OPINION AND ORDER
JOHN W. PECK, Senior Circuit Judge.This three-judge district court was convened pursuant to the en banc decision of the United States Court of Appeals for the Sixth Circuit, 925 F.2d 987 (6th Cir.1991), in order to hear this challenge to the constitutionality of the apportionment of the Ohio House of Representatives. Plaintiffs allege that the boundary between House Districts 52 and 53 in Mahoning County deliberately and effectively dilutes the minority vote, and therefore violates the Fifteenth Amendment and the Voting Rights Act of 1965. For the reasons that follow, we find that the boundary violates both the plaintiffs’ constitutional and statutory rights and order appropriate relief.
I. BACKGROUND
A. Prior Proceedings
This case was filed in the United States District Court for the Northern District of Ohio in May of 1988. The plaintiffs sought a temporary restraining order preventing the results of the May, 1988 primary election for districts 52 and 53 from being certified and a preliminary injunction prohibiting further elections in those districts until they were brought into compliance with federal law. The district court denied the temporary restraining order, ordered the injunction hearing consolidated with the trial on the merits, and referred both to a magistrate. The magistrate recommended that relief be denied because the plaintiffs could not constitute a majority in a reconfigured district, and the district court adopted the recommendation. On appeal, a panel of the United States Court of Appeals for the Sixth Circuit reversed, holding that the district court should have examined the totality of the circumstances to consider whether the political process in the contested districts is equally open to minority voters. However, the Sixth Circuit voted en banc to vacate the panel opinion, and after additional arguments the court held that the subject matter was exclusively within the jurisdiction of a three-judge district court under 28 U.S.C. § 2284. The Chief Judge of the Sixth Circuit then convened the instant court.
B. Facts
The Ohio General Assembly is composed of two bodies, a 33-member senate, and a 99-member house of representatives. The method of apportioning the districts of the assembly is established by the Ohio Constitution. The current provision was enacted in 1967, after the previous plan was held unconstitutional. See Nolan v. Rhodes, 378 U.S. 556, 84 S.Ct. 1906, 12 L.Ed.2d 1034 (1964).
The House of Representatives is composed exclusively of single-member districts of substantially equal size. No district may have a population more than five per cent greater or less than the state’s population divided by 99, except that this *1047tolerance is increased to ten per cent if it will allow the creation of a house district consisting of a single county. Ohio Const. Art. XI, § 3, § 9. When possible, house districts must be drawn to contain one or more whole counties. Id. at § 7(A). However, when this is not possible, a district is formed by combining the areas of counties, townships, municipalities, and city wards, giving preference in the order named. Id. at § 7(B). If governmental units must be divided to create districts of substantially equal size, “only one such unit may be divided between two districts, giving preference in the selection of a unit for division to a township, a city ward, a city, and a village in the order named.” Id. at § 7(C).
Mahoning County lies in northeastern Ohio along the Pennsylvania border, about sixty miles southeast of Cleveland. Much of the county is rural. However, the northeastern quadrant of the county is dominated by the City of Youngstown. Youngstown is roughly rectangular in shape, about half again as wide as it is tall, except that the southeast corner of the rectangle, approximately one-ninth of the total, falls outside the corporate limits. That area is occupied by Campbell City (formerly known as East Youngstown), Struthers, and Coitsville Township. The Mahoning River bisects the city, entering Youngstown in the northwest corner and winding its way southeast to Campbell City before it wanders into Pennsylvania.
In 1981, the population of Mahoning County was 289,487. The ideal house district population was 109,065, and therefore the maximum population for a house district that was not a single county was 114,518. Consequently, Mahoning County was required to contain two whole house districts and share one house district with a neighboring county. Additionally, the City of Youngstown, which had a population of 115,427, had to be divided among two districts.
The populations and racial balances of the districts established by the 1981 Apportionment Board are set forth below:
DISTRICT 52
Total Black % Black
City of Youngstown Ward 2 16,597 10,341 (62.3%)
Ward 7 17,485 182 (1.04%)
Campbell City 11,619 1,173 (10.1%)
Boardman Twp. (Part) 10,881 52 (0.48%)
Other cities and townships 54,793 578 (1.05%)
Total 110,975 12,326 (11.11%)
DISTRICT 53
City of Youngstown Ward 1 16,667 9,791 (58.74%)
Ward 3 16,761 7,528 (44.91%)
Ward 4 16,168 339 ( 2.1%)
Ward 5 16,430 4,985 (30.3%)
Ward 6 15,719 5,312 (33.79%)
Boardman Twp. (Part) 30,952 173 ( 0.56%)
Total 112,697 28,128 (24.96%)
The plaintiffs, black voting age residents of Districts 52 and 53, assert that the apportionment between these districts unlawfully dilutes the black vote. They assert that the boundary between the two districts divides the black population at its point of greatest concentration in a ratio of 35:65. Plaintiffs seek a reapportionment *1048plan which would allocate ninety-nine per cent of Mahoning County’s black residents to District 53. They would do this by allocating Youngstown Ward 2 and Campbell City, currently assigned to District 52, to District 53. In exchange, they would assign all of Boardman Township to District 52. Maps showing the configurations of both the current and proposed districts may be found in Appendix I.
II. DISCUSSION
Plaintiffs allege that the boundary at issue violates both the Fifteenth Amendment and the Voting Rights Act of 1965. The Fifteenth Amendment claim must be heard by a three-judge district court pursuant to 28 U.S.C. § 2284(a). Armour v. Ohio, 925 F.2d 987 (6th Cir.1991). However, once convened, “the jurisdiction of the District Court so constituted ... extends to every question involved, whether of state or federal law, and enables the court to rest its judgment on the decisions of such of the questions as in its opinion effectively dispose of the case.” Sterling v. Constantin, 287 U.S. 378, 393-94, 53 S.Ct. 190, 193, 77 L.Ed. 375 (1932); U.S. v. Georgia Public Service Commission, 371 U.S. 285, 287-88, 83 S.Ct. 397, 399, 9 L.Ed.2d 317 (1963) (“Once convened the case can be disposed of below or here on any ground, whether or not it would have justified the calling of a three-judge court.”) Therefore, consistent with the judicial preference for resolving cases without reference to questions arising under the Federal Constitution whenever possible, Hagans v. Lavine, 415 U.S. 528, 546-47, 94 S.Ct. 1372, 1383-84, 39 L.Ed.2d 577 (1974), we will first address the statutory claim.
A
Section 2 of the Voting Rights Act of 1965 was originally viewed as coextensive with the prohibition against discrimination contained in the text of the Fifteenth Amendment. However, after the United States Supreme Court ruled in City of Mobile, Alabama v. Bolden, 446 U.S. 55, 60-61, 100 S.Ct. 1490, 1496, 64 L.Ed.2d 47 (1980), that a plaintiff must show discriminatory intent to prevail on a Fifteenth Amendment claim, Congress amended Section 2 “to make clear that plaintiffs need not prove a discriminatory purpose in order to establish a violation.” S.Rep. No. 97-417, p. 27, U.S.Code Cong. & Admin.News 1982, pp. 177, 204, quoted in Chisom v. Roemer, — U.S. -, -, 111 S.Ct. 2354, 2364, 115 L.Ed.2d 348 (1991).
Section 2 as amended reads as follows:
Denial or abridgement of right to vote on account of race or color through voting qualifications or prerequisites; establishment of violation
(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 4(f)(2) [42 U.S.C. § 1973(f)(2)], as provided in subsection (b).
(b) A violation of subsection (a) is established if, based on the totality of the 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) in that its members have less opportunity than other members of the electorate to participate in the political processes 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 amending Section 2, Congress intended to provide two separate claims for relief:
Plaintiffs must either prove such intent, or alternatively, must sAr/r that the challenged system or practice, in the con*1049text of all the circumstances in the jurisdiction in question, results in minorities being denied equal access to the political process.
