Page v. Bartels

OPINION PER CURIAM. Plaintiffs filed a Complaint seeking in-junctive relief to prevent defendants from implementing a plan for redistricting of New Jersey’s Senate and General Assembly districts. The plaintiffs having failed to satisfy their burden of proof, we will deny the injunction sought by the plaintiffs and enter final judgment for the defendants. I. Article 4, Section III, ¶ 1 of the New Jersey State Constitution provides that: After the next and every subsequent decennial census of the United States, the Senate districts and Assembly districts shall be established, and the senators and members of the General Assembly shall be apportioned among them, by an Apportionment Commission consisting of ten members, five to be appointed by the chairman of the State committee of each of the two political parties whose candidates for governor receive the largest number of votes at the most recent gubernatorial election. ... The Commission, by a majority of the whole number of its members, shall certify the establishment of Senate and Assembly districts and the appor *349 tionment of senators and members of the General Assembly to the Secretary of State within one month of the receipt by the Governor of the official decennial census of the United State for New Jersey, or on or before February 1 of the year following the year in which the census is taken, whichever date is later. N.J. Const. Art. 4, § 3, ¶ 1. Pursuant to this mandate, an Apportionment Commission was appointed in November 2000, consisting of five Democrats and five Republicans. The Commission met several times in February and March 2001 but was unable to certify the establishment of Senate and Assembly districts. Accordingly, the Commission submitted a certification to this effect to the Chief Justice of the New Jersey Supreme Court pursuant to Article 4, Section III, ¶ 2 of the New Jersey State Constitution, which provides: If the Apportionment Commission fails so to certify such establishment and apportionment to the Secretary of State on or before the date fixed or if prior thereto it determines that it will be unable so to do, it shall so certify to the Chief Justice of the Supreme Court of New Jersey and he shall appoint an eleventh member of the Commission. The Commission so constituted, by a majority of the whole number of its members, shall, within one month after the appointment of such eleventh member, certify to the Secretary of State the establishment of Senate and Assembly districts and the apportionment of senators and members of the General Assembly. N.J. Const. Art. 4, § 3, ¶ 2. On or about March 26, 2001, the Chief Justice appointed Professor Larry Bartels of Princeton University as the Commission’s eleventh member. After reviewing plans and maps submitted by the Republicans and the Democrats, Professor Bartels endorsed a plan 1 that was similar to the plan submitted by the Democrats. This plan was certified by the Commission on April 11, 2001 by a vote of six to one. 2 Immediately after the Commission approved the redistricting plan, plaintiffs 3 filed a Verified Complaint on April 12. The Verified Complaint contained four counts against several defendants. 4 First, the complaint claimed that “Defendants’ actions as alleged infringe plaintiffs’ rights as protected by § 2 of the Voting Rights Act of 1965.” The second count charged that “[t]he Bartels Plan, as adopted by a 6-1 vote of the Apportionment Commission, accordingly purposefully and intentionally violates § 2 of the Voting Rights Act of 1965.” Third, the complaint alleged that “[ratification and employment of the Bartels Plan, purported to be adopted by the Apportionment Commission on April 11, 2001, violates Plaintiffs’ rights to Due Process and Equal Protection as guaranteed by the 14th Amendment of the United States Constitution.” Finally, the fourth *350 count claimed that “Defendants’ actions as alleged violate Plaintiffs’ rights under the 15th Amendment of the United States Constitution.” In connection with all four counts, the plaintiffs requested: a preliminary and permanent injunction, enjoining and restraining the defendants, their officers, agents, employees, servants, attorneys, and all those in action, concert or participation with them from a) employing, ratifying, or in any way putting into effect, directly or indirectly, the apportionment map, purportedly approved by the New Jersey Apportionment Commission on April 11, 2001; b) from printing, causing to be printed, distributing, disseminating or causing to be distributed or disseminated ballots or other means of effecting an election in connection with any primary election for New Jersey Legislative Districts. (Compl. at 24.) Additionally, plaintiffs requested generally “such other and further relief as this Court may deem just and proper, including, enjoining and restraining defendants [from] the holding of any primary election for New Jersey Legislative Districts until further Order of this Court.” Also on April 12, plaintiffs submitted a proposed Order to Show Cause with Temporary Restraints to District Court Judge Dickinson R. Debevoise of the District of New Jersey. Judge Debevoise entered the order on April 12, 2001, which “immediately and temporarily restrained [defendants] from employing, ratifying, or in any way putting into effect, directly or indirectly, the apportionment map purportedly approved by the New Jersey Apportionment Commission on April 11, 2001.” The Order stated that “the Defendants shall show cause ... why an Order should not be entered against them restraining and enjoining the defendants, and each of them, from employing, ratifying or in any way putting into effect, directly or indirectly, the apportionment map, purportedly approved by the New Jersey Apportionment Commission on April 11, 2001.” The parties were further ordered to appear before Judge Debevoise on April 16, 2001 for a hearing on the Order to Show Cause. After hearing from the parties on April 16, Judge Debevoise issued an opinion from the bench, stating that “[i]n the present case the facts compel the conclusion that reducing the concentration of black Essex County voters in the three districts does not reduce the opportunities of black voters to participate in the political process.” (Tr. 54.) Accordingly, the District Court concluded, inter alia, that “there is no likelihood that plaintiffs will ultimately prevail on the merits,” and denied plaintiffs’ application for a preliminary injunction. 5 (Tr. 57.) Judge Debevoise did, however, extend the temporary restraints he had previously entered until noon on April 17, so that plaintiffs would have time to file an emergency appeal. Plaintiffs filed an emergency appeal on the following day, which was heard by *351 Judge Leonard I. Garth of the Third Circuit Court of Appeals, sitting as a single judge of that court. At the argument, Judge Garth raised the issue of Judge Debevoise’s jurisdiction to issue the April 16, 2001 preliminary injunction inasmuch as 28 U.S.C. § 2284 requires that “[a] district court of three judges shall be convened when otherwise required by Act of Congress, or when an action is filed challenging the constitutionality of the apportionment of congressional districts or the apportionment of any statewide legislative body.” 28 U.S.C. § 2284(a). Accordingly, Judge Garth ordered briefing of the statutory court jurisdictional issue and extended the stay ordered by Judge Debevoise “until noon April 24, 2001, or until earlier and further action of a three-judge panel of the Court of Appeals.” The Court of Appeals acted rapidly on this order and heard the plaintiffs’ appeal on Monday, April 23. 6 At oral argument before the Third Circuit panel, the issue of the election schedule and the effect that any relief to plaintiffs would have on the timing of the election was discussed in detail. As originally legislated, the filing deadline for candidates was April 19, 2001, and the primary was to be held on June 5. The general election is scheduled to take place on November 6, 2001. The Attorney General explained that, during the interim period of approximately seven weeks between the filing deadline and the primary date, ballots had to be drawn and printed and the printed absentee and military ballots had to be mailed. This schedule had already suffered delays at the time of the argument before the Third Circuit panel, in that the original filing deadline of April 19 had been extended indefinitely. Indeed, on the date that the jurisdictional argument was being heard by the Third Circuit, the New Jersey Legislature enacted legislation extending the filing date to May 10, 2001 and the primary date to June 26, 2001. That bill was signed into law by the acting Governor of New Jersey on April 23. The Third Circuit panel which reviewed Judge Debevoise’s April 16 order ruled from the bench on April 23 and issued a written opinion. After discussing the requirements of 28 U.S.C. § 2284 in detail and holding that § 2284 barred Judge De-bevoise from ruling on the preliminary injunction application in the absence of a statutory