speech.” U.S. Const, amend. I. Here, Plaintiffs “seek by petition to achieve political change” and “their right freely to engage in discussions concerning the need for that change is guarded by the First Amendment.” Meyer v. Grant, 486 U.S. 414, 421, 108 S.Ct. 1886, 100 L.Ed.2d 425 (1988).

Because Oklahoma’s ban on nonresident petition circulators restricts First Amendment activity, this court must first ascertain the appropriate standard of scrutiny to apply. See Chandler, 292 F.3d at 1241. In Chandler v. City of Arvada, this court considered the validity under the First Amendment of a city ordinance banning non-residents of Arvada, Colorado, from circulating petitions within the city. Id. at 1241-44.

We stated that “petition circulation ... is core political speech, because it involves interactive communication concerning political change,” and consequently, First Amendment protection for this activity is “at its zenith.” Id. at 1241 (quotations and alteration omitted). Therefore, strict scrutiny applies “where the government restricts the overall quantum of speech available to the election or voting process ... [such as] where the quantum of speech is limited due to restrictions on ... the available pool of cir-culators or other supporters of a candidate or initiative.” Campbell v.

Buckley, 203 F.3d 738, 745 (10th Cir.2000). Like the plaintiffs in Chandler, Plaintiffs here seek to participate in petition circulation, which involves core political speech. Id. at 1241; see also Buckley v. Am. Constitutional Law Found., Inc., 525 U.S. 182, 186, 119 S.Ct. 636, 142 L.Ed.2d 599 (1999). Also as in Chandler, the state government here is limiting the quantum of this speech through its residency requirements for petition circulators. Chandler, 292 F.3d at 1241-42.

Thus, we agree with the district court that under our precedent, strict scrutiny is the correct legal standard under which to analyze Oklahoma’s ban on nonresident circulators. Id. at 1241; see also Nader v. Brewer, 531 F.3d 1028, 1036-38 (9th Cir.2008) (applying strict scrutiny to Arizona’s ban on non-resident petition cir-culators). To survive strict scrutiny, Oklahoma has the burden of proving that its ban on non-resident circulators is narrowly tailored to serve a compelling state interest. Republican Party of Minn. v. White, 536 U.S. 765, 774-75, 122 S.Ct. 2528, 153 L.Ed.2d 694 (2002).

The district court concluded Oklahoma has a “compelling interest in protecting and policing both the integrity and the reliability of its initiative process.” Assuming arguendo the district court properly identified the compelling state interest, we hold that the ban on non-resident petition circulators is not narrowly tailored to serve this interest.

The district court concluded Oklahoma’s ban on non-resident petition circulation was narrowly tailored to serve its compelling interest in protecting and policing the integrity and reliability of its petition process because (1) non-resident circulators have a demonstrated lack of integrity and propensity to flout state laws, and (2) non-residents are more difficult for those protesting signatures to locate and question.

Oklahoma first contends, and the district court agreed, that banning non-resident circulators protects the integrity and reliability of the initiative process due to the questionable integrity of non-resident cir-culators. In support of this position, the district court relied on evidence of the fraudulent practices of a handful of nonresident petition circulators, including Plaintiff Rittberg. That evidence alone, however, does not support the inference that, as a class, non-resident circulators are more likely to engage in fraud than resident circulators.

See Buckley, 525 U.S. at 204 n. 23, 119 S.Ct. 636 (“While testimony in the record suggests that occasional fraud in Colorado’s petitioning process involved paid circulators, it does not follow like the night the day that paid circulators are more likely to commit fraud and gather false signatures than other cir-culators.”) (citations omitted).

Moreover, Plaintiffs presented evidence that non-resident professional petition circulators collect higher percentages of valid signatures than resident volunteers or inexperienced workers because their livelihood as professional circulators depends upon their reputation for effective signature collection. The district court apparently did not give any weight to this evidence, and, more importantly, was unable to compare the prevalence of fraudulent activity of nonresident circulators as a class with that of resident circulators as a class because Oklahoma provided no data to this effect.