S.Rep. No. 97-417, p. 27, U.S.Code Cong. & Admin.News 1982, pp. 177, 205, (emphasis supplied), quoted in Chisom v. Roemer, — U.S. -, -, 111 S.Ct. 2354, 2363, 115 L.Ed.2d 348 (1991). But see Chisom, — U.S. at-, 111 S.Ct. at 2369 (Scalia, J., dissenting) (“As currently written, the statute proscribes intentional discrimination only if it has a discriminatory effect, but proscribes practices with discriminatory effects whether or not intentional.”) We first turn our attention to the plaintiffs’ claims under the results test.
The results test under the Voting Rights Act as amended is meant to restore the preMobile legal standard which governed cases challenging election systems or practices as an illegal dilution of the minority vote. S.Rep. No. 97-417, p. 27, quoted in Chisom v. Roemer, — U.S.-,-, 111 S.Ct. 2354, 2364 (1991). This test asks whether “as a result of the challenged practice or structure, plaintiffs do not have an equal opportunity to participate in the political process and to elect candidates of their choice.” This determination requires a searching practical evaluation of past and present reality in the region at issue.
The Senate Report lists several typical factors that may be used to establish unequal access to the political processes:
1. the extent of any history of official discrimination in the state or political subdivision that touched the right of the members of the minority group to register, to vote, or otherwise to participate in the democratic process;
2. the extent to which voting in the elections of the state or political subdivision is racially polarized;
3. the extent to which the state or political subdivision has used unusually large election districts, majority vote requirements, antisingle shot provisions, or other voting practices or procedures that may enhance the opportunity for discrimination against the minority group;
4. if there is a candidate slating process, whether the members of the minority group have been denied access to that process;
5. the extent to which members of the minority group in the state or political subdivision bear the effects of discrimination in such areas as education, employment and health, which hinder their ability to participate effectively in the political process;114
6. whether political campaigns have been characterized by overt or subtle racial appeal;
7. the extent to which members of the minority group have been elected to public office in the jurisdiction.
S.Rep. No. 97-417, pp. 28-29, U.S.Code Cong. & Admin.News, 1982, pp. 177, 207, footnotes omitted, citing with approval White v. Regester, 412 U.S. 755, 93 S.Ct. 2332, 37 L.Ed.2d 314 (1973), and Zimmer v. McKeithen, 485 F.2d 1297 (5th Cir.1973) (en banc), aff'd sub nom. East Carroll Parish School Board v. Marshall, 424 U.S. 636, 96 S.Ct. 1083, 47 L.Ed.2d 296 (1976). See also Thornburg v. Cingles, 478 U.S. 30, 46, 106 S.Ct. 2752, 2764, 92 L.Ed.2d 25 (1986). The Senate Committee noted that in some cases the following factors also had probative value:
whether there is a significant lack of responsiveness on the part of elected officials to the particularized needs of the members of the minority group;
whether the policy underlying the state or political subdivision’s use of such voting qualification, prerequisite to voting, or standard, practice or procedure is tenuous.
S.Rep. No. 97-417, p. 29, U.S.Code Cong. & Admin.News, 1982, pp. 177, 207.
*1050Finally, the Committee indicated that these factors are not exclusive, and that there is no requirement that any particular factors be proven or that a majority of them point one way or another. Id. Instead, the court must decide based on the totality of the circumstances whether the voting strength of minority voters is “minimized or cancelled out.” Id. at 29, n. 118.
The Supreme Court’s recent decision in Thornburg v. Gingles also offers guidance on the use of the results test. In that case, the Court concluded that the Senate Report places three limitations on the circumstances under which a Section 2 violation may be proven:
First, electoral devices, such as at-large elections, may not be considered per se violative of § 2. Plaintiffs must demonstrate that, under the totality of the circumstances, the devices result in unequal access to the electoral process. Second, the conjunction of an allegedly dilutive electoral mechanism and the lack of proportional representation alone does not establish a violation. Third, the results test does not assume the existence of racial bloc voting; plaintiffs must prove it.
Thornburg v. Gingles, 478 U.S. 30, 46, 106 S.Ct. 2752, 2764, 92 L.Ed.2d 25 (1986).
In the instant case, the “allegedly dilutive electoral mechanism” is the placement of a district boundary. Therefore, applying the framework supplied by the Supreme Court and the Senate Report, the plaintiffs must establish that due to racial bloc voting and the totality of the past and present circumstances of blacks in Mahoning County, the district boundary minimizes or cancels out the voting strength of the minority population.
However, the defendants argue that in this case, there is no need to examine the totality of the circumstances because the plaintiffs do not have a large enough population to constitute a majority in a single-member district, however drawn. They assert that the Supreme Court implicitly established this size as a pre-condition to all challenges to district configurations in Thornburg v. Gingles. In Gingles, the Supreme Court stated that “unless there is a conjunction of the following circumstances, the use of multimember districts generally will not impede the ability of minority voters to elect representatives of their choice.” Id. at 48, 106 S.Ct. at 2765 (emphasis supplied). The Court then set forth three pre-conditions to a Section 2 challenge to the use of multi-member districting:
First, the minority group must demonstrate that it is sufficiently large and geographically compact to constitute a majority in a single-member district. If it is not, as would be the case in a substantially integrated district, the multi-member form of the district cannot be responsible for the minority voters’ inability to elect its candidates.17 Second, the minority group must be able to show that it is politically cohesive. If the minority group is not cohesive, it cannot be said that the selection of a multi-member electoral structure thwarts distinctive minority group interests. Third, the minority must be able to show 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.
Id. at 50-51,106 S.Ct. at 2765-67 (emphasis in original) (citations omitted).
Defendants assert that the Court’s reasoning compels the extension of these preconditions to all results test cases, adopting the logic of the three-judge district court in the Gingles case:
*1051Short of [the majority] level, there is no principled basis for gauging voting strength, hence dilution of that strength. Nothing but raw intuition could be drawn upon by courts to determine in the first place the size of those smaller aggregations having sufficient group voting strength to be capable of dilution in any legally meaningful sense----
... There must obviously be some size (as well as dispersion) limits on those aggregations of voters to whom the concept can properly be applied. We do not readily perceive the limit short of the effective voting majority level that can rationally be drawn and applied.
Gingles v. Edmisten, 590 F.Supp. 345, 381 (1984).
Defendants also argue that these conditions should apply to challenges to single-member districts because if the government can be forced to defend a lawsuit protesting the configuration of single-member districts when the plaintiff group would not be large enough to bring suit over a multi-member districting plan due to the Gingles pre-conditions, then the government is effectively being punished for adopting a form of districting that is generally considered to be more favorable to minority groups.
We do not agree with defendants’ analysis. In establishing threshold conditions for challenges to multi-member districts, the Court was responding to Congress’ express concern that multi-member districts would be subject to challenge any time that minorities are not elected in direct proportion to their population. As the Court noted in Gingles, the multi-member form of districting by itself infringes on a minority group’s opportunity to participate in the political process and elect a candidate of its choice only if the group would otherwise be guaranteed the opportunity to control at least one single-member district. The Court established threshold conditions for challenges to multi-member districts in order to enable governments to maintain such systems without the constant threat of time-consuming and expensive litigation.
Furthermore, the Court expressly limited the application of these pre-conditions to situations in which plaintiffs were challenging only the multi-member districting.1 In footnotes to its opinion, the Court noted that there were issues implicated by its interpretation of the Voting Rights Act which were not before the Court and which the Court therefore would not address. For example, the Court noted that:
... We have no occasion to consider whether § 2 permits, and if it does, what standards should pertain to, a claim brought by a minority group, that is not sufficiently large and compact to constitute a majority in a single-member district, alleging that the use of a multimember district impairs its ability to influence elections.
We note also that we have no occasion to consider whether the standards we apply to respondents’ claims that multimember districts operate to dilute the vote of geographically cohesive minority groups, that are large enough to constitute majorities in single-member districts and that are contained within the boundaries of challenged multi-member districts, are fully pertinent to other sorts of vote-dilution claims, such as a claim alleging that the splitting of a large and geographically cohesive minority between two or more multimember or single-member districts resulted in the dilution of the minority vote.
