This case consolidates two appeals, both arising from the Little Rock School District's request for unitary status. First, the Joshua Intervenors1 appeal from the District Court's2 denial of their Motion for Recusal of District Judge and Vacating of Orders, Rulings, and Judgments. We review a district court's denial of recusal for abuse of discretion. See In re Hale, 980 F.2d 1176, 1178 (8th Cir.1992); United States v. Walker, 920 F.2d 513, 516 (8th Cir.1990). We conclude that Judge Wilson's representation of Judge Henry Woods at a much earlier stage of the case, and on far different issues, did not involve the same "matter in controversy" for purposes of 28 U.S.C. § 455(b)(2); thus, we affirm the denial of the Joshua Intervenors' Motion for Recusal.

The Joshua Intervenors also appeal from the District Court's judgment granting the Little Rock School District (LRSD) partial unitary status. The Joshua Intervenors assert: (1) that the District Court erred by not requiring and considering additional reports from the Office of Desegregation Monitoring (ODM); and (2) that the District Court's finding of substantial compliance with the Revised Desegregation and Education Plan was erroneous. We hold that the District Court did not err by failing to require new written reports from the ODM, and that the District Court's findings of fact are not clearly erroneous; thus, we affirm the grant of partial unitary status.

Because the facts relevant to each issue on appeal are different, we address them separately. In Part I, we address the issue of disqualification. In Part II, we address whether the District Court should have required new written reports from the ODM. Finally, in Part III, we address whether the District Court erred in finding that LRSD substantially complied with the Revised Plan in most respects.

This litigation began in 1982 and has been in and out of this Court and the District Court several times — it is complex to say the least. We briefly highlight the events relevant to the issue of the disqualification of Judge Wilson.

In 1987, LRSD and the Joshua Intervenors sought to disqualify Judge Henry Woods,3 who was then presiding over the case. The parties asserted as grounds for disqualification that during Judge Woods's private law practice, one of his partners had represented parties who participated as amici curiae in a related case, and that Judge Woods's impartiality was called into question by his comments at a meeting with students. Judge Wilson, then in private practice, represented Judge Woods for the limited purpose of the mandamus proceedings, defending Judge Woods's decision not to recuse himself.4

In the current proceeding, begun by LRSD's motion that it be released from court supervision, the Joshua Intervenors sought the recusal of Judge Wilson under 28 U.S.C. § 455(b)(2), which requires a judge to disqualify himself "where in private practice he served as lawyer in the matter in controversy." After Judge Wilson entered an order on September 13, 2002, granting LRSD partial unitary status, the Intervenors filed a Motion for a Hearing Regarding the Relevance of 28 U.S.C. § 455 to the Present Proceedings. Judge Wilson denied this motion on October 29, 2002. Little Rock Sch. Dist. v. Pulaski County Special Sch. Dist. No. 1, 2002 WL 31465311 (E.D.Ark.2002). Thereafter, on November 25, 2002, the Joshua Intervenors moved for disqualification of Judge Wilson. Judge Wilson denied this motion because, among other reasons, he had never served, in his view, as a lawyer in the "matter in controversy." Little Rock Sch. Dist. v. Pulaski County Special Sch. Dist. No. 1, No. 4:82CV00866 (E.D.Ark. Dec. 20, 2002). The Joshua Intervenors appeal.

We must determine whether Judge Wilson's representation of Judge Woods in the mandamus proceeding in 1987 involved the same "matter in controversy" as the present questions before us for purposes of 28 U.S.C. § 455(b)(2). Because the mandamus proceeding did not touch upon the merits of the case, we conclude that it was not a part of the same "matter in controversy."

The Joshua Intervenors contend that Judge Wilson's participation was part of the same matter in controversy because it was part of a single case. The language chosen by Congress, "matter in controversy," is not defined by the statute. However, Congress easily could have substituted the word "case" for the words "matter in controversy," but did not do so. This deliberate choice by Congress demonstrates an intent that the words "matter in controversy" mean something other than what we commonly refer to as a "case." In fact, Congress used the words "proceeding," "case in controversy," and "subject matter in controversy" in various other subsections of § 455(b) to describe situations where a judge must disqualify himself. Thus, we must assume that Congress ascribed a particular meaning to the words "matter in controversy," and we must try to discern that meaning.

We note that Judge Wilson represented Judge Woods at the mandamus proceedings, which were given a separate docket number from the rest of the case in this Court. This circumstance, though relevant, is not enough in itself to enable us to conclude that the disqualification proceeding was not the same "matter in controversy" as the present appeal. As we have indicated, the phrase "matter in controversy" must mean something other than the word "case," and so we do not rely on this technical distinction. Instead, we look to the substance of the issues argued and decided in the two proceedings.