three-judge panel, the opinion concluded as follows: Given the potentially disruptive effects that our actions could have on New Jersey’s electoral process, it is incumbent upon us to articulate our disposition of this appeal with surgical accuracy. Our exact disposition is as follows: We will vacate the District Court’s April 16, 2001 order, and remand the case to the District Court, so that a district court of three judges, as specified in 28 U.S.C. § 2284, can be convened to hear both the Voting Rights Act and the constitutional challenges brought by Plaintiffs. With respect to interim relief, the temporary stay issued on April 17, 2001 will expire at its scheduled time, April 24, 2001 at noon. We will grant no further interim relief in this matter. We note, however, that neither the District Court, acting as a single judge, nor the district court of three judges that will be convened, is foreclosed from acting (and issuing interim relief), provided that they comply with the limitations on their authority imposed by 28 U.S.C. § 2284. *352 Page v. Bartels, 248 F.3d 175, 198 (3d Cir.2001). Directly after oral argument and the issuance of the order vacating the District Court’s April 16 denial of a preliminary injunction, Chief Judge Becker informed the parties that Judge Debevoise had now requested the convening of a three-judge panel consistent with the Third Circuit’s ruling. Chief Judge Becker then designated a statutory three-judge court pursuant to 28 U.S.C. § 2284(b)(1). The three-judge panel consisted of Judge Garth as the presiding judge, and District Court Judges Dickinson R. Debevoise and Harold A. Ackerman. A pretrial conference was held on the following day, Tuesday, April 24. At the pretrial conference, the parties were informed that a two-day trial would be held on Monday, April 30 and Tuesday, May 1. The parties were instructed to submit exhibits to the Court by Thursday, April 26 and to conclude depositions by Sunday, April 29. An amicus brief was received from the NAACP Legal Defense & Educational Fund, Inc., et al., urging a full evidentiary trial of all claims and defenses. Such a full evidentiary trial was held on April 30 and May 1, 2001. II. Plaintiffs called six witnesses: Walter L. Fields, who had consulted with the Republicans in their formation of a proposed apportionment plan; Donald Page, an African-American local politician in Essex County who is the first named plaintiff in the caption of this case; New Jersey State Senator Ronald Rice (D), 7 of District 28; Reverend Cordy T. Vivian, who has been involved in the civil rights movement since its inception in the 1960s; Professor James Loewen, the plaintiffs’ expert witness; and New Jersey State Senator John O. Bennett (R), the New Jersey Senate majority leader and Senator for District 12. Defendants called eight witnesses: U.S. Congressman Robert Menendez (D); New Jersey Assemblywoman Nia Gill (D) of old District 27; Professor Larry Bartels; Professor Allan Lichtman, defendants’ expert witness; New Jersey Senator Wayne Bryant (D) of District 5; New Jersey Assemblyman Joseph Charles (D) of District 31; New Jersey Assemblyman Wilfredo Caraballo (D) of District 28; and New Jersey Senator Richard Codey (D) of District 27. In order to understand the positions taken by the plaintiffs and the defendants, we recount first those historical and statistical facts which neither party disputes. Initially, plaintiffs challenged four districts in the plan adopted by the 2001 Apportionment Commission (the “Bartels plan”), Districts 27, 28, 29, and 34. However, during the course of the trial, it became clear that Districts 28 and 29 were unassailable as they stood under the Bartels plan. 8 Therefore, our analysis and discussion will focus primarily on Districts 27 and 34. Under the 1991 apportionment plan, there are a total of 20 African-American and Hispanic members of the New Jersey State Legislature: 4 of the 40 current members of the New Jersey Senate are African-American, and there are no Hispanic Senators; 11 of the 80 members of the General Assembly are African-American and 5 are Hispanic. (DX 40.) *353 As constituted under the 1991 Apportionment Plan the white, African-American, Hispanic and total minority voting age populations (“WVAP”, “AAVAP”, “HVAP” and “MVAP”) were as follows: District WVAP AAVAP HVAP MVAP 27 31.4% 52.8% 9.4% 68.6% 28 20.4% 57.4% 16.8% 79.6% 29 20.9% 48.2% 26.2% 79.1% 34 76.8% 3.9 11.3% 23.2% Under the Bartels plan the boundaries of these districts were changed and, particularly in District 27, the percentage of African-American voters declined dramatically. The new racial composition of the voter age population is: District WVAP AAVAP HVAP MVAP 27 58.0% 27.5% 6.6% 42.0% 28 30.3% 48.3% 14.0% 69.7% 29 22.5% 39.2% 33.2% 77.5% 34 48.2% 35.3% 9.8% 51.8% The composition of the four districts under the 1991 plan and the Bartels plan is as follows: Dist. 1991 Plan 27 City of Orange, East Orange, West Orange, part of Newark, and Montclair 28 Irvington, Maplewood, part of Newark, and South Orange Village 29 Hillside and part of Newark 34 Bloomfield, Clifton, Glen Ridge, Little Falls, Totowa, and Wayne Bartels Plan Caldwell, City of Orange, Essex Fells, Fairfield, Livingston, Maplewood, part of Newark, North Caldwell, Roseland, South Orange Village, West Caldwell, and West Orange Belleville, Bloomfield, Irvington, and part of Newark Hillside and part of Newark East Orange, Glen Ridge, Montclair, Clifton, and West Paterson (DX 7.) Neither party disputes the above facts and figures. Where the parties diverge is in the interpretation of the voting opportunities afforded to minorities under the Bartels plan. It is the contention of the plaintiffs that the realignment of the districts under the Bartels plan and the reduction of the African-American voting age population in District 27 from 53% to 27% will reduce or eliminate the opportunity of African Americans in that district to elect legislators of their own race. The plaintiffs also contend that the new District 34, with a 35% African-American voting age population, does not afford sufficient opportunities for minorities to elect their preferred candidates. Plaintiffs assert that racial bloc voting is so prevalent in those areas that African Americans cannot count on cross-over votes from other racial groups to ensure victory of their candidates. Thus they will have less opportunity than other voters to participate in the political process and to elect representatives of their choice in violation of Section 2 of the Voting Rights Act. Plaintiffs also contend that the interests of Hispanic and African Americans are so different they cannot be considered members of the same voting bloc. On the other hand, the defendants argue that the new districts are not racially polarized, that black representatives will not only be elected from the new districts, including new District 27, but that minority representation in the state legislature *354 will be increased under the Bartels plan. In essence, the defendants seem to suggest that, although the “old” Districts 27, 28, and 29 have elected African-American representatives, some of these votes are being “wasted” under the 1991 plan, because those representatives would have been elected in any event with a much diminished minority population. Hence, if the “wasted” excess was diverted to the new districts (such as the new District 34), even more avenues for the election of minority representatives would be opened up. A. Actions of Apportionment Commission After being appointed as the eleventh member of the Apportionment Commission, Professor Bartels hired among his staff members Professor Ernest C. Reock, Jr., who was an expert in apportionment matters, and retired New Jersey Supreme Court Justice Robert L. Clifford to advise him on the Voting Rights Act and constitutional aspects of apportionment. Between April 2 and April 11, 2000, Professor Bar-tels met several times with the Republican and Democratic delegations to review procedures and standards for the Commission’s task and to discuss the plans and maps proposed by each delegation. Professor Bartels specified a number of criteria for the apportionment plan. These included: 1) minimizing population deviation between the districts in order to secure compliance with the “one person, one vote” rule established under the federal Constitution; 2) ensuring the fair representation of minority voters as required by the Voting Rights Act and the federal Constitution; 3) keeping each of the forty existing districts contiguous; 4) keeping each of the existing districts reasonably compact; 5) respecting municipal boundaries by not splitting towns smaller than Newark and Jersey City among different districts; 6) respecting voting district boundaries; 7) avoiding any bias in favor of one or the other political parties; 8) ensuring that some seats remained competitive; and 9) minimizing voter disruption. During his testimony, Professor Bartels explained that avoiding bias in favor of one party or another meant creating a plan under which, if the vote split 50% for