... In a different kind of ease, for example a gerrymander case, plaintiffs might allege that the minority group that is sufficiently large and compact to constitute a single-member district has been split between two or more multi-member or single member districts, with the effect of diluting the potential strength of the minority vote.
*1052Id. at 46 n. 12, 106 S.Ct. at 2764 n. 12; at 50 n. 16, 106 S.Ct. at 2767 n. 16. (emphasis added).
Additionally, the Court has since suggested that a dilution of minority influence may be sufficient to sustain a Section 2 results claim. In Chisom v. Roemer, — U.S.-, 111 S.Ct. 2354, 115 L.Ed.2d 348 (1991), the Supreme Court held that Section 2 of the Voting Rights Act applied to elected judges. In the course of its analysis, the Court stated that in order to establish a Section 2 claim, the plaintiffs must show both that they have less opportunity to participate in the political process and that they have less opportunity to elect representatives of their choice. Id. at-, 111 S.Ct. at 2364. Justice Scalia dissented, arguing that this reading of the statute would leave “minorities who form such a small part of the electorate in a particular jurisdiction that they could on no conceivable basis ‘elect representatives of their choice’ ” entirely without Section 2 protection and that they could therefore be denied equal opportunity to participate in the political process with impunity. Id. at -, 111 S.Ct. at 2371 (Scalia, J. dissenting). The majority responded that Justice Scalia’s argument “rest[ed] on the erroneous assumption that a small group of voters can never influence the outcome of an election.” Id. at- n. 24, 111 S.Ct. at 2365 n. 24.
Based on these statements in Chisom and the Court’s express disclaimers regarding the scope of its decision in Gingles, we cannot conclude that the Court intended the Gingles pre-conditions for challenges to multi-member districting schemes to apply to all Section 2 challenges.2 Therefore, we do not adopt them for the racial gerrymandering claim at issue here. While the bright-line test advocated by the defendants would indeed be simpler than a case by case analysis of the totality of the circumstances,
[t]he standard that should be applied in litigation under § 2 is not at issue here. Even if serious problems lie ahead in applying the totality of the circumstances described in § 2(b), that task, difficult as it may prove to be, cannot justify a judicially created limitation on the coverage of the broadly worded statute, as enacted and amended by Congress.
Chisom v. Roemer, — U.S.-,-, 111 S.Ct. 2354, 2368, 115 L.Ed.2d 348 (1991).
Moreover, we reject defendants’ argument that the state is being punished for adopting a single-member districting plan. While single-member districting is generally more favorable to large, compact minority populations, single-member districting plans possess their own unique opportunities for abuse, of which racial gerrymandering is a prime example.3 We cannot agree with the defendants that a government may with impunity divide a politically cohesive, geographically compact minority population between two single member districts in which the minority vote will be consistently minimized by white bloc voting merely because the minority population does not exceed a single district’s population divided by two. Accordingly, we proceed to an analysis of the totality of the circumstances in Mahoning Valley to determine whether the plaintiffs have proven that the boundary lines at issue minimize or cancel out the voting strength of the black community.
*1053 Totality of the Circumstances
At first blush, plaintiffs do not appear to have presented a very strong case. There are no allegations that laws in Ma-honing County have ever prohibited blacks from voting or from registering to vote. Additionally, none of the election procedures frequently used to discriminate against minorities are present here: candidates are elected from single member districts, there is no majority vote requirement and no formal slating process, and only 150 signatures are needed to obtain a place on the ballot. Therefore, the only allegation of de jure discrimination is that the boundary line at issue divides the black community.
However, the Senate Report to the Voting Rights Act directs us to examine de facto discrimination as well as that authorized by law. We begin this analysis with an examination of the history of the area in question. Although the boundary line at issue was drawn during a statewide reapportionment, the focus of the results test is on the plaintiffs rather than the drafters, and therefore our inquiry does not focus on the black experience in the entire state but rather on the political and social reality local to the Mahoning Valley. Because more than ninety-eight per cent of Mahoning County’s black population resides within the corporate limits of Youngstown, much of our discussion will necessarily be concerned with that city.
History of the Mahoning Valley
The relevant history of the Mahoning Valley is developed in detail in the record before us. It is thus established that in 1880 the City of Youngstown had a population of approximately 15,000. By the turn of the century, the population had tripled to 45,000, and by 1930 the city had swollen to over 170,000 residents due to the burgeoning steel industry that had grown up along the banks of the Mahoning River. The ethnic composition of the area reflected a similarly dramatic change. In the nineteenth century, Youngstown was composed predominantly of “old-stock” Americans of British and German descent. However, the steel industry attracted immigrants from a variety of European countries, to the point that in 1920, more than 60% of the city’s population were foreign born or first generation Americans.
During the same time period, the steel mill owners actively recruited rural southern blacks to work in the mills. Blacks were barred from membership in the steelworkers labor unions. Therefore, steel mill owners hired blacks to quell labor unrest, to break strikes, and to work at undesirable jobs that union members would not take. As a result of the recruitment by the steel mills, the black population of Youngstown increased at an even greater rate than that of the general population, jumping from less than two per cent of the population in 1900 to 6,750 in 1920 and doubling again to 14,500, over eight per cent of the city’s population, by 1930. The black workers formed communities near the steel mills that employed them, settling along the river banks. Ultimately, the steel companies began to provide housing which the workers could rent and eventually own. The company housing segregated the workers by race and national origin.
The influx of blacks combined with religious and cultural differences between the primarily Protestant original settlers and the Catholic and Jewish immigrants provided fertile ground for the organization of the Ku Klux Klan, which began to organize in Youngstown in 1922. In 1923, the year that Youngstown obtained a home rule charter, see Ohio Const., Art. XVIII, § 7, the Klan functioned as the city’s only political party, appointing precinct workers and nominating candidates.4 That year its can*1054didates won six of the seven council seats, the mayoral seat, and the majority of seats on the board of education. The Klan also was responsible for the division of the city into seven wards. Each ward began at the river and moved outward, in a wedge shape, thereby dividing the mill laborers who resided along the river banks and submerging them into the general population. These wards exist essentially unchanged today.
While the Klan fell from power almost as quickly as it rose, disappearing as a major influence by 1926, racial segregation and inequality continued. Discrimination pervaded all aspects of Youngstown life. The police routinely rounded up blacks in the 1920’s as a means of investigating crimes. Theaters directed black patrons to the balcony. Most restaurants denied them any but carryout service. Hotels did not always permit blacks to register. All of the city-operated swimming pools but one were restricted for white use.5
Discrimination was also conspicuous in city employment practices. Prior to 1960, there were no black members of the fire department. Blacks fared little better in Youngstown’s Police Department: in 1976, their relative low numbers and lack of advancement opportunities were such that a number of black police officers and candidates for appointment to the department filed a lawsuit alleging racial discrimination in the department’s hiring and promotional practices. Williams v. Vukovich, 720 F.2d 909 (6th Cir.1983). In that same year, the Office of Revenue Sharing, in response to a complaint lodged against the department, launched an investigation and found that, while Youngstown boasted a black population of 25 percent, only 7.6 percent of the police department was black. Id. at 914 n. 5. The investigation further uncovered that only two (6 percent) of the thirty-four officers hired the previous year were black, and that only 5 percent of the sergeants and 2.6 percent of the detectives on the police force were black.6
Discrimination was also evident in the city’s school system. The Youngstown Board of Education had a policy against hiring black high school teachers until 1956. Black teachers also found it difficult to obtain administrative positions.7 In 1977, the United States District Court for the Northern District of Ohio specifically found that the Youngstown public schools were racially segregated, although no relief was ordered because the segregation was not proven to result from discriminatory intent. See Alexander v. Board of Education, 675 F.2d 787, 795 n. 7 (6th Cir. 1982). The court also found that the school system disproportionately assigned black teachers and administrators to predominantly black schools and ordered appropriate relief. Id. at 790 & n. 2.
