Riori Tampubolon (“Tampubolon”) and his wife, Erlinda Silitonga (“Silitonga”) petition for review of the Board of Immigration Appeals (“BIA”) decision affirming the immigration judge’s (“IJ”) denial of their applications for asylum, withholding of removal, and cancellation of removal from Indonesia. We deny the petition with respect to the asylum and cancellation of removal claims, but we grant the petition with respect to the withholding of removal claim. We hold that the BIA erred in failing to apply disfavored group analysis to petitioners’ withholding claim because the record compels a finding that Christians in Indonesia are a disfavored group. We remand in accordance with INS v. Orlando Ventura, 537 U.S. 12, 16 , 123 S.Ct. 353 , 154 L.Ed.2d 272 (2002).

I. BACKGROUND

Petitioners, devout Protestant Christians, are natives and citizens of Indonesia. In 1989, when he was 31 years old, Tampubolon entered the United States on a tourist visa. Likewise, in 1992, when she was 22 years old, Silitonga entered the United States on a tourist visa. Tampubolon and Silitonga met in the United States and married in 1995. They have two U.S. citizen daughters, ages 12 and 14. The entire family is active in their local church. They attend weekly services, Bible study, and prayer meetings. Silitonga teaches Sunday School, and the children sing in the choir. Tampubolon and Silitonga are both employed in the healthcare industry and pay taxes; neither has been arrested for any crime.

Petitioners came to the attention of the Department of Homeland Security (“DHS”) when they conformed to the National Security Entry-Exit Registration System (“NSEERS”) and registered with Immigration and Customs Enforcement. In 2003 and 2004, respectively, DHS issued Notices to Appear to Tampubolon and Silitonga, charging them with removability for overstaying their visas. Tampubolon and Silitonga conceded removability and applied for relief from removal, including cancellation of removal, asylum, withholding of removal, and protection under CAT.

The IJ denied all applications for relief. The IJ denied petitioners’ application for asylum because they failed to file their applications within one year of arrival, and also failed to demonstrate changed circumstances. The IJ denied petitioners’ application for withholding of removal because neither had suffered past persecution and both had similarly situated siblings living in Indonesia who had not experienced problems practicing their Christian faith. The IJ denied petitioners’ application for cancellation of removal because they failed to demonstrate that removal would result in exceptional and extremely unusual hardship to their two U.S. citizen daughters. The BIA adopted and affirmed the decision of the IJ, citing Matter of Burbano, 20 I. & N. Dec. 872 (B.I.A.1994). Petitioners timely petitioned for review. We have jurisdiction over the petition pursuant to 8 U.S.C. § 1252 (a).

II. STANDARD OF REVIEW

Where, as here, the BIA cites its decision in Burbano, and does not express disagreement with any part of the IJ’s decision, we review the IJ’s decision as if it were the BIA’s decision. See Cinapian v. Holder, 567 F.3d 1067, 1073 (9th Cir.2009). The BIA’s determination that petitioners have not established eligibility for asylum or withholding of removal is reviewed for substantial evidence. See, e.g., Zehatye v. Gonzales, 453 F.3d 1182, 1184-85 (9th Cir.2006).

Under the substantial evidence standard, the BIA’s determinations will be upheld “if the decision is ‘supported by reasonable, substantial, and probative evidence on the record considered as a whole.’ ” Zhao v. Mukasey, 540 F.3d 1027, 1029 (9th Cir.2008) (quoting Abebe v. Gonzales, 432 F.3d 1037, 1039-40 (9th Cir.2005) (en banc)). Reversal, however, is appropriate when “the evidence in the record compels a reasonable factfinder to conclude that the [BIA’s] decision is incorrect.” Id.

Purely legal questions, including jurisdictional questions, are reviewed de novo. See Taslimi v. Holder, 590 F.3d 981, 984 (9th Cir.2010).