Republicans and 50% for Democrats, the legislature would be equally divided between representatives of the two parties. He stated that minimizing voter disruption entailed minimizing the situations where voters would have to vote for different incumbents. A principal issue arose out of the configuration of districts in Essex County. In essence, the Republican delegation sought to maintain the 1991 majority-minority districts in Essex County, while the Democratic delegation sought to change the configuration of these districts, reducing the minority population in majority-minority districts and increasing the minority population in neighboring districts. On April 11, 2001, the Commission adopted the Bartels plan (PX 34), which was a modified version of an apportionment plan earlier submitted by the Commission’s Democratic membership. Walter Fields, who had consulted with the Republicans during the apportionment process, testified that he felt that the plan proposed by the Democrats, and eventually adopted by Bartels with some revisions, shuffled African-American voters around to create more Democratic districts but less African-American districts. (Tr. 32-33.) Fields testified that he engaged in several communications with Bartels during the apportionment process regarding the importance of retaining majority African-American districts in Essex County. (Tr. 49.) He stated that Bartels indicated *355 that he felt that “minorities would have opportunities in the new districts,” but Bartels did not explain the basis for his opinion. (Tr. 49.) Additionally, Senator John Bennett (R), who served on the Apportionment Commission, testified that he attended a public hearing at which Assemblyman Wilfredo Caraballo and U.S. Congressman Robert Menendez indicated that the Hispanic community should be kept together in the new plan and cautioned against diluting and fragmenting that community. (Tr. 269.) Nevertheless, Senator Bennett testified, Professor Bartels did not find it “offensive to take a 50 percent majority black district and divide it into ... two 25 percent African-American districts and he felt that would give equal opportunity to minorities.” (Tr. 279.) On the date that the Bartels plan was approved, Professor Bartels received two letters from Laughlin McDonald of the American Civil Liberties Union Foundation. In the first letter, Mr. McDonald explained the legal significance of Section 2 of the Voting Rights Act and observed that “a plan that destroyed the existing majority-minority district would arguably violate Section 2.” (PX 11.) Additionally, he stated that “one cannot assume that districts combining various minority groups would withstand [a] Section 2 challenge on the assumption that they were majority-minority.” (PX 11.) Later that day, Mr. McDonald sent another letter to Professor Bartels explaining that he had not seen any of the proposed plans and was therefore unable to comment on their merits. (DX 14.) Professor Bartels also received a letter on April 11, 2001 from Martin L. King III of the Southern Christian Leadership Conference. In that letter, Mr. King urged the Apportionment Commission “to preserve the present configuration of legislative districts (Districts 27, 28 and 29) in Essex County,” because “[w]e strongly believe that any alteration of these district lines will dilute minority representation in the New Jersey Legislature.” (PX 12.) However, on April 25, 2001, Mr. King sent a second letter to Professor Bartels stating: The [SCLC] is committed to ensuring that African-American voters in New Jersey, and throughout the Nation, have a fair and equal opportunity to elect their preferred candidates to state legislatures and other public offices. I understand that the legislative redistricting plan that you and the New Jersey Apportionment Commission approved earlier this month does precisely that, and I further understand that the Commission’s plan actually enhances the opportunities for increased minority representation in the New Jersey Legislature. (DX 22.) Based on the testimony and exhibits in connection with the Apportionment Commission’s formulation of the Bartels plan, the Court finds that the Commission did not intend for the Bartels plan to diminish the voting rights of minorities. B. Minority Opportunities for Election under the Bartels Plan There was extensive testimony at trial that is relevant to the issue of the opportunities for African Americans and Hispanics to elect the candidates of their choice under the Bartels plan. Evidence was adduced by minority and white legislators about the opportunities for minorities to be elected in new Districts 27 and 34. There was testimony about the extent to which the African-American and Hispanic communities share common concerns and tend to vote together. Finally, expert witnesses testified for both sides, analyzing data from previous elections to determine *356 the effect of the redistricting on minority election opportunities. 1. Legislators’ Testimony Ronald Page, a witness for the plaintiffs, testified that he felt that the Bartels plan “makes less of an opportunity for African Americans.” (Tr. 88.) As to District 27, Page testified that “I do not believe we’ll ever elect again another assembly person nor a state senator on the strength of the black vote in the 27th District” (Tr. 89.) Senator Ronald L. Rice (D), an African-American Senator from District 28 who testified for the plaintiffs, seemed somewhat ambivalent about the Bartels plan. He testified that he does not “buy into the notion that some whites will not vote for African Americans than others, but in order for whites to vote for African Americans, my experiences have been that they have to get to know us.” (Tr. 103). Thus Senator Rice’s objection to the Bartels plan is that it puts new municipalities in his district in which people do not know him and his chance of being reelected is lessened because he will have to run “off the Democratic line” 9 in the new District 28. However, Senator Rice testified on cross-examination that the candidate who would be running for Senator on the Democratic line in new District 28 was Willie Brown, an African American. On the other hand, U.S. Congressman Robert Menendez (D), who testified for the defendants, stated that, when he first ran for Congress in 1992, his Congressional District included a 42% Hispanic population (with a 25% Hispanic voting population), a 14% African-American population (with an 8% African-American voting population). He prevailed with 68% of the vote against a white opponent. He has been elected in each election since then with ever-increasing margins. Congressman Menendez testified, and the results of his elections confirm, that he has received wide support from the African-American community and leaders and from white voters. He himself is Cuban, but there are in his district Hispanics of other derivations, including large numbers of Puerto Ricans. Assemblywoman Nia Gill is an African-American Democratic member of the Assembly from former District 27. She testified that she had always received support from Hispanic and white voters. Under the Bartels plan, she will be in the newly configured District 34 where she will run for the Senate this year and expects to prevail with a 35% African-American voting age population in that district. Assemblywoman Gill testified that she is confident that an African American can be elected to an Assembly position in new District 27 and that she, or another African American, will be elected to the Senate in the new District 34. She noted that, as of the date of the trial, the only people who had filed to ran in the Democratic primary for State Senator in new District 34 were African-American women. She explained that the districts in Essex County have a Democratic majority. Therefore, the real contest in those districts is in the Democratic primary. According to Gill, whoever prevails in the Democratic primary is likely to be elected in the general election. In addition, even if another non-minority candidate were to file to run in the Democratic primary in District 34, Gill testified that she is still confident that an African American (likely herself) would be elected. The defendants also produced as witnesses several other minority legislators who were elected in districts where as few as 30% of the voters were from their eth- *357 nie/racial communities. Hispanic Assemblyman Wilfredo Caraballo testified that he had been elected to the New Jersey Assembly from District 28 on three occasions. In 1995, when Caraballo first ran for the General Assembly, District 28 was 16% Hispanic, 54% African-American and 30% white. He asserted that he could not have been elected without substantial African-American and white support. He gave his opinion at trial as to the effect of reducing the number of African Americans in District 27: “... the only way that we can focus on the reduction in the 27th is to also understand what happened in the 34th. Because two of the big towns [East Orange and Montclair] out of the 27th are now in the 34th. So both of these districts have to be looked at together to understand how .., in fact, the opportunities ... for African Americans in particular were not, in fact, diminished but, in fact, augmented when we took the towns of the 27th, and were able to split them up into the 27th and the 34th”. (Tr. 555). Caraballo also confirmed what other witnesses described as the process of “coalition building.” In each legislative district three candidates are elected, two members of the Assembly and a Senator. This leads to coalition building, ensuring that the ticket includes representatives of racial groups that reside in the district in significant numbers. Based on the record, including testimony from New Jersey legislators, we are persuaded that the following changes in minority representation may reasonably be anticipated in the New Jersey Legislature under the Bartels plan: Representatives, 1991 Plan Representatives, Bartels Plan District (1999 election results) (predicted 2001 election results) white black Hisp. white black Hisp. 27 28 29 34 Total OO |tO W to H CO | HA bO T — I | 1 i -q | co to co | CO H^ The evidence demonstrates that the 2001 election will most likely yield a gain of one minority (African-American) legislator in the four relevant districts. 