In addition to city employment, Youngstown blacks have faced discrimination in other fundamental areas throughout this century. Restrictive covenants prohibited real estate from being sold to blacks. Even after these covenants were declared unenforceable, Shelley v. Kraemer, 334 U.S. 1, 68 S.Ct. 836, 92 L.Ed. 1161 (1948), black families were unable to purchase *1055housing in white neighborhoods.8 Consequently, Youngstown’s housing pattern has remained intensely segregated. While the fourth and seventh wards of the city are less than two per cent black, the 1980 and 1990 census data show that many areas in Youngstown are more than eighty per cent black.9 The court takes judicial notice of the segregated white-black housing assignment policies maintained by the Youngstown Metropolitan Housing Authority. This was pursuant to the policy of the Federal Public Housing Administration that existed from the 1930’s into the 1960’s. Moreover, social activities remain closed to blacks. At the time of the trial in 1988, no blacks had ever been members of any of the city’s three country clubs or of the Youngstown Club, the city’s primary business club.
Effects of Discrimination
The effects of discrimination are apparent in the data compiled during the U.S. Department of Commerce’s 1980 Census of Population and Housing (Census) and in studies conducted by plaintiffs’ expert witness, Dr. Terry Buss.10 The Census showed that in Mahoning County, the mean income for blacks in 1979 was $14,118, substantially lower than the mean income for whites of $20,259. Likewise, the Census showed that 27.3% of black families lived below the poverty level while the corresponding figure for whites was only 5.8%. Of the blacks in Mahoning County 25 years of age or older, nearly half (48.6%) had not completed high school, and only sixteen per cent had completed one or more years of college. For whites in Mahoning County, less than one-third had not completed high school, and almost one-quarter had completed at least one year of college. Finally, the Census showed that the median value of housing for blacks in Mahoning County in 1980 was $20,200, while the median value of housing for whites was $39,-700. See Appendix II for census data.
Similarly, a 1984 study by Dr. Buss, entitled “Inequities in the Distribution of Unemployment in the Youngstown/Warren SMSA”,11 indicated that blacks were much more likely to be unemployed than whites. This study, prepared for the Ohio Bureau of Unemployment Services using the methodology used by the U.S. Bureau of the Census during its monthly population survey, showed that the unemployment rate for blacks was 37.2% while that for whites was only 15.0%. Young blacks, ages 34 and under, had the highest unemployment rate, nearly 50%. Additionally, blacks comprised more than one-quarter of the discouraged workers,12 nearly three times the amount that would have been equivalent to whites given the racial makeup of the region. Dr. Buss testified that he had studied black and white unemployment data for the period between his 1984 study and the time of the trial and concluded that while unemployment for both groups vacillates, *1056blacks consistently experienced greater unemployment rates than whites, with the rate for blacks reaching double that of whites on some occasions.13
While the effects of discrimination are apparent in areas of education, employment, income and health care as discussed above, the effect of discrimination against blacks in the Mahoning Valley is perhaps most apparent in political life. In the most vivid example, McCullough Williams, a leader in Youngstown’s black community and one of the first black members of Youngstown’s school board, testified that he received threats as a result of his civil rights activities. These threats culminated in his house being bombed in 1970. Compelling testimony was also provided by Herman “Pete” Starks regarding his candidacy for Mayor in 1985. Starks was endorsed by the party in his campaign for the Democratic primary, the only time a black candidate has received a party endorsement in a city-wide or county-wide election. Nonetheless, throughout the primary race, the media focused on Starks’ race, consistently describing him as the black candidate for Mayor. Additionally, the opposition campaign used overt racial appeals. They campaigned door-to-door telling voters that if Starks was elected, his cabinet would be black. In addition, a soundtrack canvassed the city announcing that “If Pete Starks is elected Mayor, we will have a black police chief, we will have a black fire chief, and we cannot have that.” Furthermore, although party rules required party officials to support the endorsed candidates, and although each precinct had committeemen who were paid to campaign for the endorsed candidates, Starks received zero votes in four precincts, including one precinct where two party officials resided. No sanctions were taken by the party against the officials or the committeemen who refused to support Starks.
Finally, the record establishes that blacks in Mahoning County participate in the political process at a lower rate than whites. From the data previously discussed indicating disparity in income, education, and employment between whites and blacks in Mahoning County, we conclude that the depressed minority political participation is the result of past discrimination. See S.Rep. No. 97-417, 28 n. 114, U.S.Code Cong. & Admin.News 1982, pp. 177, 207, quoted supra, at p. 1049.
Black Candidates in Elections
In the history of Mahoning County, eight blacks have run for state representative. With the exception of W.R. Stewart’s election in Í904, before the population boom and the emergence of the Ku Klux Klan, none has been successful. No black has ever won a county-wide election. Only one black candidate has ever won a city-wide election other than for school board. The Honorable Lloyd Haynes was appointed to municipal court in 1972, and subsequently won election in 1978 and again in 1984. Five black candidates have won elections for the Youngstown school board. The school board election is conducted on a multi-member city-wide basis.
Racial Polarization and Political Cohesiveness
The Supreme Court’s decision in Gingles indicates that plaintiffs must prove white bloc-voting and black political cohesiveness in order to prevail on a Section 2 claim. In this case, the plaintiffs have proven both.
Plaintiffs’ expert witness, Dr. Buss, testified first to general trends in Mahoning County. Overall, blacks in Mahoning County will vote for the Democratic party candidate in the general election at a rate of between eighty and ninety per cent. Ten per cent of white voters in Mahoning County vote consistently for the Republican ticket. The remaining white voters tend to split evenly between independent candidates and the Democratic candidate.
Dr. Buss collected and evaluated data showing the voting patterns in the City of *1057Youngstown for 14 elections in which a black candidate had sought election. Dr. Buss conducted extreme case analysis and regression analysis to determine whether blacks and whites in Youngstown differed in their voting behavior. These techniques yielded data concerning the voting patterns of each racial group, including estimates of the percentages of members of each race who voted for each candidate.14 His analysis showed that the relationship between the candidate’s race and the race of the voter was consistently near linear15, indicating that there is a relationship between tine race of the voter and the likelihood that he will vote for a black candidate. Additionally, the analysis showed that the percentage of votes received increased as the percentage of blacks in the precinct increased.16 Homogeneous precinct analysis, or extreme case analysis, produced similar results.
Dr. Buss’s analyses showed that, on average, black voters supported black candidates at a rate of almost eighty per cent, while white voters supported black candidates at a rate of about twelve per cent.17 *1058The rate of white support dropped to less than seven percent when school board races were excluded.
From these data, and Dr. Buss’s observation that black voters in Youngstown generally vote for the Democratic candidate at a rate between eighty and ninety per cent, we find that plaintiffs constitute a politically cohesive voting unit. However, white voters in Youngstown do not support black candidates. Therefore, in the usual course of events white bloc voting will result in the defeat of the minority group’s candidates.
Responsiveness
The Senate Report indicates that another factor that is sometimes useful in the totality of the circumstances analysis is the responsiveness of elected officials to plaintiffs’ group. In this case, we find that the state representatives in Districts 52 and 53 have not been sensitive to the needs of the black community in Youngstown. First, we note that the representatives have little incentive to consider black voters. As demonstrated above, the black community is safely Democratic. In the absence of a black candidate, over eighty per cent of black voters will vote strictly along party lines. However, white voters in Mahoning County are less consistent, and therefore the representatives must cater to their needs in order to secure re-election. The uncontradicted evidence suggests that this is in fact the practice in Mahoning County: very little campaigning for the state representative positions has been directed toward the black community. Furthermore, black voters in these districts testified that their representatives are not responsive to minority issues, and surveys by Dr. Buss of black residents and leaders in the black community show that the representatives are perceived as indifferent to the needs of the minority community. These surveys also showed that black citizens feel that their representatives do not seek input from the black community. For example, one of Youngstown’s state representatives introduced a bill into the state legislature to eliminate one of Youngstown’s three municipal judgeships, the only judgeship ever held by a black, and the only city-wide position to which a black had ever been elected by a majority vote.18
State’s Interest
Finally, the Senate Report suggests that this court may find it useful to examine the policy underlying the state’s use of the challenged practice. Where the underlying policy is tenuous, the challenged practice is more likely to violate Section 2. In this case, there is simply no defensible basis for the current boundaries. The state’s apportionment policies established within its constitution require that the integrity of political subdivisions be respected whenever possible. In this case, those policies were clearly violated. Article XI of the state constitution states that a house district should be formed by combining the areas of governmental units. Ohio Const., Art. XI, § 7(B). When a unit must be divided in order to create house districts of substantially equal size, “only one such unit may be divided between two districts.” Id. at § 7(C). In this case, as more fully appears in Section B below, both the City of Youngstown and Boardman Township were divided between districts 52 and 53. Since the current configuration of the districts violates the state’s own constitutional requirements, we find that the state has no interest in maintaining the current configuration.