In Little Rock Sch. Dist. v. Pulaski County Special Sch. Dist. No. 1, 839 F.2d 1296 (8th Cir.1988), we discussed, but did not decide, whether a matter in controversy could extend beyond a single case. Even if a matter in controversy could be more extensive than a single case, we concluded that the facts before us did not support such a conclusion because the cases involved, "to a large extent, different issues and different remedies." Id. at 1302. We think this reasoning is useful in determining whether a matter in controversy may be less extensive than a case.

Judge Wilson's representation of Judge Woods was restricted solely to the issue of recusal and did not go to the merits of the case. Judge Wilson was involved in the case solely for the mandamus proceedings and, in the course of his representation, never addressed the merits of the case or expressed any opinion about them. The issues before Judge Wilson in the present matter are wholly unrelated to his prior representation of Judge Woods.

Although the case law is slim in this area, we find support for our position in In re Apex Oil Co., 981 F.2d 302 (8th Cir.1992). In Apex Oil, Judge Loken found his recusal unnecessary where he and his former law firm were previously involved with plaintiffs' claim for damages from an oil spill and where, later, his law firm filed claims on behalf of plaintiffs in Apex Oil's bankruptcy proceedings. Id. at 304-05. The question was whether the plaintiffs' claim for damages constituted the same matter in controversy as the later claims in bankruptcy when both resulted from the same oil spill. Id. at 303. Although acknowledging that bankruptcy proceedings are atypical because they are conducted under an umbrella proceeding, Judge Loken concluded that the cases were not "sufficiently related" so as to constitute the same matter in controversy. Id. at 304. Applying this analysis to our situation, we conclude that there is not a sufficient relationship between the recusal proceedings with respect to Judge Woods and the issues now before us on the merits to make them the same "matter in controversy." Nor do we think that any impartial observer could reasonably think that Judge Wilson's impartiality should be called into question. Not only was his prior representation of Judge Woods wholly distinct; the issues before the Judge in the current proceeding involved the current version of the parties' agreement to settle the underlying case, an agreement that was never before Judge Woods, and that was not even in existence until long after he voluntarily relinquished the case.

As we have noted, this appeal arises from an interdistrict desegregation case filed by LRSD in 1982. As part of that case, the parties agreed to a settlement plan in 1989. However, as time passed, portions of that plan proved unworkable, and the parties agreed to the Revised Desegregation and Education Plan. This plan was approved by the District Court and this Court.

On March 15, 2001, LRSD asked the District Court to declare it unitary under § 11 of the Revised Plan. On July 25, 2001, the Joshua Intervenors filed an opposition to this request. The opposition, App. of Appellants 185-86, made the following argument, among many others:

The Joshua Intervenors believe further that the Court must have before it a written response to the district's plan or other written analysis regarding that plan from the Court's Office of Desegregation Monitoring (ODM) before the Court can issue a final opinion regarding the matter. Otherwise, any assessment by the Court would be incomplete and not keeping with the expectations of the Eighth Circuit Court of Appeals when it required the establishment of the ODM to assist the Court in determining and effectuating desegregation compliance.

This opposition was filed while the case was still before Chief Judge Wright (who had taken the case after Judge Woods had removed himself from it). She then conducted five and one-half days of evidentiary hearings, ending on November 20, 2001. On January 3, 2002, Chief Judge Wright withdrew from the case, and it was reassigned to Judge Wilson. He held three additional days of evidentiary hearings on July 22, 23, and 24, 2002.

The Joshua Intervenors' second major argument on appeal is that the District Court erred in making findings and entering judgment without directing ODM to prepare additional monitoring reports on LRSD's compliance with the Revised Plan. The Joshua Intervenors point out that the ODM was created in the first place at the direction of this Court. See Little Rock Sch. Dist. v. Pulaski County Special Sch. Dist. No. 1, 921 F.2d 1371, 1388 (8th Cir. 1990).

The District Court had before it some relevant materials from the ODM: a report on LRSD's preparations for implementation of the Revised Plan, filed August 11, 1999, and a report of disciplinary sanctions in the Little Rock School District, filed on June 14, 2000. As to the first report, the Court observed that it "indicated that, overall, LRSD was doing a satisfactory job of implementing the Revised Plan." Little Rock Sch. Dist. v. Pulaski County Special Sch. Dist. No. 1, 237 F. Supp. 2d 988, 1048 (E.D.Ark.2002). The District Court did not view the second report as having much value. The Joshua Intervenors argue that the District Court should have had the ODM prepare an additional report or reports before making any findings. They point out that "ODM had gained considerable expertise, preparing at least 49 reports." Brief for Appellants 39.

In response, LRSD argues that this point was not properly raised in the District Court. Certainly it is true that the Court never entered a written order expressly disposing of the request that additional monitoring reports be prepared. Before the case was transferred to Judge Wilson, however, Chief Judge Wright effectively denied the Intervenors' request, saying:

And of course, you are free, Mr. Walker, to call the Office of Desegregation Monitoring as witnesses, as well, I mean, those people as witnesses to the extent you think they have knowledge on the matters at issue. And furthermore, and I talked with Ms. Marshall [the head of ODM] about this, I don't mind Ms. Marshall telling you, sharing with you the information that she has, but if she does that I want her to share it with everyone else too.