2. Common Interests of African-American and Hispanic Communities The plaintiffs contend that the African-American and Hispanic communities cannot be considered as a bloc and that, therefore, it is irrelevant to their claims that the combined minority voting age population of the two communities — African-American and Hispanic — is 34.1% in District 27, 62.3% in District 28, 72.4% in District 29, and 45.1% in District 34. Defendants, on the other hand, argue that the two communities often vote together and that, therefore, the total minority voting age populations of each district must be considered. A number of witnesses spoke to this issue. Both Fields and Page, appearing for the plaintiffs, testified that the issues of immigration and bilingual education were of primary concern to the Hispanic community but not to the African-American community. They also asserted that *358 the two communities do not consistently vote together. Dr. Loewen, plaintiffs expert, claimed that, because different ethnic groups within the Hispanic community had varied interests and political affiliations, even among themselves, there was no consistency in voting between Hispanics and Afincan Americans. In contrast, Congressman Menendez testified that “I believe that minority voters will join together in coalition and vote for each other.” Further, he stated that “I find that the challenges they face as communities, whether it be in education, in housing, in a wide variety of issues they find common ground on, and will seek to be supportive of candidates from those communities.” (Tr. 314.) He mentioned two examples of coalitions of African-American and Hispanic candidates for the New Jersey State Assembly that have joined together and been elected together. (Tr. 314-15.) Additionally, though Congressman Menendez is a Cuban-American Democrat, he testified that he enjoys significant support from the Puerto Rican community. He explained that Cuban-Americans do have a tendency to vote Republican in national elections, but there are many Democratic Cuban Americans in state and local government positions in New Jersey and in other states. (Tr. 327.) Congressman Menendez also testified that he did not believe that immigration issues divided the Hispanic and African-American communities and pointed out that the Congressional Black Caucus and the Hispanic Caucus have joined together in Congress on immigration issues. (Tr. 317-18.) The other African-American and Hispanic legislators testifying for the defendants also indicated that African Americans and Hispanics have common interests and that they work together on issues of interest to their two communities in the Black and Latino Caucus of the New Jersey Legislature. Though we recognize and respect that the African-American and Hispanic communities have several differing sociological and political interests, the evidence of minority candidate coalitions and testimony about cross-over support by both Hispanic and African-American legislators strongly suggests, and we so find, that the African-American and Hispanic communities often vote as a bloc — a fact which may be considered in assessing the ability of either community to elect the candidate of its choice. 3. Dr. Loewen, Plaintiffs’ Expert Plaintiffs’ expert witness, Dr. James W. Loewen, was asked by plaintiffs to render an opinion as to whether the Bartels plan gave a fair opportunity to African-American and Hispanic voters to elect candidates of their choice for the New Jersey Legislature. (Tr. 135.) Dr. Loewen testified that there is a “chilling effect” on racial minorities’ participation in the political process when they make up less than 50% of the voters in a given district: the minority community’s impression that it does not have a reasonable opportunity to elect a candidate of its choice leads to less political mobilization, fewer candidates, less money donated to campaigns, less voter registration, and lower voter turn-out. 10 (Tr. 151-52.) Dr. Loewen testified specifically that reducing the percentage of African Americans in District 27 from greater than 50% to 27% would have a “chilling effect.” (Tr. 153.) To analyze the effect of the Bartels plan on minority voting and election of minori *359 ty-preferred candidates, Dr. Loewen asked plaintiffs’ attorneys to provide him with information “on all black-white contests in county-wide elections and for state legislature within Essex County for the last ten years.” (PX 49; Tr. 231.) However, Dr. Loewen received complete data from only four elections: 1) the 1994 Democratic Primary for County Executive; 2) the 1994 General Election for County Executive; 3) the 1995 Democratic Primary for General Assembly from District 28; and 4) the 1998 General Election for Essex County Executive. Only one of the elections provided to Dr. Loewen was a legislative election — the elections implicated in the present litigation. Dr. Loewen analyzed the data for the four elections using a methodology which he testified was the same as was used by defendants’ expert, Dr. Allan Lichtman. 11 Using the results of two of the four elections described above (and with reference to Dr. Lichtman’s analysis of a third election), Dr. Loewen projected the outcomes of elections in the new Districts 27 and 34 under the Bartels plan. He projected that in District 27 under the Bartels plan, a white candidate would receive 55% to 65% of the votes and a black candidate would get 40% to 45% of the votes. (Tr. 192.) He also projected that in District 34 under the Bartels plan, a white candidate would receive 52% to 62% of the votes. Dr. Loewen noted that, even taking into account the fact that African Americans have an advantage in Democratic primaries, based on his analysis, African Americans “can’t win these districts.” (T 194.) Dr. Loewen concluded from those analy-ses of the election results provided to him that, under the Bartels plan, Districts 27 and 34 are “influence districts” 12 that do not give African Americans a fair opportunity to elect candidates of their choice (Tr. 173) and that Hispanics have only one influence district. (Tr. 219.) Dr. Loewen concluded that, under the Bartels plan, Districts 28 and 29 were not unfair to African Americans and provide African Americans with a fair opportunity to elect candidates of their choice. (Tr. 172-73, 219.) While we do not attempt to summarize the full extent of Dr. Loewen’s expert testimony, any more than we will attempt to summarize the full extent of the testimony of Dr. Lichtman, it suffices for present purposes that we acknowledge that, after *360 careful study and examination of the full record, we are not persuaded by Dr. Loewen’s analysis and the conclusions that he reached. 13 4. Dr. Lichtman, Defendants’ Expert Dr. Allan Lichtman produced reports and testified about minority opportunities under the Bartels plan. Dr. Lichtman first considered and testified about the current composition of the New Jersey Legislature and the racial make-up of the districts in which minority legislators have been elected. Additionally, Dr. Lichtman engaged in a comprehensive election analysis, whereby he considered New Jersey Senate and General Assembly elections, analyzed the results of those elections in terms of the voting patterns of whites, African Americans, and Hispanics, and used that data to project the likely outcome of elections in the new Districts 27 and 34. Dr. Lichtman analyzed the results of every general election and some of the primary elections for the New Jersey Senate and General Assembly held between 1991 and 1999 in those eleven districts from which minority legislators have been elected, and in one additional district that is heavily minority. 