Conclusions
From the totality of the above evidence, we conclude that segregation and racial discrimination have been a way of life in Mahoning County since blacks settled in *1059the area at the turn of the century. One effect of these practices has been to foreclose the area’s black residents from the political processes leading to nomination and election of candidates for any office in the Mahoning Valley, with a corresponding depression in black registration and voting rates. Additionally, the current configuration of the Ohio House of Representatives’ districts deprives blacks of the opportunity to elect a candidate of their choice in either house district. They cannot elect a black representative because white voters will not support a black candidate, whether or not he has a party endorsement, and blacks are not sufficiently numerous in either district to carry an election without white support. Therefore, white race-based bloc voting works in conjunction with the division of the black voters to permit and indeed compel political parties to ignore minority candidates, and to discourage black candidates from seeking office. Defendants argue that blacks have not run in the primary elections, and that therefore they have no basis for claiming that they cannot prevail in the current districts. However, we can take notice of the tremendous expense, monetary and otherwise, of running for office, and we find that the circumstances in these two districts are such that most blacks could not reasonably have believed that they would have a chance of making a substantial showing, much less of winning.
With regard to white candidates, political reality does not encourage candidates in the current district configurations to take positions that are responsive to the needs of the black community. In the absence of a black candidate, the Democratic candidate can be assured of more than eighty per cent of the black vote simply by virtue of his party endorsement, and the Republican candidate can be assured that he will receive less than ten per cent. Therefore, candidates from both parties must, and do, focus their campaigns on winning the “swing vote” — which in these districts is primarily composed of wealthy suburban whites. Given the totality of the circumstances in this region, we can take judicial notice that the expectations of wealthy suburban voters are qualitatively different from those of the impoverished, urban black community. As a result, we conclude that under the current configurations, because the black voting strength is divided into two districts each of which contains a large population of white suburban voters, black voters in Mahoning county do not have an equal opportunity to elect a candidate of their choice for state representative.
The next question that we address is whether black voters could elect a candidate of their choice in a redrawn district.19 For the reasons set forth below, we find that plaintiffs have shown that in the proposed district they will be able to elect a candidate of their choosing.
Defendants go to great lengths to demonstrate that based upon racial voting patterns plaintiffs will not be able to elect a black candidate without a majority of black voters in the redrawn district. However, defendants misapprehend the requirements of the Voting Rights Act. The issue is not whether the plaintiffs can elect a black candidate, but rather whether they can elect a candidate of their choice. We believe that they can. In a reconfigured district, plaintiffs will constitute nearly one-*1060third of the voting age population and about half of the usual Democratic vote. Therefore, the Democratic Party and its candidates will be forced to be sensitive to the minority population by virtue of that population’s size. Moreover, in a district composed only of Youngstown and Campbell, candidates and representatives will not find themselves in conflict between the interests of wealthy suburbs and the impoverished urban communities they serve. Since black voters consistently vote eighty to ninety per cent Democratic and white voters vote consistently almost fifty per cent Democratic, we find that plaintiffs could elect a candidate of their choice, although not necessarily of their race, in a reconfigured district.20
Therefore, based on our previous findings, we conclude that the plaintiffs have proven that the current configuration of Ohio House of Representatives Districts 52 and 53 violates the provisions of Section 2 of the Voting Rights Act of 1965, 42 U.S.C. § 1973.
B
Our finding that the plaintiffs have proven their claim under the results test of Section 2 of the Voting Rights Act would ordinarily allow us to proceed to the question of relief without reaching plaintiffs’ other claims. However, due to the paucity of authority regarding the application of the results test to minority populations that are not large enough to form a majority in a single-member district, we are compelled by principles of sound judicial administration to address plaintiffs’ second claim under the Voting Rights Act: that the state apportionment board intentionally split the black population of Mahoning County into two districts in order to dilute the effectiveness of the minority vote. Because this claim is indistinguishable from a claim under the Fifteenth Amendment, we address these claims together.
The Fifteenth Amendment states:
Section 1. Right of citizens to vote— Race or color not to disqualify.
The right of citizens 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.
Section 2. Power to enforce amendment.
The Congress shall have power to enforce this article by appropriate legislation.
This amendment prohibits states from intentionally discriminating on the basis of race in matters having to do with voting. City of Mobile v. Bolden, 446 U.S. 55, 61, 100 S.Ct. 1490, 1496, 64 L.Ed.2d 47 (1980).
Plaintiffs have produced strong evidence proving that the drafting of the boundary between District 52 and District 53 was racially motivated. First, Robert Dykes, a political consultant hired by the Ohio Democratic Party to draft an apportionment plan, testified that the team that designed the plan adopted by the state sought out information regarding the location of substantial black populations within the state. When the drafting team determined that there was a large enough black population to form a black majority for a district, they shifted boundaries to more closely approximate the black population. This was accomplished notwithstanding the mandate of the Ohio Constitution that “district boundaries established by the preceding appor*1061tionment board shall be adopted to the extent reasonably consistent with the requirements of section 3 of this article.” Art. XI, § 7(D). However, if the black population could not constitute a majority in a district, Dykes testified that the drafters did not attempt to conform the boundaries to black concentrations. Furthermore, the Secretary of State testified that the apportionment board was contacted by incumbents who wanted their districts “protected.” Finally, the evidence showed that the incumbent legislators were white, these incumbents were supported by white voters, voting in the districts was racially polarized, and the districts were safely Democratic.
From this evidence, we find that the line dividing Youngstown between districts 52 and 53, when it was originally drawn in 1971 and when it was left in place in 1981, was intended to split the black community in order to dilute the potential effectiveness of the black vote, to the obvious benefit of the incumbents. Although courts are reluctant to provide relief on claims that a district has been gerrymandered to protect an incumbent’s seat, see Davis v. Bandemer, 478 U.S. 109, 138-43, 106 S.Ct. 2797, 2813-15, 92 L.Ed.2d 85 (1986) and 478 U.S. at 143-60, 106 S.Ct. at 2815-24 (O’Connor, J., concurring), this rule does not hold when the manipulations were conducted on a race-conscious basis. Like the Seventh Circuit, we see “little point ... in distinguishing discrimination based on an ultimate objective of keeping certain white incumbents in office from discrimination borne of pure racial animus.” Ketchum v. Byrne, 740 F.2d 1398, 1406-10 (7th Cir. 1984), cert. denied, 471 U.S. 1135, 105 S.Ct. 2673, 86 L.Ed.2d 692 (1985). See also Garza v. City of Los Angeles, 918 F.2d 763, 771 (9th Cir.1990), cert. denied, — U.S. -, 111 S.Ct. 681, 112 L.Ed.2d 673 (1991) (Fifteenth Amendment violation was proven when officials chose to fragment the Hispanic vote in order to preserve incumbencies). See also Gomillion v. Lightfoot, 364 U.S. 339, 346, 81 S.Ct. 125, 130, 5 L.Ed.2d 110 (1960) (“When the legislature thus singles out a readily isolated segment of a racial minority for special discriminatory treatment, it violates the Fifteenth Amendment.”)