Tr. of June 29, 2001, 27-28. The Joshua Intervenors, in response to this invitation or otherwise, did not call anybody from the ODM as a witness.

As we have noted, the request that additional monitoring reports be required was not the subject of a separate motion, but rather a matter mentioned, almost in passing, in a pleading filed by the Joshua Intervenors. App. of Appellants 185-86. As far as we can tell, the request was never renewed on the record, either in writing or in open Court, during the days of evidentiary hearings conducted by Judge Wilson, or in any other manner. We nevertheless assume for present purposes that the point is properly before us, and we hold that it is without merit. The ODM, as the Joshua Intervenors point out, was created at the direction of this Court, at the time of our initial approval of the settlement agreement, but the ODM was to be under the supervision of the District Court and to act as an arm of that Court in ensuring that the settlement agreement was followed. It was and remains the job of the District Court, in its discretion, to determine how the ODM should be used. A choice to rely on the existing materials prepared by the ODM, and to eschew the preparation of additional reports, is certainly not an abuse of discretion. Two further points are important. First, the Joshua Intervenors could have, but did not, call someone from the ODM to testify. Second, no offer of proof was made. We do not know what ODM's position would have been if it had been asked. In this situation, it is simply impossible to say that the decision not to request the production of additional papers had any effect on the outcome of this case.

The Revised Plan "supersede[s] and extinguish[es] all prior agreements and orders" in the case, with limited exceptions. App. of Appellants 87. Unlike the previous settlement agreement, the Revised Plan contains a specific procedure by which LRSD can attain unitary status. Section 11 of the Revised Plan provides:

At the conclusion of the 2000-01 school year, the district court shall enter an order releasing LRSD from court supervision and finding the LRSD unitary with regard to all aspects of school operations provided that LRSD has substantially complied with its obligations set forth in this Revised Plan. In anticipation of release, LRSD shall issue a report on March 15, 2001 indicating the state of LRSD's compliance with the Revised Plan. Any party challenging LRSD's compliance bears the burden of proof. If no party challenges LRSD's compliance, the above-described order shall be entered without further proceedings.

App. of Appellants 110.

Although not required by § 11 of the Revised Plan, one year before the final report required by § 11 was due, LRSD filed an interim report to demonstrate its progress toward compliance. App. of Appellee 71. On March 15, 2001, as required by the Revised Plan, LRSD filed its final report, which supplemented and updated the information provided in the interim report. App. of Appellee 245. The Joshua Intervenors filed objections to this report on June 25, 2001, challenging LRSD's substantial compliance with various sections of the Revised Plan.App. of Appellants 185.

After holding evidentiary hearings on the Joshua Intervenors' objections, the District Court issued an order granting LRSD partial unitary status. See Little Rock Sch. Dist., 237 F.Supp.2d at 1086. The District Court denied LRSD unitary status under § 2.7.1 of the Revised Plan, requiring LRSD to assess annually the academic programs promulgated under § 2.7. Id. at 1081-82. LRSD has not cross-appealed the District Court's ruling on § 2.7.1, and it is not before us. This issue remains pending in the District Court.

On appeal, the Joshua Intervenors argue that the District Court erred in granting partial unitary status to LRSD. Specifically, the Joshua Intervenors challenge the District Court's finding of substantial compliance with the following sections: (1) § 2.1, Good Faith; (2) §§ 2.5-2.5.4, Student Discipline; (3) § 2.6, Extracurricular Activities; and (4) §§ 2.6-2.6.2, Advanced Placement Classes.

We review the District Court's findings of fact for clear error. See Nash Finch Co. v. Rubloff Hastings, L.L.C., 341 F.3d 846, 850 (8th Cir.2003). Thus, we must affirm unless the findings are, in our opinion, clearly erroneous, which means that we must have a "definite and firm conviction" that the District Court was mistaken. Ibid. If "there are two permissible views of the evidence, the factfinder's choice between them cannot be clearly erroneous." Id. at 851 (quoting Anderson v. City of Bessemer City, 470 U.S. 564, 574, 105 S. Ct. 1504, 84 L. Ed. 2d 518 (1985)).

We also note that the Joshua Intervenors bear the burden of proof. Under § 11 of the Revised Plan, "[a]ny party who challenges the Little Rock School District's compliance bears the burden of proof." App. of Appellants 110. Section 11 of the Revised Plan also compelled the District Court to enter an order granting unitary status to LRSD unless the Joshua Intervenors met this burden. Ibid. We hold that the District Court did not clearly err in finding that the Joshua Intervenors had not met their burden with respect to the four subject-matter areas on appeal. Thus, we affirm.

The Joshua Intervenors appeal from the District Court's judgment granting LRSD unitary status under § 2.1 of the Revised Plan, which provides:

LRSD shall in good faith exercise its best efforts to comply with the Constitution, to remedy the effects of past discrimination by LRSD against African-American students, to ensure that no person is discriminated against on the basis of race, color or ethnicity in the operation of the LRSD and to provide an equal educational opportunity for all students attending LRSD schools.