14 (Tr. 444.) He also analyzed six county-wide elections held from 1994 through 1999 in Essex County in which African-American and white candidates squared off. (Tr. 445.) In total, he analyzed over one hundred and fifty separate elections. (Tr. 445.) Dr. Loewen, by contrast, analyzed just three county-wide elections in Essex County (the 1994 Democratic primary for county executive, the 1994 general election for county executive, and the 1998 general election for county executive) and one state legislative primary (the 1995 Democratic primary for assembly member in District 28). (Tr. 446.) Dr. Lichtman distinguished his analysis from Dr. Loewen’s in two significant respects. First, Dr. Lichtman testified that a large sample of elections is preferable to a smaller sample because only analysis of a large number of elections will eliminate the “special circumstances” of each election. (Tr. 446-47.) Second, Dr. Lichtman testified that state-wide, district-by-district elections provide a more reliable measure of voter behavior in future statewide, district-by-district elections — the very elections for Senate and General Assembly to be held under the Bartels plan-than do county-wide elections. (Tr. 447.) Dr. Lichtman testified about the results of all general elections for Senate from 1991 through 1997 in what may be called the twelve “minority districts.” For the Democratic candidate — the candidate preferred by a majority of African Americans and Hispanics in all elections in all districts — the mean African-American vote for the Democratic candidate was 85%; the mean Hispanic vote was also 85%; and the mean white vote was 52%. (Tr. 459; DX 46.) Dr. Lichtman’s results specific to Districts 27, 28, and 29 reveal that for the Democratic candidate, the mean African-American vote was 95%; the mean Hispanic vote was 97%; and the mean white vote was 48%. (DX 46.) Dr. Lichtman concluded from these results that the vote in these elections was not racially polarized. (Tr. 460.) He fur *361 ther concluded that the majority of voters in Districts 27, 28, and 29 voted for the candidate preferred by both African Americans and Hispanics. (Tr. 461-62.) The white vote was nearly evenly split between Democratic and Republican candidates. (Tr. 463.) In general elections for General Assembly from 1991 through 1999 in the twelve minority districts, Dr. Lichtman’s results reveal that, for the Democratic candidates, the mean African-American vote by ballot 15 was 85%; the mean Hispanic vote was 83%; and the mean white vote was 54%. (DX 46.) His results specific to Districts 27, 28, and 29 show that for the Democratic candidates, the mean African-American vote by ballot was 97%; the mean Hispanic vote was 91%; and the mean white vote was 54%. (DX 46.) His results specific to District 27 reveal that in the 1999 general election, in which two African-American Democratic candidates were opposed by two white Republican candidates, the two Democratic candidates together received 60% of the white vote and 99% of the African-American vote. (Tr. 467; DX 45.) These results further confirmed Dr. Lichtman’s opinion that neither the vote in District 27 nor the vote statewide is racially polarized. (Tr. 467-68, 475-76.) Dr. Lichtman further testified that, in more than thirty biracial general elections for the General Assembly — contests between African-American Democrats and white Republicans — the African-American Democratic candidates received 86% of the mean African-American vote, 79% of the mean Hispanic vote, and 55% of the white vote. (Tr. 471; DX 46.) Dr. Lichtman made special mention of the mean white majority voting for African-American candidates and the lack of “fall off’ among white Democratic voters when faced with the choice between an African-American Democrat and a white Republican. (Tr. 472.) In sum, Dr. Lichtman concluded from these results that the most salient factor in determining the winner of these elections was not the race of the candidate, but the party of the candidate. (Tr. 476.) In extrapolating from the results of his analyses of general elections for the Senate and General Assembly in order to project voting results under the Bartels plan, Dr. Lichtman accounted for two key variables: the effect that the presence or absence on the ticket of African-American, Hispanic, or white candidates had on the voter turnout of those groups, and the effect of white cross-over voting, i.e., white votes for minority candidates. (Tr. 481-87.) The “worst-case scenario” yielded by these projections was that the lowest white vote percentage for a minority candidate would be 54 %. (Tr. 484.) Specifically, Dr. Lichtman projected that the Democratic candidate in District 27 who wins the Democratic primary and enters the general election for Senate will win, whether he or she is African-American or white. (Tr. 489.) Finally, it was Dr. Lichtman’s expert opinion that the Bartels plan afforded reasonable opportunities afforded to African Americans and Hispanics to participate in the political process and to elect representatives of their choice to the New Jersey Legislature. C. Findings of Fact Taking into account all of the evidence in the record, we find that: *362 1. The new Districts 27 and 34 each have a sufficient African-American voting-age population to provide African Americans with a reasonable opportunity to elect candidates of their choice, even without considering that Hispanics usually vote with African Americans in those districts; 2. African Americans and Hispanics often vote in support of each other’s candidates; 3. There is no evidence that the white majority in the challenged districts votes sufficiently as a bloc to enable it to defeat the minority’s preferred candidates; 4. The reduction of the African-American population in District 27 from 53% to 27% will not impair the opportunity for African Americans and Hispanics to be elected in the new District 27, nor will it chill the African-American or Hispanic votes in any of the challenged districts; 5. In light of the significant increase in the minority voting population in District 34, the Bartels plan will increase rather than reduce the opportunity for minorities to be elected; and 6. There is no evidence of discriminatory intent from which it could be inferred that the Bartels plan was designed to discriminate against minorities. III. As stated above, plaintiffs have made claims under the Voting Rights Act and the Constitution, all in connection with the 2001 redistricting plan (the Bartels plan) approved by the New Jersey Apportionment Commission. However, apportionment is primarily the domain of the States, not the federal courts. Voinovich v. Quitter, 507 U.S. 146, 156, 113 S.Ct. 1149, 122 L.Ed.2d 500 (1993); Growe v. Emison, 507 U.S. 25, 34, 113 S.Ct. 1075, 122 L.Ed.2d 388 (1993) (“reapportionment is primarily the duty and responsibility of the State through its legislature or other body, rather than of a federal court.”); Reynolds v. Sims, 377 U.S. 533, 586, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964) (“legislative apportionment is primarily a matter for legislative consideration and determination”). For this reason, the Supreme Court has repeatedly held that federal courts should exercise extreme caution in reviewing state redistricting decisions. Miller v. Johnson, 515 U.S. 900, 915, 115 S.Ct. 2475, 132 L.Ed.2d 762 (1995) (holding that “[flederal court review of districting legislation represents a serious intrusion on the most vital of local functions.”); see also Hunt v. Cromartie, — U.S.-, 121 S.Ct. 1452, 149 L.Ed.2d 430 (2001). Until a plaintiff makes a sufficient showing to the contrary, the good faith of a state legislature is presumed. Cromartie, 121 S.Ct. 1452. A federal court must exercise “extraordinary caution in adjudicating claims that a State has drawn district lines on the basis of race.” Cromartie, 121 S.Ct. 1452. As the Court held in Cromar-tie, caution is especially appropriate where, as here, the state has “articulated a legitimate political explanation for its districting decision, and the voting population is one in which race and political affiliation are highly correlated.” 121 S.Ct. at 1458. A. Voting Rights Act 1. Section 2 of the Voting Rights Act, on which plaintiffs’ statutory claims are based, provides: (a) No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color, or in contravention of the guarantees set forth in section *363 1973b(f)(2) of this title, 16 as provided in subsection (b) of this section. (b) A violation of subsection (a) of this section is established if, based on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (a) of this section in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered: Provided, That nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population. 42 U.S.C. § 1973. In this case, plaintiffs claim essentially that the Bartels plan violates the Voting Rights Act because it disperses African Americans and Hispanics “ ‘into districts in which they constitute an ineffective minority of voters.’ ” Voinovich v. Quitter, 507 U.S. 146, 154, 113 S.Ct. 1149, 122 L.Ed.2d 500 (1993) (quoting Gingles, 478 U.S. at 46 n. 11, 106 S.Ct. 2752). The Supreme Court has made clear that such dispersion is not a per se violation of § 2. Indeed, the Court stated in Voinovich that “Section 2 contains no per se prohibitions against particular types of districts ... Only if the apportionment scheme has the effect of denying a protected class the equal opportunity to elect its candidate of choice does it violate § 2.” 507 U.S. at 155, 113 S.Ct. 1149. Although the plaintiffs advocate that Bartels should have accepted the Republican plan, which retains the three majority-minority districts from the 1991 plan, § 2 by no means requires creation (or retention) of majority-minority districts where it is possible to do so. The Court has pointed out the detriments and benefits of majority-minority districts: “On the one hand, creating majority-black districts necessarily leaves fewer black voters and therefore diminishes black-voter influences in predominantly white districts. On the other hand, the creation of majority-black districts can enhance the influence of black voters.” Voinovich, 507 U.S. at 154, 113 S.Ct. 1149. In Johnson v. De Grandy, the Court emphasized that majority-minority districts are not always necessary to ensure that minority groups be able to elect the candidates of their choice: If the lesson of Gingles [discussed below] is that society’s racial and ethnic cleavages sometimes necessitate majority-minority districts to ensure equal political and electoral opportunity, that should not obscure the fact that there are communities in which minority citizens are able to form coalitions with voters from other racial and ethnic groups, having no need to be a majority within a single district in order to elect candidates of their choice. 512 U.S. 997, 1020, 114 S.Ct. 2647, 129 L.Ed.2d 775 (1994). The Supreme Court’s clear instruction that the fact of the existence or dissolution of majority-minority districts does not constitute a per se violation of § 2 requires us to look at the factual, circumstances surrounding the adoption of the *364 Bartels plan and the likely effect of the plan on minority representation. Section 2(b) of the Voting Rights Act provides that we must consider “the totality of circumstances” in assessing whether a dis-tricting plan violates § 2. However, in Thornburg v. Gingles, 478 U.S. 30, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986), the Supreme Court set out three threshold requirements that a plaintiff must prove before a court need consider the “totality of the circumstances”: 1) “the minority group must be able to demonstrate that it is sufficiently large and geographically compact to constitute a majority in a single-member district”; 2) “the minority group must be able to show that it is politically cohesive,” i.e., that it votes as a racial bloc; and 3) “the minority must be able to demonstrate that the white majority votes sufficiently as a bloc to enable it — in the absence of special circumstances, such as the minority candidate running unopposed — usually to defeat the minority’s preferred candidate.” Gingles, 478 U.S. at 50-51, 106 S.Ct. 2752 (internal citations omitted). If a plaintiff does not prove the three “Gingles factors,” the § 2 claim fails as a matter of law, as does the plaintiffs’ claim here. Though the “Gingles factors” were conceived in the context of a challenge to a multi-member district, the Court held in Growe v. Emison, 507 U.S. 25, 40, 113 S.Ct. 1075, 122 L.Ed.2d 388 (1993), that those factors must also be applied as threshold requirements for § 2 challenges to single-member districts. Indeed, the Court explained: The “geographically compact majority” and “minority political cohesion” showings are needed to establish that the minority has the potential to elect a representative of its own choice in some single-member district. And the “minority political cohesion” and “majority bloc voting” showings are needed to establish that the challenged districting thwarts a distinctive minority vote by submerging it in a larger white voting population. Unless these points are established, there neither has been a wrong nor can be a remedy. Growe, 507 U.S. at 40-41, 113 S.Ct. 1075 (internal citations omitted). In connection with racial bloc voting, relevant to both the second prong (minority bloc voting) and the third prong (white bloc voting) of the Gingles test, the Court has stated that: A showing that a significant number of minority group members usually vote for the same candidates is one way of proving political cohesiveness necessary to a vote dilution claims.... And, in general, a white bloc vote that normally will defeat the combined strength of minority support plus white “crossover” votes rises to the level of legally significant white bloc voting. Gingles, 478 U.S. at 56,106 S.Ct. 2752. As the Third Circuit has noted, a plaintiff need not demonstrate the white majority’s “unbending or unalterable hostility” to the minority’s candidate of choice, but only that the usual result of bloc voting is the defeat of the minority-preferred candidate. Jenkins v. Red Clay Consol. School Dist. Bd. of Ed., 4 F.3d 1103, 1123 (3d Cir.1993). We derive our instruction in analyzing plaintiffs’ § 2 claim from Thornburg v. Gingles, 478 U.S. 30, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986). However, we feel it more expeditious in this case to consider the third threshold element of the Gingles test out of turn. It was this element of the test that the bulk of the evidence and testimony addressed and on which we primarily focused. We hold that plaintiffs have not satisfied their burden under the third Gin- *365 gles element, in other words, they have not proved that the Bartels plan will impair minorities’ ability to elect their preferred candidate. Drawing all the threads of the evidence together, we are satisfied that the Bartels plan is designed not to prevent or interfere with the election of minority representatives but rather will enhance and expand the opportunity for African Americans and Hispanics to participate in a meaningful way in the political process. We have found that even the reduction of the African-American voting age population in District 27 from 53% under the 1991 plan to 27% will not impair or prevent minorities from electing their preferred candidates from that district to the New Jersey Senate or General Assembly. Moreover, we have also found, based on the voluminous record produced, that minorities will be elected and constitute a significant voting force in the new District 34. Hence, rather than diluting the franchise participation of minorities, as plaintiffs claim, the Bartels plan will ensure that the minority voting age population remains an effective political force. We review briefly some of the evidence that influenced our conclusion. Dr. Loewen, the plaintiffs’ expert, found that: 1) 69.5% of white voters voted for the white candidate in the 1994 Democratic primary for the County Executive; 2) 72% of white voters voted for the white candidate in the 1994 general election for the County Executive; 3) 34% of white voters voted for the white candidate in the 1995 Democratic primary in District 28; and 4) 83% of white voters voted for the white candidate in the 1998 Democratic primary for the County Executive. (PX 51.) He concluded, based on this small sample, that racial polarization “seems to be a problem in Essex County.” (Tr. 187.) Using the statistics from those four elections and from the 1991 Democratic primary in District 27, Dr. Loewen projected that, in the new District 27, a white candidate would likely receive from 55% to 64.6% of the votes and a black candidate would likely receive 40% to 45% of the votes at most. He projected that a white candidate in District 34 would receive 52% to 62% of the votes. (PX 56.) The defendants’ expert, Dr. Lichtman, on the other hand, had an entirely different view of the data. At the outset, he noted that eight of the fifteen African Americans currently in the New Jersey legislature were elected from districts with less than 30% African-American voting age population. This suggested, in Dr. Lichtman’s view, that there must be significant cross-over voting in New Jersey. Analyzing the data from just Districts 27, 28, and 29, Dr. Lichtman found that 54% of white voters in those districts tended to vote Democratic and that 56% of white voters voted Democratic in biracial elections. (DX 46.) There was only one general election for Senate in one of the relevant districts (District 29) pitting an African-American Democrat against a white Republican, and, in that election, whites voted 61% for the white candidate and 39% for the African-American candidate. (PX 44.) Using the data he had gathered to project the outcome of biracial elections in new districts 27 and 34, Dr. Lichtman concluded that, in Senate elections in new District 27, 54% to 65% of voters would vote for the candidate preferred by African-American voters, and, in new District 34 Senate elections, 58% to 69% of voters would vote for the candidate preferred by African Americans. He found that, in a General Assembly election in new District 27, 55% to 66% of voters would vote for the candidate that