Furthermore, the districts as drawn in 1981 violate the express command in the Ohio Constitution that only one governmental unit be divided between two districts. The governmental unit that was unnecessarily divided was Boardman Township, an area with a population of 41,510 persons, only 225 of whom are black. It is apparent from the record that any reasonable division between the two districts that did not split any entity other than Youngstown would have created a district with a substantially greater black population than District 53 as currently drawn. Recognizing the admonition in Gomillion v. Lightfoot, 364 U.S. 339, 342, 81 S.Ct. 125, 127, 5 L.Ed.2d 110 (1960), that “the Fifteenth Amendment nullifies sophisticated as well as simple-minded modes of discrimination”, we find that the deliberate combination of over 30,000 persons from a 99% white township with areas of Youngstown that were nearly half black in flagrant disregard of the state constitution’s apportionment rules was not color blind. Accordingly, we find that the plaintiffs have proven that the current apportionment violates the Fifteenth Amendment.
C
The Supreme Court stated in Constantin, supra at page 1048 and Railroad Commission of California v. Pacific Gas & Electric Co., 302 U.S. 388, 58 S.Ct. 334, 82 L.Ed. 319 (1938) that once a three-judge court is properly convened, it has jurisdiction to determine “all the questions in the case, local as well as federal.” We recognized above that the boundary line at issue in this case violates the Ohio Constitution. While issues requiring the resolution of ambiguities in state statutes or constitutions are best reserved to state courts, there is no ambiguity here. The state constitution mandates that “only one [governmental] unit may be divided between two [house] districts.” Ohio Const. Art XI, § 7(C). The City of Youngstown was more than five per cent larger than the state’s population divided by 99, and it therefore *1062had to be divided to meet the requirements of Article XI, section 3, limiting the permissible variation in size between districts. Because Youngstown had to be divided, the state constitution compelled that all other governmental units in the two districts at issue be allocated wholly to one district or the other. The division of Boardman Township unquestionably violated this mandate.
III. RELIEF
Plaintiffs seek declaratory and injunctive relief from the boundary between Districts 52 and 53. In particular, plaintiffs request an order enjoining the State from holding future elections using the current boundaries and directing the State to adopt plaintiffs’ proposed districts. Finally, plaintiffs ask this court to order a special election to be held in November of 1991 to elect representatives from constitutionally drawn districts. We address these requests in sequence.
First, we find that plaintiffs are entitled to declaratory relief for the reasons discussed above, and accordingly, we declare that the current boundaries of Ohio House of Representatives Districts 52 and 53 violate the Voting Rights Act of 1965 as well as the constitutions of the United States and the State of Ohio. “[Ojnce a State’s legislative apportionment scheme has been found to be unconstitutional, it would be the unusual case in which a court would be justified in not taking appropriate action to insure that no further elections are conducted under the invalid plan.” Reynolds v. Sims, 377 U.S. 533, 585, 84 S.Ct. 1362, 1393, 12 L.Ed.2d 506 (1964). In this ease, the current house terms do not expire until January 1, 1993, and the Ohio Constitution requires a new statewide apportionment of the General Assembly to be published this year. Therefore, we find that there are no special circumstances which would compel us to withhold this form of injunctive relief. Consequently, we enjoin the defendants from using the current house district configurations for future elections.
However, we conclude that an order requiring the state to adopt plaintiffs’ proposed districts would not be appropriate at this time. “[Ljegislative reapportionment is primarily a matter for legislative consideration, and ... judicial relief [is] appropriate only when the legislature fails to reapportion according to federal constitutional requisites in a timely fashion after having had an adequate opportunity to do so.” Reynolds v. Sims, 377 U.S. at 586, 84 S.Ct. at 1394; see also Wise v. Lipscomb, 437 U.S. 535, 539-40, 98 S.Ct. 2493, 2496-97, 57 L.Ed.2d 411 (1978). The Ohio Constitution requires the apportionment board to publish a new apportionment plan for the Ohio General Assembly by October 5, 1991 based on the 1990 decennial census. We therefore retain jurisdiction of this case to consider further injunctive relief; however, we defer a hearing on this issue until after the October deadline has passed. If at that time the reapportionment plan has not been published, or if the board’s plan violates plaintiffs’ statutory or constitutional rights, we will then entertain an application for an order fashioning an appropriate remedy.
Finally, with regard to plaintiffs’ request for a special election from constitutionally drawn districts for the house seats expiring in January of 1993, equitable considerations require that this relief be denied. The burden of requiring the State to hold an election for these districts one month after the apportionment is published combined with the burden on the potential candidates far exceeds the burden on the plaintiffs of being represented for an additional year by legislators whom they might not have chosen had the election been held in constitutionally drawn districts. Moreover, while our broad remedial powers may permit us to order the apportionment published before the October deadline, we decline to do so. We do not believe that an earlier deadline would give the board adequate opportunity to fashion a statewide plan that meets the federal constitutional requirements announced in this opinion as well as the requirements imposed by the Ohio Constitution. Similarly, we decline to order a special election for the house seats at a later date. Although a later election would give the candidates and the state *1063time to prepare, the incremental cost of a special election imposes its own burdens, and the benefit to the plaintiffs diminishes as the time remaining in the existing house term expires. Therefore, we decline to compel a special election for the house districts at issue.
IV. ATTORNEY’S FEES
Section 1973Í of United States Code Title 42 provides:
(e) Attorney’s fees. In any action or proceeding to enforce the voting guarantees of the fourteenth or fifteenth amendment the court, in its discretion, may allow the prevailing party, other than the United States, a reasonable attorney’s fee as part of the costs.
Under this statute, a prevailing plaintiff should ordinarily recover an attorney’s fees unless special circumstances would render such an award unjust. Hensley v. Eckerhart, 461 U.S. 424, 429, 433 n. 7, 103 S.Ct. 1933, 1936, 1939 n. 7, 76 L.Ed.2d 40 (1983). We find that there are no circumstances that would render an award of fees unjust in this case, and we therefore grant plaintiffs’ request for fees and costs. Plaintiffs should submit a fee request with appropriate documentation to the court. Defendants will then have the opportunity to file objections to any of the charged hours if they so desire.
V. CONCLUSION
For the reasons stated above, we find that plaintiffs’ statutory and Fifteenth Amendment claims have merit, and we order the relief announced above.
While we retain jurisdiction of this case to consider plaintiffs’ request for an order directing the state to adopt plaintiffs’ proposed districts until after the state apportionment board publishes a new apportionment plan and to determine the amount of attorney’s fees, we find that there is no just reason for delay in the entry of judgment ordering declaratory relief and enjoining future elections using the current districts, and we therefore direct that judgment be entered on those claims pursuant to Rule 54(b).
So ordered.
*1064APPENDIX I
[[Image here]]
*1065[[Image here]]
*1066[[Image here]]
*1067[[Image here]]
*1068APPENDIX II *
The following data are excerpted from tables published in the 1980 Census of Population and Housing Report for the Youngstown-Warren, Ohio Standard Metropolitan Statistical Area.