App. of Appellants 88. This section places an independent duty on LRSD to exercise its "best efforts" and to act in "good faith" in attempting to remedy the effects of discrimination.

The Joshua Intervenors argue that LRSD did not act in good faith. As evidence, they allege that Central High School is still functionally segregated, although the building itself has been integrated. See Brief for Appellants 44-46. Specifically, the Joshua Intervenors argue that the advanced-placement program segregates students into different classrooms, which are the functional equivalent of different schools. Ibid. Moreover, they assert that the teachers are assigned to advanced-placement courses in a racially segregated manner — white teachers teaching advanced-placement classes and African-American teachers teaching regular classes. Ibid. The Joshua Intervenors also suggest that segregation seeps outside of the classroom and into extracurricular activities. Ibid.

The obligation of good faith under § 2.1 of the Revised Plan is separate from, and independent of, other affirmative obligations undertaken by LRSD pursuant to § 2 of the Revised Plan. Thus, it is possible for LRSD to have acted in good faith, meeting its obligation under § 2.1, even though it did not meet other affirmative obligations imposed by the Revised Plan.

After the Revised Plan was adopted, the Little Rock School Board enacted fifteen different policies related to its obligation of good faith and took steps to ensure that all administrators and teachers were aware of these new policies. LRSD also hired Dr. Terrence Roberts, Tr. of July 24, 2002, at 615-16, and Dr. Steven Ross, Tr. of July 23, 2002, at 539, as desegregation experts. Dr. Roberts testified that he had been actively involved in reviewing policies and procedures. Tr. of July 24, 2002, at 619-20. He also testified that he had developed training programs for teachers and other staff members. Ibid. Dr. Roberts testified that he told the Board that LRSD had directed much energy and effort toward meeting all the criteria in the Revised Plan and that LRSD had the potential for being a model school district for the nation. Id. at 647. Dr. Roberts criticized LRSD for having a "compliance mentality" because some individuals were interested only in meeting the requirements of the Revised Plan. Id. at 630-31. However, as explained by the District Court, compliance was exactly the issue at hand. LRSD was under constant scrutiny and had to be very careful that it met its obligations. Little Rock Sch. Dist., 237 F.Supp.2d at 1045.

Under § 8.2 of the Revised Plan, a detailed procedure for addressing compliance issues was established whereby the parties would attempt to solve compliance issues before submitting them to the District Court for resolution. The Board paid the Joshua Intervenors to monitor LRSD's compliance with the Revised Plan. During the term of the Revised Plan, the Joshua Intervenors raised only five compliance issues, which were all resolved without resorting to the District Court. App. of Appellee 415. None of the issues raised in opposition to the final report was previously raised by the Joshua Intervenors. The District Court found that the purpose of the dispute mechanism under § 8.2 was to avoid any surprises when LRSD filed the final report, and that LRSD reasonably relied on the Joshua Intervenors to raise any problems in a timely fashion. Little Rock Sch. Dist., 237 F.Supp.2d at 1043. The District Court also found that the interim report placed the Joshua Intervenors on notice of all the problems, but they did not respond. Ibid. Although § 11 does not require that any objections be previously raised under § 8.2, the District Court found that Intervenors' failure to raise these issues was a factor to consider in deciding whether LRSD substantially complied with the Revised Plan. Id. at 1043-44.

For the reasons stated above, we find no clear error in the District Court's finding of substantial compliance with § 2.1 of the Revised Plan.

The Joshua Intervenors also appeal from the District Court's judgment granting LRSD unitary status under §§ 2.5-2.5.4, relating to student discipline. Although §§ 2.5.1-2.5.4 impose specific obligations with regard to discipline, the Joshua Intervenors assert in particular that LRSD did not meet its obligation under § 2.5, which provides:

LRSD shall implement programs, policies and/or procedures designed to ensure that there is no racial discrimination with regard to student discipline.

App. of Appellants 90. This section requires LRSD to create and implement programs and policies designed to eliminate discriminatory practices from student discipline. It does not require, however, that LRSD in fact absolutely eliminate racial disparity from student discipline.

The Joshua Intervenors argue that the District Court improperly found that LRSD had substantially complied with § 2.5 because the Court misconstrued the meaning of the words "to ensure." Brief for Appellants 40. Interpretation of the Revised Plan is a question of law, which we review de novo, and we hold that the District Court did not err in construing the obligation imposed by § 2.5.

The Joshua Intervenors argue that "to ensure" means to make sure that racial discrimination does not occur. Ibid. If "to ensure" were the only operative phrase in the provision, the argument might be well taken. But § 2.5 does not require LRSD to ensure anything. It merely requires that LRSD "implement programs, policies, and/or procedures designed to ensure...." (Emphasis ours.) The thrust of the provision is that certain programs with the purpose of ensuring that there is no racial discrimination with regard to student discipline be instituted. This does not mean that the programs must be perfectly efficacious. In addition, the object is to eradicate discrimination, which is not necessarily the same thing as disparity. Racial disparity may exist without discrimination. Discrimination, of course, can cause disparity, but it is not the only possible cause.