III. ASYLUM

To qualify for asylum, an applicant must file her application within one year after arrival in the United States. 8 U.S.C. § 1158 (a)(2)(B). The limitations period will be tolled if the applicant can establish changed circumstances that materially affect her eligibility for asylum. 8 U.S.C. § 1158 (a)(2)(D); 8 C.F.R. § 208.4 (a)(4)(i). Contrary to the government’s argument, we have jurisdiction to review an agency’s changed circumstances determination. See Ramadan v. Gonzales, 479 F.3d 646, 648 (9th Cir.2007) (per curiam) (holding that the Real ID Act restored jurisdiction over the “changed circumstances” question because this question involved the application of a statutory standard to undisputed facts).

Petitioners argue that Sael v. Ashcroft, 386 F.3d 922 (9th Cir.2004), and Lolong v. Gonzales, 400 F.3d 1215 (9th Cir.2005) (“Lolong I”), rev’d, 484 F.3d 1173 (9th Cir.2007) (en banc) (“Lolong II”), constituted changed circumstances because they changed United States law in a way that materially affects their eligibility for asylum. 8 C.F.R. § 208.4 (a)(4)(i)(B). We review the IJ’s changed circumstances determination for substantial evidence. See Ramadan, 479 F.3d at 657 . Petitioners’ argument fails because both Sael and Lolong were decided after petitioners filed their asylum applications. Accordingly, those decisions could not have tolled the one-year statute of limitations. We affirm the IJ’s denial of petitioners’ asylum application as untimely.

IV. WITHHOLDING OF REMOVAL

Unlike asylum, there is no statutory time limit for filing a withholding application. Himri v. Ashcroft, 378 F.3d 932, 937 (9th Cir.2004) (citing 8 U.S.C. § 1231 (b)(3)). Accordingly, petitioners’ application for withholding of removal is not time-barred. Petitioners argue that the BIA erred by failing to apply disfavored group analysis to their withholding claim because Christians are a disfavored group in Indonesia. We agree and remand to the BIA to use the disfavored group analysis in determining whether petitioners are entitled to withholding of removal.

A. Christians Are a Disfavored Group in Indonesia

A “disfavored group” is “a group of individuals in a certain country or part of a country, all of whom share a common, protected characteristic, many of whom are mistreated, and a substantial number of whom are persecuted” but who are “not threatened by a pattern or practice of systematic persecution.” Wakkary, 558 F.3d at 1052, 1062 (brackets and internal quotation marks omitted). Although we held in Sael, 386 F.3d at 927 , that the ethnic Chinese are a disfavored group in Indonesia, and suggested in Wakkary, 558 F.3d at 1063, that Chinese Christians are disfavored in Indonesia, we have never determined whether Christians who are not Chinese are a disfavored group in Indonesia.

In Sael , we held that Indonesia’s ethnic Chinese minority are a “disfavored group” because of Indonesia’s long history of anti-Chinese violence and official discrimination. 386 F.3d at 927 . We noted that official policies of ethnic tolerance and decreased numbers of racially motivated attacks did not diminish the “disfavored” status of ethnic Chinese because official discrimination continued and Chinese-Indonesians experienced “centuries of persecution.” Id. at 929 . Here, petitioners have also submitted compelling evidence that Christians in Indonesia are subject to violence and official discrimination.

Indonesia is the world’s most populous Muslim country. Nearly 90% of Indonesians are Muslim, while Protestant Christians only account for 6 % of the population. According to the 2003 U.S. State Department Country Report on Human Rights Practices in Indonesia (“Country Report”), “[t]he Government ... frequently failed to protect adequately the fundamental rights ... of religious minorities.”