As a consequence, the record does not support the district court’s conclusion that non-resident circulators as a class engage in fraudulent activity to a greater degree than resident circulators. See id,.; see also Meyer, 486 U.S. at 426, 108 S.Ct. 1886 (“[W]e are not prepared to assume that a professional circulator — whose qualifications for similar future assignments may well depend on a reputation for competence and integrity — is any more likely to accept false signatures than a volunteer who is motivated entirely by an interest in having the proposition placed on the ballot.”).

This court therefore concludes Oklahoma has failed to prove the ban is narrowly tailored to protect the initiative process due to a higher rate of non-resident circulator fraud. The district court also concluded the ban was narrowly tailored to protect the integrity of the initiative process due to the difficulty of locating and questioning nonresident circulators within the ten-day protest period. The Ninth Circuit recently addressed similar issues in the context of reviewing Arizona’s residency requirement for petition circulators. Nader, 531 F.3d at 1037.

In Nader, Arizona argued its ban on non-resident petition circulators was “narrowly tailored to ensure that circula-tors are subject to the state’s subpoena power, and that the state can locate them within the ten-day period allotted for petition challenges.” Id. The Ninth Circuit rejected this argument, reasoning that “[fjederal courts have generally looked with favor on requiring petition circulators to agree to submit to jurisdiction for purposes of subpoena enforcement, and the courts have viewed such a system to be a more narrowly tailored means than a residency requirement to achieve the same result.” Id.

The court then explained that Arizona had submitted insufficient evidence to support its contention that this system would be unworkable, and thus the ban violated the First Amendment. Id. Oklahoma contended, and the district court agreed, that an approach similar to that favored by the Ninth Circuit, requiring petition circulators to sign a written agreement to return to Oklahoma should a protest arise, was an ineffective alternative because such agreements are unenforceable contracts between the circulator and proponent and Oklahoma does not have subpoena power over non-residents.

The district court accepted Oklahoma’s assertion that striking the signatures gathered by circulators who fail to return for questioning was unacceptable because it would punish and disenfranchise Oklahoma voters who had the misfortune of signing a non-resident circulator’s petition. On these bases, the district court upheld the blanket ban on non-resident petition circu-lators as narrowly tailored to further a compelling interest.

Even if Oklahoma adequately established its contentions that the ability to question non-resident circulators during the protest periods is necessary to prevent fraud and that non-resident circulators are more difficult to locate and question, it failed to prove the ban is narrowly tailored. Oklahoma could require that in order to circulate petitions, non-residents enter into agreements with the state, rather than the initiative proponent, wherein the circulators provide their relevant contact information and agree to return in the event of a protest. See Chandler, 292 F.3d at 1242-44.

In addition, Oklahoma could provide criminal penalties for circulators who fail to return when a protest occurs. Oklahoma contends such agreements would be more difficult and costly to enforce than a resident subpoena. Even if true, Oklahoma has not proved that, as a class, non-resident petition circulators who sign such agreements are less likely to submit to questioning than residents. Therefore, requiring non-residents to sign agreements providing their contact information and swearing to return in the event of a protest is a more narrowly tailored option that Oklahoma has failed to prove would be ineffective.

Ashcroft v. ACLU, 542 U.S. 656, 665, 124 S.Ct. 2783, 159 L.Ed.2d 690 (2004) (“[T]he burden is on the Government to prove that the proposed alternatives will not be as effective as the challenged statute.”); see also Krislov v. Rednour, 226 F.3d 851, 866 n. 7 (7th Cir.2000) (holding a residency requirement for circulators unconstitutional under the First Amendment and suggesting a state may legitimately ensure the integrity of the process through a requirement that non-residents agree to submit to the state’s jurisdiction).