Disparity in discipline is a nation-wide problem. The District Court cited something called "total suspension index." The total suspension index demonstrates disparity in discipline and is calculated by dividing the percentage of African-American students expelled or suspended by the percentage of African-American students in the population, and comparing this number with that for white students. The District Court found that LRSD's suspension index was between 1.25-1.31 for the years 1997-2001. Little Rock Sch. Dist., 237 F.Supp.2d at 1054. In other words, to take 1997 as an example, African-American students were 1.25 times as likely, so to speak, to be disciplined or suspended than white students. The national index for 1998 was 2.24, and the Arkansas index was 2.16. The District Court specifically found that the Joshua Intervenors did not meet their burden of proving that disproportionate discipline imposed on African-American students was the result of discrimination. Little Rock Sch. Dist., 237 F.Supp.2d at 1057. This finding is not clearly erroneous.

LRSD enacted several policies to implement its obligations regarding student discipline and created a Compliance Plan, which outlined how LRSD planned to implement the Revised Plan and who bore responsibility for such implementation. Under the Compliance Plan, Junious Babbs was responsible for monitoring student discipline. An ombudsman, James Washington, was appointed pursuant to § 2.5.3 to ensure that students were treated fairly throughout the discipline process. The ombudsman was charged with shepherding students through the discipline process, including making students aware of the rules, acting as an advocate for students involved in the disciplinary process, and investigating parental and student complaints of discrimination.

The interim and final reports issued by LRSD focused on the decrease in overall suspensions and expulsions, due in part to programs developed by LRSD, such as behavior modification programs and alternative learning centers. App. of Appellee 85-87, 273-74. Although the reduction in suspensions for African-American students was not so large as that of white students, the District Court found that the proportion of suspensions received by African-American students remained the same. Little Rock Sch. Dist., 237 F.Supp.2d at 1051. Neither the interim report nor the final report focused on the fact that racial disparity existed among the students who received suspensions or expulsions, and the District Court found that LRSD could have sorted the data in such a way as to give a more meaningful analysis. Id. at 1051-52. However, the District Court found that the Joshua Intervenors had access to the raw data and never raised the issue. Id. at 1052. More specifically, the District Court found that the reports did not mislead the Joshua Intervenors. Ibid.

The ODM produced a Report on Disciplinary Sanctions in LRSD, which showed that African-American students received a disproportionate number of suspensions and expulsions. However, the District Court specifically found that this report was not intended to address the effectiveness of any programs that were instituted to address fairness in discipline. Id. at 1052-53. The District Court also noted that the report suggested that factors outside of the schools might affect which students receive discipline, such as home environment, family values, and whether the home is a single-parent home. Id. at 1052. The report did not contain a specific analysis of the facts of each suspension or expulsion to help determine whether discrimination occurred. Id. at 1052-53. However, the report did conclude that the racial disparity meant that LRSD "has certainly not eliminated nor even abated racial discrimination in suspensions...." Id. at 1053 (quoting Report on Disciplinary Sanctions in LRSD, June 14, 2000). The District Court rejected this conclusion as speculative because it was based on raw statistics. Ibid.

Dr. Linda Watson, the Assistant Superintendent for Student Hearings, was responsible for monitoring compliance with the Student Handbook. She reviewed every long-term suspension or expulsion and all appeals from short-term suspensions. Tr. of Nov. 19, 2001, 36-37. If the procedures of the Student Handbook were not followed, Dr. Watson overturned the punishment and removed it from the records. Ibid. Although Dr. Watson acknowledged that African-American students were more frequently suspended than white students, she believed this was due to the fact that they more frequently engaged in conduct prohibited by the Student Handbook. Id. at 83-84. She also testified that she believed this was due primarily to socioeconomic factors. (Some of these factors may be caused by or related to racial discrimination, but they are not the fault of the present administration of LRSD.) The District Court specifically found that the testimony of all the administrators involved in the disciplinary process was credible. Little Rock Sch. Dist., 237 F.Supp.2d at 1050.

For these reasons, we find no clear error in the District Court's finding of substantial compliance with § 2.5 of the Revised Plan.

The Joshua Intervenors also appeal from the District Court's judgment granting LRSD unitary status under §§ 2.6 and 2.6.3, relating to extracurricular activities. Although § 2.6.3 imposes a specific obligation with regard to transportation for extracurricular activities, the Joshua Intervenors assert that LRSD did not meet its obligation under § 2.6, which provides:

LRSD shall implement programs, policies and/or procedures designed to promote participation and to ensure that there are no barriers to participation by qualified African-Americans in extracurricular activities....

App. of Appellants 90-91.