Muslims and Christians lived together in relative peace until the 1990s when President Suharto began courting militant Islamic groups to maintain his political power. See Mieke Kooistra, Minority Rights Group Int'l, Indonesia: Regional Conflicts and State Terror 14 (2001). Consequently, militant Islam increased in strength and political influence. Id. During this time, Suharto purged his cabinet and army of Christians and replaced them with fundamentalist Muslims. Id. Even after Suharto’s regime ended, the military and political elite continued to protect violent Muslim militia groups, such as Laskar Jihad, whose goal is to kill, convert, or drive out all non-Muslims from certain parts of Indonesia. Id. at 20; see also Paul Marshall, Jihad Comes to Indonesia, The Weekly Standard, Dec. 31, 2001, at 20.

The Indonesian government’s support of, or at the very least, acquiescence in, militant expressions of Islam has subjected Christians to violent persecution in Indonesia. The record demonstrates that Christian churches throughout Indonesia have been burned, bombed, and vandalized by Muslim extremists. These attacks are often accompanied by threats, such as: “God has no son. Jesus could not help you. Until doomsday, Muslims will not make peace with Christians. Death to all Christians.” See Seth Mydans, Jihad Seethes, and Grows, on Indonesian Island, N.Y. Times, Jan. 10, 2002, at A3. Petitioners’ family members who remain in Indonesia have been affected by these attacks. For example, Silitonga testified that Muslim extremists threatened to kill people who attended worship service at her sister’s church in Sumatra on Christmas Eve in 2000. According to the 2003 U.S. Department of State International Religious Freedom Report on Indonesia (“Religious Freedom Report”), at least 25 churches were destroyed in 2003. Although mosques are also subject to attack, only one mosque was destroyed during the same reporting period.

Christians living in some areas, such as the Maluku Islands, are subject to even greater violence, such as forced conversions and ritual circumcisions. See, e.g., Catherine Philp, Indonesian Christians Butchered in Islamic ‘Crusades’, The Scotsman, July 8, 2001, at 18. In these areas, Christian villages are attacked and Christians are forced to flee their homes. Another concern is that Christians, who like all Indonesians must carry identification cards denoting their religious affilia tion, are murdered at unofficial identification checkpoints.

Laskar Jihad is responsible for much of the Christian-targeted violence in Indonesia. The activities of this paramilitary group escalated the bilateral Muslim-Christian tensions to a one-sided religious cleansing and slaughter of Christians between 1998 and 2002. Marshall, supra at 20. According to local eyewitnesses, the military participates in attacks on churches, and Indonesians widely believe that the government cannot or will not protect Christians against Muslim attacks. See, e.g., Indonesian Military Deny Allegations Soldiers Took Part in Church Attack, Associated Press, Apr. 29, 2002 (witnesses stated that uniformed soldiers participated in an attack on a church); Philp, supra at 18 (noting that the government’s reaction to forced conversions and circumcisions was “astonishingly weak”); Kooistra, supra at 20 (noting that “it is obvious that [Laskar Jihad] is well-armed, well-financed and well-protected by the military or powerful people with connections to the former ruling elite”). Despite reductions of interreligious violence, the Religious Freedom Report notes that the government failed to hold accountable religious extremists who killed and terrorized Christians, even as it fully prosecuted and punished Christian leaders for subversion.

In addition to condoning or turning a blind eye to the persecution of Christians by private individuals, evidence in the record indicates that the Indonesian government discriminates against Christians. For example, according to the Religious Freedom Report, Christians complained that the government made it harder for them to build houses of worship than Muslims. Religious minorities also contend that they are excluded from prime civil service postings and slots at public universities.

Christian Indonesians have also suffered private discrimination and marginalization by the general populace. For example, the 2003 Country Report noted that estates housing only Muslims have grown increasingly popular. Moreover, according to a 2002 survey in Jakarta, 79.6% of respondents believed that the government should outlaw groups that follow faiths other than Islam, 72.5% believed that members of minority religions should not be permitted to teach in public schools, 58% support an Islamic state, and 42% would not allow churches to be built in their neighborhood. See Devi Asmarani, Many Muslim Indonesians Say They Want an Islamic State, The Straits Times (Singapore), Jan. 2, 2002, at 1-2.