Oklahoma has failed to prove the ban on non-resident circulators is narrowly tailored to protect the integrity of the initiative process. The evidence presented by Oklahoma and relied upon by the district court consisted of the allegedly fraudulent or uncooperative practices of a handful of non-resident circulators. From this limited evidence, the district court made unwarranted conclusions about non-resident circulators as a class.

Because the record contains insufficient evidence to conclude that non-residents, as a class, threaten the integrity or reliability of the initiative process, Oklahoma has failed to prove that banning all non-resident circulators is a narrowly tailored means of meeting its compelling interest. Oklahoma has also failed to prove the ineffectiveness of plausible alternatives to the blanket ban on non-residents. Oklahoma’s ban on nonresident circulators therefore violates the First and Fourteenth Amendments of the United States Constitution. B.

Alternative Constitutional Claims Because Oklahoma’s ban on non-resident circulators violates the First and Fourteenth Amendments, this court need not decide whether it also violates the Privileges and Immunities Clause or dormant Commerce Clause. IV. CONCLUSION For the reasons discussed above we reverse the decision of the district court and remand for further proceedings consistent with this opinion. .

These tactics included telling individuals they had to sign three copies of the petition they wished to support, when in fact they were signing one copy of that petition and two separate petitions in support of non-related issues. . The district court did not address the other compelling interest proposed by Oklahoma, i.e., "restricting the process of self-government to members of its own political community.” Oklahoma correctly contends the Supreme Court has recognized a state's interest in restricting the right to vote or hold office to residents. Supreme Court of N.H. v.

Piper, 470 U.S. 274, 282 n. 13, 105 S.Ct. 1272, 84 L.Ed.2d 205 (1985) ("A State may restrict to its residents, for example, both the right to vote, and the right to hold state elective office." (citation omitted)). Oklahoma, however, provides no case law supporting the proposition that states may restrict non-resident speech, such as petition circulation, simply because the speech may indirectly affect the political process through the solicitation of resident participation. Supreme Court precedent seems to indicate there is no compelling interest in restricting such speech. See Meyer v.

Grant, 486 U.S. 414, 424-28, 108 S.Ct. 1886, 100 L.Ed.2d 425 (1988) (holding Colorado’s ban on paid petition circulators unconstitutional and stating that while Colorado could wholly ban initiatives, it could not ban the speech of a class of circulators). To accept the wholesale restriction of the petition process to residents of Oklahoma as a compelling state interest would have far-reaching consequences. For example, the prohibition of non-residents from driving voters to the polls would seemingly be a logical extension.

This court is unwilling to approve as a compelling state interest the restriction of core First Amendment rights in this manner. Under the circumstances of this case, we reject Oklahoma's broad purpose of "restricting the process of self-government to members of its own community” as a compelling interest in the context of interdicting non-resident circu-lators. . The validity of these contentions is far from clear. It was not obvious from the record that the ability to question circulators after a petition is submitted significantly aids in protecting the integrity of the initiative process.

In addition, the bulk of Oklahoma’s evidence on the difficulty of locating and questioning non-resident circulators again consisted of information about the practices of only a handful of non-resident circulators who were difficult to locate or uncooperative in the past. As discussed above, this evidence is insufficient to prove non-resident circulators as a class are more difficult to locate and question than resident circulators. . The Sixth Circuit recently addressed the constitutionality of Ohio’s ban on non-resident circulators. Nader v. Blackwell, 545 F.3d 459 (6th Cir.2008).

There, the court held the ban violated the First Amendment, but the right was not clearly established for purposes of qualified immunity. Id. at 476-78. . Plaintiffs also appeal the denial of their claims challenging the validity of the criminal provision of the ban. Because the ban is unconstitutional, we need not address the criminal provision. In addition, Defendants submitted a motion to strike certain evidence introduced by Plaintiffs for the first time on appeal. Because Plaintiffs did not present this evidence before the district court, the motion to strike is granted.