The Joshua Intervenors argue that racial discrimination occurred in extracurricular activities, evidenced by the fact that many extracurricular activities did not have a proportionate share of African-American participants. Brief for Appellants 46. Certain activities' participants, such as tennis, swimming, quiz bowl, mock trial, and cheerleading, were predominantly white. The Joshua Intervenors also assert that there were barriers to participation, including costs of participation and lack of transportation.

The Joshua Intervenors argue that racial disparities in extracurricular activities are the result of discrimination. However, as noted by the District Court, nothing in § 2.6 of the Revised Plan required LRSD to impose quotas on extracurricular activities. Little Rock Sch. Dist., 237 F.Supp.2d at 1058. LRSD undertook to promote the participation of African-American students and to eliminate barriers to participation. As we noted above with respect to § 2.5, this provision does not make LRSD an insurer. It requires only that the District "implement programs, policies and/or procedures designed to promote participation and to ensure," et cetera. (Emphasis ours.)

The final report noted a marked increase in African-American students' participation in extracurricular activities following the enactment of the new policies. App. of Appellee 276-77. The final report also demonstrated that LRSD attempted to eliminate barriers to participation by having buses transport students to and from extracurricular activities. Id. at 278. Although the record does not establish which students took advantage of the extra buses, the final report stated that "no extracurricular activity transportation request made by an eligible student has been denied." Ibid. As noted by the District Court, the Joshua Intervenors bore the burden of proof on this issue, and they did not provide a single witness to testify that African-American students were unable to participate because of a lack of transportation. Little Rock Sch. Dist., 237 F.Supp.2d at 1059.

The Joshua Intervenors also assert that the costs of certain activities create a barrier to participation. Although there are costs associated with certain activities, Dr. Marian Lacey, Assistant Superintendent of Secondary Schools, testified that each school had a discretionary fund which could be used to help students pay the costs of extracurricular activities. Tr. of July 24, 2002, 775-76. The District Court also found that the Joshua Intervenors presented no testimony that any student was denied an opportunity to participate because of costs. Little Rock Sch. Dist., 237 F.Supp.2d at 1059-60.

The Joshua Intervenors asserted that certain schools, which were primarily African-American, did not have the same extracurricular activities as other schools, and that this violated LRSD's duty to promote participation. However, the District Court found that each school determined which extracurricular activities to offer on the basis of student interest, and if enough interest existed, each school offered a stipend to sponsors of those activities. Id. at 1060. The District Court concluded that certain activities were missing at certain schools not because of discrimination but instead because of lack of student interest. Ibid.

The Joshua Intervenors presented several students' testimony to support their assertion that African-American students were not encouraged to participate or were prevented from participating in extracurricular activities. The District Court did not find this testimony impressive. Id. at 1061. Questions of credibility and inferences to be drawn from facts must generally be left to the trial court.

The Joshua Intervenors bore the burden of proving that LRSD was not implementing programs, policies, or procedures designed to promote participation and ensure there were no barriers to participation by qualified African-Americans in extracurricular activities. We hold that the District Court did not err in determining that the Joshua Intervenors failed to meet this burden.

The Joshua Intervenors also appeal from the District Court's order granting LRSD unitary status under §§ 2.6-2.6.2, relating to advanced-placement classes and honors programs. While §§ 2.6.1 and 2.6.2 impose specific duties on LRSD to provide training programs for teachers to identify and encourage qualified African-American students to participate in advanced-placement programs and to assist African-American students in being successful in advanced-placement programs, the Intervenors do not complain that these specific provisions were violated. Instead, they focus on § 2.6, which imposes a more general duty:

LRSD shall implement programs, policies and/or procedures designed to promote participation and to ensure there are no barriers to participation by qualified African-Americans in ... advanced placement courses, honors and enriched courses and the gifted and talented program.

App. of Appellants 90-91. The phraseology of this provision is similar to others discussed above.

The Joshua Intervenors assert that the District Court erred in finding no barriers to participation in advanced-placement courses. The low number of African-American teachers assigned to advanced-placement courses, they say, is a barrier to participation. Brief for Appellants 43-44. The Joshua Intervenors rely primarily on the testimony of Dr. Michael Faucette, an English teacher at Central High School. Dr. Faucette testified that although there were eight African-American teachers and eight white teachers in Central High's English Department, African-American teachers taught only a few of the advanced-placement sections. Tr. of July 22, 2002, 176-80. Dr. Faucette, an African-American teacher, did not teach any of the advanced-placement sections. Id. at 177.

The Little Rock School District Board created a regulation setting forth criteria to help teachers identify African-American students for participation in advanced-placement courses. Although this was one factor used in identifying students for participation in advanced-placement courses, enrollment was still open to any student who showed the proper level of motivation and commitment. App. of Appellee 279. Teachers were then required to monitor performance and behavior to ensure that students placed in those courses would remain there.