Tables H-2 and H-3 Occupancy, Utilization and Financial Characteristics of Housing Units in Mahoning County
PERSONS IN UNIT
White Black
1 person ...................................................... 18,849 3,051
2 persons...................................................... 27,780 3,484
3 persons...................................................... 15,629 2,581
4 persons...................................................... 13,980 2,071
5 persons...................................................... 7,135 1,259
6 persons...................................................... 2,867 626
7 persons...................................................... 1,094 376
8 or more persons............................................. 457 237
Median, occupied housing units................................. 2.40 2.62
Median, owner-occupied housing units........................... 2.67 2.87
Median, renter-occupied housing units........................... 1.70 2.33
VALUE
Specified owner-occupied housing units.................... 58,276 6,707
Less than $15,000 ............................................ 3,546 1,916
$15,000 to $19,999 ............................................ 3,204 1,366
$20,000 to $24,999 ............................................ 4,537 1,121
$25,000 to $29,999 ............................................ 5,703 695
$30,000 to $34,999 ............................................ 6,535 548
$35,000 to $39,999 ............................................ 6,018 308
$40,000 to $49,999 ............................................ 10,467 379
$50,000 to $59,999 ............................................ 7,153 169
$60,000 to $79,999 ............................................ 7,490 162
$80,000 to $99,999 ............................................ 2,293 34
$100,000 to $149,999 .......................................... 1,038 6
$150,000 to $199,999 .......................................... 198 2
$200,000 or more............................................. 94 1
Median....................................................... $ 39,700 $ 20,200
CONTRACT RENT
Specified renter-occupied housing units.................... 20,498 5,595
Median....................................................... $ 180 $ 107
Tables P-12 & P-14 Social and Labor Force Characteristics for Mahoning County: 1980
SCHOOL ENROLLMENT
Persons 3 years old and over enrolled in school........... 61,122 12,625
Nursery School............................................... 2,699 606
Kindergarten................................................. 3,022 775
Elementary School (1 to 8 years).............................. 28,083 6,436
High School (1 to 4 years).................................... 16,121 3,236
College....................................................... 11,197 1,572
YEARS OF SCHOOL COMPLETED
Persons 25 years of age or older .......................... 153,435 21,364
Elementary: 0 to 4 years.................................... 3,564 1,518
5 to 7 years.................................... 7,813 2,076
8 years ........................................ 10,996 1,353
*1069White Black
YEARS OF SCHOOL COMPLETED
High School: 1 to 3 years.................................... 27,279 5,443
4 years ........................................ 65,766 7,527
College: 1 to 3 years.................................... 20,021 2,535
4 or more years................................ 17,996 912
LABOR FORCE STATUS
Persons 16 years and over................................. 191,484 28,328
Labor force.................................................. 110,316 14,822
Percent of persons 16 and over......................... 57.6 52.3
Employed................................................ 99,320 11,229
Unemployed.............................................. 10,923 3,583
Percent of civilian labor force........................... 9.9 24.2
LABOR FORCE STATUS IN 1979
Persons 16 years and over, in labor force in 1979.......... 119,214 16,321
Percent of persons 16 and over......................... 62.3 57.6
Worked in 1979 .............................................. 116,806 14,673
40 or more weeks.......................................... 86,678 9,607
Usually worked 35 or more hrs/week..................... 73,220 8,226
50 to 52 weeks............................................. 70,106 7,067
Usually worked 35 or more hrs/week..................... 60,844 6,179
With unemployment in 1979................................... 25,307 6,112
Percent of those in labor force in 1979.................. 21.2 37.4
Unemployed 15 or more weeks.............................. 9,176 3,225
Mean weeks of unemployment.............................. 14.5 20.6
Tables P-13 and P-15 Occupation, Income in 1979, and Poverty Status in 1979 for Mahoning County: 1980
Census Tracts
[400 or More White Persons and 400 or More of a Specified Racial Group] White Black
INCOME IN 1979
Households................................................ 87,969 13,752
Less than $5,000 ............................................. 10,352 3,922
$5,000 to $7,499 .............................................. 5,993 1,386
$7,500 to $9,999 .............................................. 6,260 1,159
$10,000 to $14,999 ............................................ 12,167 1,797
$15,000 to $19,999 ............................................ 13,312 1,691
$20,000 to $24,999 ............................................ 13,570 1,407
$25,000 to $34,999 ............................................ 15,889 1,509
$35,000 to $49,999 ............................................ 7,351 721
$50,000 or more.............................................. 3,075 160
Median....................................................... $ 18,401 $11,047
Mean........................................................ $ 20,259 $14,118
Families................................................... 67,398 10,356
Median income............................................... $ 21,245 $13,552
Mean income................................................. $ 23,176 $15,896
Unrelated individuals 15 years and over ......................................... 24,394 4,450
Median income............................................... $ 6,635 $ 4,452
Mean income................................................. $ 9,186 $ 6,722
Per capita income............................................ $ 7,323 $ 4,696
Households................................................ 87,969 13,752
With earnings...........................■..................... 69,349 9,338
Mean earnings ............................................. $ 21,099 $16,436
With Social Security income................................... 27,184 3,761
Mean Social Security income................................ $ 4,427 $ 4,131
With public assistance income................................. 5,465 4,112
Mean public assistance income.............................. $ 2,515 $ 2,546
*1070INCOME IN 1979 BELOW POVERTY LEVEL
White Black
Families.................................... 3,906 2,823
Percent below poverty level................ 5.8 27.3
Householder worked in 1979 ................... 1,926 848
INCOME IN 1979 BELOW POVERTY LEVEL
With related children under 18 years........... 2,877 2,477
Female householder, no husband present....... 1,663 2,073
Householder worked in 1979 ................. 655 562
With related children under 18 years......... 1,482 1,941
With related children under 6 yrs.......... 784 1,141
Householder 65 years and over................ 462 211
Unrelated individuals for whom poverty status is determined................ 5,285 1,860
Percent below poverty level................ 22.0 42.3
65 years and over ............................. 1,943 549
Persons for whom poverty status is determined............................... 18,818 12,238
Percent below poverty level................ 7.8 30.2
Related children under 18 years................ 6,418 5,828
Related children 5 to 17 years............... 4,655 3,959
60 years and over ............................. 3,850 1,311
65 years and over ........................... 2,799 884
Percent of persons for whom poverty status is determined
Below 75 percent of poverty level............ 5.4 25.0
Below 125 percent of poverty level........... 10.9 35.6
Below 200 percent of poverty level........... 23.3 51.7
The courts have recognized that disproportionate educational employment [sic], income level and living conditions arising from past discrimination tend to depress minority political participation. Where these conditions are shown, and where the level of black participation in politics is depressed, plaintiffs need not prove any further causal nexus between their disparate socio-economic status and the depressed level of political participation, [citations omitted].
... The single-member district is generally the appropriate standard against which to measure minority group potential to elect, because it is the smallest political unit from which representatives are elected. Thus, if the minority group is spread evenly throughout the multimember district, or if, although geographically compact, the minority group is so small in relation to the surrounding white population that it could not constitute a majority in a single-member district, these minority voters cannot maintain that they would have been able to elect representatives of their choice in the absence of the multimember electoral structure.
. Indeed, there is nothing in Gingles that would prevent minority plaintiffs in a multi-member district that prohibited "bullet” voting or that required a majority vote to prevail in an election from challenging the latter requirements even if the group were not large enough to challenge the multi-member form of the district. See Gingles, 478 U.S. at 56, 106 S.Ct. at 2769.
. We recognize that some courts have applied the Gingles preconditions to cases in jurisdictions where a majority vote was not required to win the election. Brewer v. Ham., 876 F.2d 448, 455 (5th Cir.1989); McNeil v. Springfield Park District, 851 F.2d 937, 944 (7th Cir.1988), cert. denied, 490 U.S. 1031, 109 S.Ct. 1769, 104 L.Ed.2d 204 (1989); Collins v. Norfolk, 883 F.2d 1232 (4th Cir.1989), cert. denied, — U.S.-, 111 S.Ct. 340, 112 L.Ed.2d 305 (1990); Buckanaga v. Sisseton Indep. School Dist., 804 F.2d 469, 475 (8th Cir.1986). Because our analysis in this case is not founded on Ohio’s plurality election requirement, we need only recognize that these cases involved challenges to multi-member districting and therefore are not relevant to the issues presented in this case.
. Indeed, given the unique facts of this case, the plaintiffs might well benefit from a multi-member districting scheme because in a multi-member district the plaintiff group could not easily be divided.
. The primary focus of the Klan in Youngstown appears to have been on moral reforms — enforcement of prohibition and the Sunday blue laws. However, white supremacy and racial segregation were still part of the platform. For example, the Youngstown Vindicator reported that blacks and ethnic women were harassed when they tried to register to vote in 1923. Similarly, Herbert Armstrong testified that Klansmen wearing white hoods and riding horses had driven his father, a black merchant, into bankruptcy in 1926 because his clothing store was primarily patronized by white men. He also testified that he had personally observed *1054Klan activity in Youngstown, including the burning of crosses. It appears that the main reason that racial issues did not have the primacy in Youngstown that they had in other areas of the country dominated by the Klan was that there was a system of de facto racial segregation firmly in place before the Klan arrived.