African-American voters preferred, and the General Assembly elections in new District 34 would yield *366 59% to 70% votes for the candidate preferred by African Americans. (DX 46.) In addition to the evidence from the experts, there was extensive testimony from minority legislators who supported the Bartels plan and felt confident that the plan would increase minority election opportunities over the next ten years. The plaintiffs’ evidence did not rebut these projections. Another relevant factor emerged at trial as well: there was testimony from several witnesses that many of the white voters who had been moved into District 27 and who remained in District 34 voted Republican. Therefore, in those districts where the Democrats still had a clear majority, the additional white Republican votes were not likely to affect election outcomes. Evidence was also presented that African Americans are more likely than white Democrats to vote in the Democratic primary, and that the primary appears to be the more significant race in most of Essex County. Once again, we feel that, in this connection, Gingles is instructive. In Gingles, the evidence of white bloc voting found by the district court included: In the primary elections, white support for black candidates ranged between 8% and 50%, and in the general elections it ranged between 28% and 49%. The court also determined that, on average, 81.7% of white voters did not vote for any black candidate in the primary elections. In the general elections, white voters almost always ranked black candidates either last or next to last in the multicandidate field, except in heavily Democratic areas where white voters consistently ranked black candidates last among the Democrats, if not last or next to last among all candidates. The court further observed that approximately two-thirds of white voters did not vote for black candidates in general elections, even after the candidate had won the Democratic primary and the choice was to vote for a Republican or for no one. 478 U.S. at 59, 106 S.Ct. 2752. While this evidence in Gingles certainly supported a conclusion that “a substantial majority of white voters would rarely, if ever, vote for a black candidate,” 478 U.S. at 59, 106 S.Ct. 2752, the evidence of white bloc voting in the present case does not even begin to rise to that level. Indeed, there was no credible evidence presented that “a white bloc vote [in the new Districts 27, 28, 29 and 34] normally will defeat the combined strength of minority support plus white ‘crossover’ votes.” Gingles, 478 U.S. at 54, 106 S.Ct. 2752. Because the plaintiffs have failed to prove that the change in minority voting age populations under the Bartels plan would impair or prevent minorities from electing their chosen representatives, we hold that the Bartels plan does not violate § 2 of the Voting Rights Act and, on that basis, we refuse to grant the sought-for injunction against the Bartels plan. B. Fourteenth Amendment 1. Plaintiffs also claim that the Bartels plan violates the Equal Protection Clause of the Fourteenth Amendment, which provides that “[n]o State shall ... deny to any person within its jurisdiction the equal protection of the laws.” U.S. Const. Am. XIV, § 1. In the context of redistricting, the Supreme Court has held that electoral systems which are adopted with a discriminatory purpose and have the effect of diluting minority voting strength, are violative of the Fourteenth Amendment. 17 *367 See e.g., Rogers v. Lodge, 458 U.S. 613, 616-17, 102 S.Ct. 3272, 73 L.Ed.2d 1012 (1982); City of Mobile v. Bolden, 446 U.S. 55, 66, 100 S.Ct. 1490, 64 L.Ed.2d 47 (1980); White v. Regester, 412 U.S. 755, 765-66, 93 S.Ct. 2332, 37 L.Ed.2d 314 (1973); Whitcomb v. Chavis, 403 U.S. 124, 141 — 42, 91 S.Ct. 1858, 29 L.Ed.2d 363 (1971). This requirement of purposeful discrimination distinguishes an equal protection claim from a claim under the Voting Rights Act. To prove discriminatory purpose, it is not sufficient to show that the minority group allegedly discriminated against has not elected representatives in proportion to the population of that minority group. See White v. Regester, 412 U.S. at 765-66, 93 S.Ct. 2332; Whitcomb, 403 U.S. at 149-50, 91 S.Ct. 1858; see also Mobile v. Bolden, 446 U.S. at 75-76, 100 S.Ct. 1490 (“The Equal Protection Clause of the Fourteenth Amendment does not require proportional representation as an imperative of political organization.”). Likewise, a state action is not unconstitutional simply because it results, in a racially disproportionate impact. See Village of Arlington Heights v. Metropolitan Housing Devel. Corp., 429 U.S. 252, 264-65, 97 S.Ct. 555, 50 L.Ed.2d 450 (1977), Of course, the fact that an action bears more heavily on one race than another may be a basis for inferring discriminatory intent. See 429 U.S. at 264-65, 97 S.Ct. 555. “But where the character of a law is readily explainable on grounds apart from race,. ... disproportionate impact alone cannot be decisive, and courts must look to other evidence to support a finding of discriminatory purpose.” Mobile v. Bolden, 446 U.S. at 69,100 S.Ct. 1490. Importantly, the burden of proof on a party challenging a districting decision on equal protection grounds is a “demanding one.” Cromartie, 121 S.Ct. at 1458 (citing Miller v. Johnson, 515 U.S. 900, 915, 115 S.Ct. 2475, 132 L.Ed.2d 762 (1995)). The plaintiffs burden is to show that “race was the predominant factor motivating the legislature’s decision to place a significant number of voters within or without a particular district.” Miller, 515 U.S. at 916, 115 S.Ct. 2475 (emphasis added); Cromartie, 121 S.Ct. at 1458. To make this showing, a plaintiff must prove that the legislature subordinated traditional race-neutral districting principles, including but not limited to, compactness, contiguity, and respect for political subdivisions, to racial considerations. See Cromartie, 121 S.Ct. at 1458. In other words, a plaintiff must demonstrate that race was the “dominant and controlling” or “predominant” consideration in the districting decision. See Cromartie, 121 S.Ct. at 1466. *368 Stated succinctly, to sustain a claim under the Equal Protection Clause of the Fourteenth Amendment, a plaintiff must prove the following two elements: 1) that the disputed plan was “conceived or operated as [a] purposeful devic[e] to further racial ... discrimination,” Whitcomb v. Chavis, 403 U.S. at 149, 91 S.Ct. 1858; and 2) that the scheme does, in fact, result in a dilution of the minority group’s voting strength. Whitcomb, 403 U.S. at 144, 91 S.Ct. 1858. In this case, there was only the most minute amount of evidence even hinting at a discriminatory intent on the part of Professor Bartels and the Democrats. Dr. Loewen indicated in his report entitled “Racial Bloc Voting Analysis” that “Democrats like to draw districts that are 20%-40% black, because they know such districts are unlikely to elect a Republican and unlikely to elect a black.” (PX 49.) This statement and a few comments at trial by Fields and Senator Bennett that the Democrats wanted to keep the apportionment process secret from the public and that Professor Bartels did not explain his reasons for being satisfied with the racial compositions of new Districts 27 and 34 constitute the entire universe of evidence that the Bartels plan was drawn with a racially discriminatory purpose. By contrast, Professor Bartels explained in a certification and testified at trial that, because of the Voting Rights Act, he “paid close attention to racial and ethnic data and electoral opportunities.” (DX 3.) He went on to explain other criteria to which he “paid close attention” as well: (1) minimizing the population deviations among the districts, as required by the “one person, one vote” rule; (2) keeping each of the forty districts contiguous; (3) keeping each of the forty districts reasonably compact; (4) respecting municipal boundaries by not splitting any towns other than Newark and Jersey City (which are too large to fit in one legislative district); (5) respecting voting-district boundaries; (6) avoiding any bias in the map that might favor one political party over the other in an election year when the two parties were tightly competitive; (7) ensuring that some seats were competitive so that the composition of the Legislature would be responsive to shifts in votes from one party to another; and (8) minimizing voter disruption, so as not to deny too many New Jersey citizens the opportunity to vote for incumbents who had served them well or against incumbents who had not done so. (DX 3.) In support of Professor Bartels’ representations, all of the witnesses who testified about the apportionment process confirmed that Professor Bartels set out several criteria with which he was concerned, and his criticisms of each party’s maps were not race-based. Race per se was not the predominant motive in the approval of the Bartels plan. Rather, the Bartels plan was adopted based on a variety of factors, most notably the doctrine of political fairness, Constitutional and statutory requirements, and the concerns of minorities. Thus, we hold that the evidence is not sufficient to support a claim for violation of the Equal Protection Clause. 