LRSD studied methods to increase enrollment in advanced-placement courses and determined that pre-advanced-placement courses were necessary to prepare students better and earlier. LRSD implemented pre-advanced-placement courses for sixth and seventh-grade students. These programs have been highly successful, and the District Court found that as a result of these programs, LRSD has added over 600 African-American students to its advanced-placement courses for juniors and seniors. Little Rock Sch. Dist., 237 F.Supp.2d at 1063.

LRSD has also implemented the SMART Program, a summer program designed to teach algebra to students to prepare them for algebra in the eighth grade. App. of Appellee 112. The District Court found that during the term of the Revised Plan, at least 95% of the students attending the SMART Program were African-American. Little Rock Sch. Dist., 237 F.Supp.2d at 1063. Evaluations of the SMART Program determined that it was a success. Tr. of July 24, 2002, 678.

LRSD also instituted a "Teachers of Color" program to increase the number of African-American advanced-placement teachers. Id. at 671. The principal at each middle school and high school determined who would be assigned to teach each class. However, the principals were constrained by the collective-bargaining agreement, which required consideration of a teacher's experience and seniority. Tr. of July 22, 2002, 90. An advanced-placement teacher also needed to be qualified through the state. Although Dr. Faucette testified about the racial composition of advanced-placement teachers in Central High School's English Department, he did not know about other advanced-placement sections at Central High School. Little Rock Sch. Dist., 237 F.Supp.2d at 1065. The District Court found Dr. Faucette's testimony unreliable. Ibid.

The Joshua Intervenors also point to racial disparity in the Hall High School University Studies program, a program developed in conjunction with the University of Arkansas at Little Rock that provided an opportunity for students to earn college credit for classes taken at Hall High School. Admission requirements were developed by the University of Arkansas. Tr. of July 24, 2002, 727-28. In order to receive college credit for the courses, students were required to pay tuition of approximately $150 per course. Tr. of July 22, 2002, 114. The Joshua Intervenors assert that the tuition payments created a barrier to participation for African-American students. Brief for Appellants 42-43.

The District Court found that during the 1999-2000 school year, 58% of the students participating in Hall High School's University Studies Program were African-American, while African-American students comprised 71% of all students at Hall High School. Little Rock Sch. Dist., 237 F.Supp.2d at 1066. During 2000-2001, only 35% of the students in the University Studies Program were African-American, while African-American students comprised 72% of all students at Hall High School. Ibid. However, the Court found that the Joshua Intervenors presented no evidence that any student was denied admission to the University Studies Program because of inability to pay. Ibid. Testimony also indicates that the school solicited a donation to cover the cost for at least one African-American student who wished to participate but was unable to pay. Tr. of July 24, 2002, 802.

For these reasons, we hold that the District Court did not err in finding that LRSD substantially complied with its obligations under § 2.6 of the Revised Plan.

* * * * * *

The judgment is affirmed. It goes without saying, but we say it anyway, that LRSD remains fully subject to the Constitution and all other applicable laws, and that these obligations are enforceable by appropriate legal action.

HEANEY, Circuit Judge, concurring.

I concur in every aspect of the majority's opinion except insofar as it holds that the LRSD has implemented "programs, policies and/or procedures designed to ensure that there is no racial discrimination with regard to student discipline," as required by section 2.5 of the Revised Plan. In my view, the LRSD has failed to meet this obligation.

It is true that the LRSD has implemented several programs with regard to student discipline: the LRSD provided every student, parent, teacher, and administrator with a copy of the Student Handbook; the LRSD trained students, teachers, and administrators on provisions in the Handbook; the LRSD created the position of Ombudsman to investigate student complaints of race-based mistreatment in student discipline; Dr. Linda Watson, the Assistant Superintendent who was responsible for implementing section 2.5 of the Revised Plan, reviewed every long-term suspension and expulsion, and any short-term suspensions that were appealed; Dr. Watson prepared and reviewed quarterly Discipline Management Reports from each school, used these reports to identify problems, and met with the schools' administrators to discuss solutions; the LRSD established alternative learning environments to allow students with behavioral problems to remain in school; the LRSD offered training in classroom management and effective discipline; and the LRSD followed a progressive discipline approach by imposing lesser sanctions before suspending students.

It is also true that the LRSD has reduced the total number of disciplinary sanctions of students during the time of the Revised Plan from 5,312 total sanctions in 1998, to 5,080 total sanctions in 2001.5 During that same period, however, the number of black students receiving disciplinary sanctions actually increased. During the 1998-99 school year, there were 4,470 disciplinary sanctions of black students compared to 842 disciplinary sanctions of white students. Put another way, in the first year of the Revised Plan, 65% of the student population in the LRSD was black, while 84% of the disciplinary sanctions were of black students. By 2001, the year the LRSD sought unitary status, the disparity was even greater. In the 2000-01 school year, there were 4,534 disciplinary sanctions of black students compared to 546 disciplinary sanctions of white students. In other words, black students consisted of 68% of the student population, but accounted for 89% of the disciplinary sanctions. Therefore, from 1998 to 2001, disciplinary sanctions of black students increased from 84% to 89%. It is undisputed that the programs instituted by the LRSD to address disciplinary issues have had no positive impact on the racial disparity of student discipline in the district.