. Eventually, another white pool was constructed less than 300 feet away from the only black pool.
. On the date the Williams case was filed, only 6.8 percent of the police department was black. A decade later, the city entered into a consent decree agreeing to an affirmative hiring and promotion plan.
. Dr. Herbert Armstrong testified that he was not appointed to a position as principal until 1965, nine years after qualifying for the position and thirteen years after joining the school system, despite a Master’s Degree in School Administration and Bachelor of Science degrees in both Education and Business Administration. He testified that the average time for a non-black teacher of similar background and experience to become a principal would have been about five years. Dr. Armstrong’s eventual appointment made him the school system’s first black elementary school principal.
. For example, McCullough Williams testified that he had contracted to purchase a house on Volney Road and obtained financing. However, at closing, the realtor refused to sell him the property directly, stating that he would lose his real estate license if he did so. He suggested that Williams instead place the property in trust or have it purchased by a white person. Williams testified that he attempted to look at a home on Goleta Avenue and that the owner refused to show the house. Finally, he testified that when he sought to purchase property on which to establish a funeral home, there were only three properties that were properly zoned that realtors would show him. Similarly, A.J. Carter testified that when he came to Youngstown in 1925, he moved in with his uncle because people refused, explicitly on the basis of race, to rent to him.
. The district court in the school desegregation case made the same observation in 1977. See Alexander v. Youngstown Board of Education, 675 F.2d 787, 795 n. 7 (6th Cir.1982) (quoting 454 F.Supp. 985, 1066).
. At the time of trial, Dr. Buss was the Director of the Department of Urban Studies at the University of Akron. Dr. Buss testified that he had conducted between three and four hundred surveys related to the demographics and economy of the Mahoning Valley area since obtaining his Ph.D. in political science and mathematics in 1976.
. Standard Metropolitan Statistical Area, as classified by the U.S. Bureau of the Census during the 1980 Census of Population and Housing.
. Discouraged workers are persons who want to work but have not sought employment in the last four weeks.
. Dr. Buss also conducted a survey of "Health and Human Service Needs of People in Poorer Neighborhoods" in 1985. This study focused on the twenty poorest census tracts in Youngstown. The demographics of the sample indicated that almost sixty per cent of the persons surveyed were black. The survey indicated that blacks in poor neighborhoods in Youngstown were alienated from the health and human services care delivery system.
. There are, however, certain deficiencies in plaintiffs’ data. First, the trial revealed an error in the ecological bivariate regression analysis. This analysis determines whether there is a relationship between two variables, for example the-race of the candidate and the race of the voter, and if so, attempts to quantify the correlation. One common approach to this analysis in voting cases is to plot the percentage vote received by a black candidate in each precinct against the percentage of black residents in that precinct. If the resulting graph is approximately linear, there is a relationship between the race of the voter and the likelihood that he would vote for a black candidate. A line of x = y indicates complete racial polarization: no whites voting for the black candidate but all blacks voting for the black candidate. In the instant case, the testimony indicated that instead of plotting votes received by the black candidate against the percentage of black residents, Dr. Buss plotted the votes received against the concentration of "non-whites” in a precinct. We find that this error most likely results in an understatement of the degree of polarization.
Additionally, we note that Dr. Buss used only data from the City of Youngstown. While city data is useful because the plaintiff class resides almost exclusively within the city limits and because the district proposed by plaintiffs is nearly coextensive with the city limits, it would have been appropriate to include all the data from the state representative elections and from Pete Starks’ race for County Auditor.
Finally, it would have been useful to have statistical proof of black voting patterns in the absence of a black candidate.
. The correlation coefficient was usually between 0.75 and 1.0. Significantly, the lowest correlation coefficient, about 0.2, appeared in the race of Benson for state representative. In this race, Benson only received 12% of the black vote. He received 6% of the white vote.
. The slope of the graphs was consistently positive and approached one.
. These figures do not include the race of Benson because in that election the regression analysis indicated low correlation between the race of the voter and the percentage of votes the candidate received.
The following table shows plaintiffs’ extrapolations from Dr. Buss’s regression analyses:
Candidate Black Vote White Vote R2
Frost, 1981 mayoral primary 70% 20% 0.87
Pincham, 1983 school board 65% 33% 0.92
Simon, 1983 school board 60% 8% 0.77
Rogers, 1983 school board 73% 12% 0.91
Frost, 1983 mayoral general 64% 3% 0.96
Benson, 1984 State Rep. 12% 7% 0.20
Jackson, 1984 Pres, primary 98% 2% 0.96
Simon, 1985 school board 90% 6% 0.97
Starks, 1985 Mayoral Primary 98% 6% 0.97
Starks, 1986 County Auditor 67% 2% 0.67
Armour, 1986 State Rep. 48% 4% 0.83
Hightower, 1987 school board 90% 20% 0.93
Pincham, 1987 school board 80% 35% 0.91
Jackson, 1988 Pres, primary 98% _9% 0.95
Average (Total/14) 72% 12% 0.80
Without elections for which R 2 <0.50 77% 12% 0.90
The result of Armour's 1986 campaign for State Representative is particularly significant because in that race, the black candidate was running in a general election against an endorsed Democratic candidate. Since black voters in Mahoning County usually identify with the Democratic candidate in a general election *1058at a rate of eighty to ninety per cent, for an independent candidate to obtain nearly fifty per cent of the black vote is a remarkable showing.
. Plaintiff Ezell Armour, who was active in many leading black community organizations including the A. Philip Randolph Institute, the NAACP, and the Ministerial Alliance, testified that the representative had not consulted with the black community regarding this legislation although he had "a lot of contact with other people.” Although the record shows that the legislation was not pursued after the black community protested, we agree with the late Mr. Armour that “this was very insensitive.”
. The plaintiffs, relying on Gingles, 478 U.S. at 46 n. 12, 106 S.Ct. at 2764 n. 12, suggest that § 2 of the Voting Rights Act may permit an action based on a minority group’s "ability to influence” an election in a single-member district which is independent of that group's ability to elect the candidate of its choice. In Gingles, the Supreme Court stated:
We have no occasion to consider whether § 2 permits, and if it does, what standards should pertain to, a claim brought by a minority group, that is not sufficiently large and compact to constitute a majority in a single-member district, alleging that the use of a multimember district impairs its ability to influence elections.
478 U.S. at 46 n. 12, 106 S.Ct. at 2764 n. 12 (emphasis supplied). The plaintiffs argue that the Supreme Court’s dicta here recognizes the possibility of such an action under § 2. We need not reach the question of whether such an action may be viable under the Voting Rights Act because we find that the plaintiffs have met their burden of demonstrating an ability to elect a candidate of their choice.
. The State of Ohio argues that the plaintiffs’ claim of an ability to elect based on an alleged correlation between the outcome of the Democratic primary election and victory in the general election is flawed. We agree. Plaintiffs contend that in a redrawn district they will be able to elect a black candidate in the Democratic primary by a plurality and, because the Democratic candidate always wins the general election, they will therefore be able to elect a black candidate to office. The problem with this reasoning is that plaintiffs have demonstrated persuasively that white voters in Mahoning County do not usually vote for black candidates. Thus, there is no assurance that a black candidate winning the primary couid muster sufficient white votes to be elected to office. However, our conclusion that blacks in a redrawn district could elect a candidate of their choice depends neither upon an historical correlation between the outcome of the primary and the general election nor upon speculation as to how a black can be elected, and therefore the inconsistency in plaintiffs’ theory does not affect our conclusion.
The Census Bureau reports include data for each of the larger municipalities within Mahoning County and for the area which the source describes as "Remainder,” using separate tables for data about blacks and whites. We have combined the tables, showing only the "Total” for Mahoning County for each race. Additionally, we have omitted some of the data categories provided by the census bureau.