Christian Indonesians are targeted for violence and experience official and unofficial discrimination in many facets of their lives. Although the record demonstrates that the Indonesian government does not promulgate facially discriminatory laws against Christians, this fact does not address its discriminatory enforcement of facially neutral laws or its acquiescence in widespread private discrimination and violence against Christians. See Sael, 386 F.3d at 929 . The record demonstrates that Christian Indonesians are mistreated, and some are subject to persecution. See Wakkary, 558 F.3d at 1052. Accordingly, any reasonable factfinder would be compelled to conclude on this record that Christian Indonesians are a disfavored group.

B. Disfavored Group Analysis in the Withholding Context

As we recognized in Wakkary, disfavored group analysis is an evidentiary concept that applies when a petitioner at tempts to show that she will be individually singled out for persecution. Id. at 1063. Evidence of both individual and group targeting are relevant to demonstrate the likelihood that a particular individual will be persecuted. Id. at 1062-63. Therefore, “[t]he more evidence of group targeting an ... applicant proffers, the less evidence of individually specific evidence [s]he needs to [satisfy her ultimate burden of proof].” Id. at 1064. Nevertheless, a petitioner’s membership in a disfavored group is not sufficient by itself to meet her ultimate burden of proof; “some evidence of individualized risk is necessary for the petitioner to succeed.” Id. at 1065.

Here, the BIA failed to analyze petitioners’ withholding claim according to disfavored group analysis. Because the record compels the conclusion that Christians in Indonesia are a disfavored group and petitioners are Christians, we must remand to the BIA for it to determine whether the combination of disfavored group evidence and evidence of individualized risk is sufficient to establish a clear probability that petitioners will be persecuted if removed to Indonesia. Wakkary, 558 F.3d at 1067.

V. CANCELLATION OF REMOVAL

The IJ denied petitioners’ application for cancellation of removal because petitioners did not demonstrate that their removal would result in exceptional and extremely unusual hardship to their two U.S. citizen daughters. See 8 U.S.C. § 1229b(b)(1)(D). This court lacks jurisdiction to review an agency’s “exceptional and extremely unusual hardship” determination. See Romero-Torres v. Ashcroft, 327 F.3d 887, 888 (9th Cir.2003) (holding that the “ ‘exceptional and extremely unusual hardship’ determination is a subjective, discretionary judgment that has been carved out of our appellate jurisdiction”). This court retains jurisdiction, however, to review due process challenges and other questions of law. See, e.g., Martinez-Rosas v. Gonzales, 424 F.3d 926, 930 (9th Cir.2005). We review our own jurisdiction de novo. Taslimi, 590 F.3d at 984 .

Petitioners argue that the BIA violated their due process rights by failing to consider the effect of Sael and Lolong I. The BIA violates an alien’s due process rights if: “(1) the proceeding was so fundamentally unfair that the alien was prevented from reasonably presenting his case, and (2) the alien demonstrates prejudice, which means that the outcome of the proceeding may have been affected by the alleged violation.” Ibarra-Flores v. Gonzales, 439 F.3d 614, 620-21 (9th Cir.2006) (citations and internal quotation marks omitted).

We decided Sael and Lolong I after the IJ rendered his decision. Both cases involved applications for asylum and neither dealt with cancellation of removal. Accordingly, neither case changed the law for cancellation of removal. The BIA’s failure to address two irrelevant cases does not render the proceeding fundamentally unfair. Because petitioners fail to state a colorable due process claim, we lack jurisdiction to review the IJ’s denial of petitioners’ application for cancellation of removal.

VI. CONCLUSION

In summary, we hold that on this record, Christians are a disfavored group in Indonesia and conclude that the BIA erred in failing to analyze petitioners’ withholding claim according to disfavored group analysis. Therefore, we grant the petition for review and remand for the BIA to apply disfavored group analysis. We affirm the BIA and deny the petition with respect to all other applications for relief. Costs on appeal awarded to petitioners.