If you compare the discipline statistics in the individual high schools for the same period they track in very similar ways with almost all of the schools experiencing an increase in disparity. It is worth noting, however, that Parkview High School, the most integrated high school in the district, has the lowest racial disparity in student discipline in the district. In 1998-99, Parkview's student population was 51% black and the percentage of disciplinary sanctions of black students was 49%. In 2000-01, Parkview's black student population was still 51%, but the percentage of disciplinary sanctions of black students rose to 66%. Even at 66%, however, Parkview still had the lowest disparity in student discipline in the district that year.

I agree that the Revised Plan does not require the LRSD to absolutely eliminate racial disparity from student discipline. The majority and the district court, however, rely heavily on the fact that section 2.5 requires the LRSD to implement programs "designed to ensure" that there is no racial discrimination in student discipline. The implication is that because the LRSD implemented programs which would effect student discipline, the actual impact of those programs does not matter. I disagree. It is not enough for the LRSD to list the programs it implemented to address the disparity in student discipline, when the result of those programs was an increase in the racial disparity in student discipline. The mere implementation of programs, no matter how many or how impressive sounding, that have virtually no impact on the racial disparity in student discipline is not enough to meet the district's obligations under the Revised Plan.

This lack of impact on the disparity in discipline is really no surprise when you review the testimony of Dr. Watson. Dr. Watson testified that: she was never instructed that there needed to be a reduction in the racial impact of suspensions in the district; she never prepared a monitoring report with regard to disparities in discipline; she did not prepare any reports which track whether certain teachers or administrators have a pattern of disciplinary actions based on race; nor did she recommend any programs to address the continued disparate impact of discipline. (Nov. 19, 2001, Unitary Status Hr'g Tr. at 25-163.) Dr. Watson also testified that the percentage of black students being suspended did not decrease, that disparate patterns of discipline still exist based on race, that there are no plans to reduce the disparate impact of student discipline in the district, and that the LRSD is not even looking at student discipline based on race. (Id.)

The majority, and the district court, seem to take solace in the fact that racial disparity in student discipline is a national problem. According to the district court, in 1998, the national "total suspension index" was 2.24 and the Arkansas "total suspension index" was 2.16, whereas the LRSD's "total suspension index" remained constant at 1.26 from 1997-2000. Little Rock Sch. Dist. v. Pulaski County Special Sch. Dist. No. 1, 237 F. Supp. 2d 988, 1054 (E.D.Ark.2002). The majority and the district court consider the fact that the LRSD's index is lower than that of the nation and the state significant, and the fact that the LRSD's index did not change over the period of the Revised Plan insignificant. I disagree. The Revised Plan said nothing about the LRSD's racial disparity in student discipline in comparison to the state or the nation. The Revised Plan did, however, require the LRSD to implement programs designed to ensure that the racial disparity in student discipline in the district would decrease. This, they failed to do.

The majority and the district court also assert that Joshua did not meet its burden in proving that the racial disparity in student discipline was the result of discrimination. This was not Joshua's burden. According to section 11 of the Revised Plan, Joshua bears the burden of proving that the LRSD failed to comply with its obligations as set forth in the plan. Joshua met this burden by showing that the programs the LRSD implemented to address the racial disparity in discipline were ineffective. As I read the Revised Plan, it was the LRSD's obligation to determine whether the continued disparity in discipline was the result of racial discrimination or merely socioeconomic factors as suggested by Dr. Watson. Here again, the LRSD failed to meet its obligation and rested merely on the fact that it implemented programs. Programs that, in the end, had no effect on the racial disparity in student discipline.

It is true that Joshua could have done more to raise concerns about the failure of the LRSD's programs earlier, but this does not remove all responsibility from the LRSD. The statistics compiled and reports filed by the LRSD lack valuable data. I have found no useful statistics on recidivism among students to determine how many students, and of what race, are receiving multiple disciplinary sanctions. The record does not contain statistics that separate offenses involving the discretionary judgment of staff from objective offenses. The record lacks any reports which show whether there is a correlation between the race of the teacher administering the discipline and the race of the student receiving it, or whether certain teachers have a higher rate of discipline than others. Dr. Watson testified that she was able to access some of this information and that she knew which schools had high rates of disciplinary sanctions and which teachers issued more suspensions than others, but I cannot agree that her personal, undocumented knowledge was sufficient to meet the court's mandate that the district implement programs, policies, and procedures designed to ensure that there is no racial discrimination with respect to student discipline.

Absent the necessary records, there is no way the district court, or this court, can reach an informed conclusion as to whether blacks are disciplined more frequently for legitimate reasons or because they are judged by different standards than white students, at least by some teachers. I would remand this case to the district court on the disciplinary issue, along with the issue of student achievement retained by the district court, to require the district to comply with our original mandate.