18 *369 C. Fifteenth Amendment The Fifteenth Amendment states, in relevant part: “The right of citizens of the United States to vote shall not be denied or abridged by the United States or by any State on account of race, color, or previous condition of servitude.” U.S. Const. Amend. XV, § 1. The Supreme Court has explained that “the [Fifteenth] Amendment prohibits all provisions denying or abridging the voting franchise of any citizen or class of citizens on the basis of race.” Rice v. Cayetano, 528 U.S. 495, 512, 120 S.Ct. 1044, 145 L.Ed.2d 1007 (2000). However, the case law suggests that a case such as this one cannot succeed on Fifteenth Amendment grounds. In Voino-vich v. Quitter, the Court observed, “we never have held any legislative apportionment inconsistent with the Fifteenth Amendment.” 507 U.S. 146, 159, 118 S.Ct. 1149, 122 L.Ed.2d 500 (1993). Put another way, the Court stated in Reno v. Bossier Parish School Board that “we have never held that vote dilution violates the Fifteenth Amendment.” 528 U.S. 820, 334 n. 3, 120 S.Ct. 866, 145 L.Ed.2d 845 (2000). These statements by the Supreme Court strongly suggest that the Bartels Plan cannot be found to violate the Fifteenth Amendment. However, even if we were to apply the Fifteenth Amendment to plaintiffs’ claim, plaintiffs would have to demonstrate under the Fifteenth Amendment that there was a discriminatory purpose behind the Bartels plan. See Reno v. Bossier Parish School Board, 520 U.S. 471, 481, 117 S.Ct. 1491,137 L.Ed.2d 730 (1997) (“Since 1980, a plaintiff bringing a constitutional vote dilution challenge, whether under the Fourteenth or Fifteenth Amendment, has been required to establish that the State or political subdivision acted with a discriminatory purpose.”). We have not found such a purpose in connection with plaintiffs’ Fourteenth Amendment claim, nor do we discern any such purpose in connection with their Fifteenth Amendment claim. IV. We are convinced that the Bartels plan as structured will encourage franchise participation of New Jersey voters, including African-American and Hispanic voters, and will do so without (Fluting or impairing minority participation. This being so, and the plaintiffs in this case having failed to prove essential elements of their statutory and Constitutional claims, judgment will be entered for the defendants, and we will deny all injunctive relief sought by the plaintiffs. Footnotes [1] Throughout this opinion, we will refer to the plan approved by Professor Bartels and the apportionment map which resulted therefrom as the “Bartels plan." [2] Only one of the five Republican members of the Commission was present at the meeting and voted against the Bartels Plan. [3] Plaintiffs in this action are: 1) African-American registered voters and residents of Essex County; 2) Hispanic registered voters and residents of Essex County and Hudson County; and 3) the Republican members of the New Jersey Senate and General Assembly. [4] Defendants in this action are: 1) the Apportionment Commission; 2) the five Democratic members of the Apportionment Commission; 3) Professor Larry Bartels, the "eleventh member" of the Apportionment Commission; 4) the Secretary of State for New Jersey; and 5) the Attorney General of New Jersey. [5] There is some confusion as to whether the plaintiffs were moving before Judge Debe-voise for a temporary restraining order or a preliminary injunction. However, the Order to Show Cause concluded by setting a time by which defendants were to file responses to “plaintiff's application for a preliminary injunction.” (Emphasis added.) Additionally, in making its decision, the District Court stated that it was "an application for a preliminary injunction'' and explicitly applied the preliminary injunction test, ruling that “[a]ll four criteria [of the test] weigh against issuance of a preliminary injunction in this case.” (Tr. 57.) In this connection, see Page v. Bar-tels, 2001 WL 417146 (3d Cir. Apr.23, 2001), and p. 7, infra. [6] The Court of Appeals panel consisted of Chief Judge Edward R. Becker, and Judges Leonard I. Garth and Morton I. Greenberg [7] When party identification is indicated throughout this opinion, we will use "R” for Republican and "D” for Democrat. [8] Indeed, even the plaintiffs’ expert, Dr. Loewen, made clear that new Districts 28 and 29 do not diminish minority voting opportunities. (Tr. 172-73, 219.) [9] In other words, he will not have the Democratic endorsement on the ballot. [10] Reverend Cordy T. Vivian also testified that African Americans experience a chilling effect on their voting when they comprise less than 50% of the population in a voting district. [11] First, Dr. Loewen looked at precincts that were overwhelmingly African-American, overwhelmingly white, and overwhelmingly Hispanic. Both Dr. Loewen and Dr. Licht-man examined precincts that were more than 90% African-American, 90% white, and 80% Hispanic (because there are no districts in New Jersey that are more than 90% Hispanic). (Tr. 175.) Next Dr. Loewen examined how those populations voted. Thereafter, Dr. Loewen conducted "overlapping percentages” analyses, looking at how the non-African Americans and non-whites in the more than 90% precincts (and non-Hispanics in the more than 80% precincts) voted, and adjusted the results for each precinct. (Tr. 175-76.) Dr. Loewen conducted a correlation analysis to determine whether there was a relationship between racial composition of a precinct and the election outcome. (Tr. 176.) Finally, Dr. Loewen performed an ecological regression analysis, which looks at the percentage of the voting-age population by race in each precinct as it relates to the percentage of votes for each candidate. (Tr. 180.) [12] Dr. Loewen defined an "influence district” as one where minority voters "have some possibility to elect what we might call the white person of their choice, but they do not have a reasonable opportunity to elect the candidate of their choice, the candidate who might be more identified with their community.” (Tr. 140.) Dr. Loewen did, however, acknowledge the possibility, if not the likelihood, that Districts 27 and 34 could elect African-American Assembly members. (Tr. 144, 246.) [13] Factors which contribute to our conclusion are: Dr. Loewen’s sample size; special circumstances that affected at least two of the elections considered; the fact and relevancy of incumbencies; and his focus on countywide, not legislative, elections. [14] Although Dr. Lichtman’s testimony and conclusions drawn from the minority membership in the New Jersey Legislature are relevant to his election analysis, we do not find it necessary to dwell on that component of his testimony. [15] Voters choose two candidates from four on the ballot in the general election for the General Assembly. (Tr. 464-65.) [16] The text of 42 U.S.C. § 1973b(£)(2) is not reproduced here because the guarantees provided therein have to do largely with language difficulties in voting, a subject not relevant to this action. [17] We should note that the Supreme Court has recognized two "analytically distinct” *367 equal protection claims that may be raised in challenging a state’s drawing of congressional districts. Miller v. Johnson, 515 U.S. 900, 911, 115 S.Ct. 2475, 132 L.Ed'.2d 762 (1995); Shaw v. Reno, 509 U.S. 630, 652, 113 S.Ct. 2816, 125 L.Ed.2d 511 (1993). First, a state may not use race as the basis for separating voters into districts. See Shaw v. Reno, 509 U.S. at 652, 113 S.Ct. 2816; see also Hunt v. Cromartie, 121 S.Ct. 1452 (2001). Thus, a reapportionment plan violates equal protection principles when, on its face, it "cannot be understood as anything other than an effort to segregate citizens into separate voting districts on the basis of race without- sufficient justification.” 509 U.S. at 652, 113 S.Ct. 2816. Second, and more applicable to plaintiffs' claims in this case, equal protection principles are violated when a state has enacted a particular districting scheme as a purposeful device "to minimize or cancel out the voting potential of racial or ethnic minorities.” Mobile v. Bolden, 446 U.S. 55, 66, 100 S.Ct. 1490, 64 L.Ed.2d 47 (1980); United Jewish Orgs. of Williamsburgh, Inc. v. Carey, 430 U.S. 144, 97 S.Ct. 996, 51 L.Ed.2d 229 (1977). Put simply, a state may not take an action which disadvantages, and is intended to disadvantage, voters of a particular race. [18] We base our holding on plaintiffs' failure to satisfy the "intent” or "purpose” prong of the Equal Protection Clause test. We note, however, that, in light of our holding above that plaintiffs have not shown that white voters will vote as a bloc so as to prevent minorities from voting for their preferred candidates in Districts 27, 28, 29, and 34, plaintiffs also would be unable to prove that the Bartels plan does, in fact, result in a dilution of the *369 minority group's voting strength. See Whit-comb v. Chavis, 403 U.S. 124, 144, 91 S.Ct. 1858, 29 L.Ed.2d 363 (1971).