PETITION GRANTED IN PART, DENIED IN PART, AND REMANDED.

. "Special Registration” refers to the National Security Entry-Exit Registration System ("NSEERS”). Following September 11, 2001, the Attorney General instituted NSEERS, which required the collection of data from aliens upon entry and periodic registration of certain aliens, from majority-Muslim states and North Korea, present in the United States. See Rajah v. Mukasey, 544 F.3d 427, 433 (2d Cir.2008); see also 8 C.F.R. § 264.1 (f)(4) (2003) (enabling regulation for the NSEERS program). It appears that NSEERS is used to identify undocumented immigrants and penalize those who voluntarily comply with the law. . Neither the IJ nor the BIA made an adverse credibility finding. Accordingly, this court must take petitioners' testimony as true. See Kalubi v. Ashcroft, 364 F.3d 1134, 1137 (9th Cir.2004). . In their opening brief, petitioners failed to raise any issues or advance any arguments in favor of relief under CAT. Accordingly, petitioners waived review of their CAT claim. See, e.g., Husyev v. Mukasey, 528 F.3d 1172, 1183 (9th Cir.2008) (concluding that the petitioner waived CAT claim by failing to advance any arguments in favor of relief); Martinez-Serrano v. INS, 94 F.3d 1256, 1259-60 (9th Cir.1996) (concluding that petitioner waived a claim that was referenced in the statement of the case, but not discussed in the body of the opening brief). . Furthermore, neither Sael nor Lolong I materially affected petitioners' eligibility for asylum. The panel's decision in Lolong I is not precedential because it was vacated, reheard en banc, and not adopted in Lolong II. See 9th Cir. R. 35-3, Advisory Note 3 ("The three-judge panel opinion shall not be cited as precedent ..., except to the extent adopted by the en banc court.”). Moreover, the holding in Sael v. Ashcroft, 386 F.3d 922 (9th Cir.2004), that the ethnic Chinese minority in Indonesia are a disfavored group, does not apply here because petitioners are not ethnically Chinese. In addition, disfavored group analysis existed well before we decided Sael. See Wakkary v. Holder, 558 F.3d 1049, 1062-63 (9th Cir.2009) (reviewing the development of "disfavored group" analysis beginning with Kotasz v. INS, 31 F.3d 847 (9th Cir.1994)). Therefore, Sael does not constitute a changed circumstance that materially affects petitioners' eligibility for asylum. . Two avenues are available for establishing eligibility for withholding of removal: (1) past persecution, which creates a presumption of eligibility for withholding of removal; and (2) a clear probability of future persecution. See 8 C.F.R. § 208.16 (b); see also Al-Harbi v. INS, 242 F.3d 882, 888-89 (9th Cir.2001) (discussing the "clear probability standard” for withholding of removal). A clear probability of future persecution may be established in one of two ways. First, the applicant may demonstrate that it is more likely than not that she will be singled out individually for persecution on account of a protected ground. See 8 C.F.R. § 1208.16 (b)(2). Second, in the absence of any individualized targeting, the applicant may demonstrate that she is a member of a group that is subject to a "pattern or practice of persecution” in her home country. Wakkary, 558 F.3d at 1060 . Disfavored group analysis is triggered only when a petitioner attempts to show that she will more likely than not be singled out individually for persecution on account of a protected ground. That is the precise situation in this case. Here, petitioners concede that they did not experience past persecution and do not argue that Christians in Indonesia are subject to a pattern or practice of persecution. Rather, petitioners rely on the "individualized risk of persecution” avenue to establish eligibility for withholding. . The record reflects that petitioners’ family members continue to live in Indonesia, but that unlike their family members, petitioners would return Westernized to Indonesia after living in the United States for over twenty years and raising two U.S. citizen children.