MEMORANDUM OPINION
CHASANOW, District Judge.Presently pending and ready for resolution in this action brought pursuant to 42 U.S.C. § 1981 are the following motions: (1) Defendants’ motion to strike class action allegations; (2) Plaintiffs’ motion for class certification; (3) Defendants’ motions to exclude all testimony and evidence from Plaintiffs’ expert witnesses Venkatesh Shankar, Richard Feinberg and Charles Calhoun; (4) Defendants’ motion to waive the local rule limiting the length of memoranda; and (5) Defendants’ motion for summary judgment. The issues have been fully briefed, and no hearing is deemed necessary. Local Rule 105.6. For the reasons set forth in this opinion, the motion to strike the class allegations will be denied; the motion for class certification will be granted; the motion to waive the local rule limiting the length of memoranda will be granted; the motion to exclude the testimony of Calhoun will be granted in part and denied in part; the motion to exclude the testimony of Shankar and Feinberg will be granted on relevance grounds; and the motion for summary judgment will be granted.
I. Background
Plaintiffs are five African-American individuals, Avis E. Buchanan, Dr. Albert R. Conley, Ardelia Crawford, Carolyn Korne-gay-Belton and Yvette D. Tate, who seek to be certified as class representatives in this action. Defendant Consolidated Stores Corp. (“Consolidated Stores”) is a Delaware corporation which, from May 5, 1996 to December 7, 2000, controlled or owned over one *182thousand KB Toy Stores across the United States. Since December 7, 2000, Defendant KB Consolidated, Inc. (“KB Consolidated”), an Ohio corporation, has controlled or owned the corporations doing business as KB Toy Stores. Plaintiffs’ complaint alleges that Consolidated Stores and/or KB Consolidated discriminated against them and other African-American customers when several KB Toy Stores located in predominately African-American neighborhoods in Maryland refused to accept checks to pay for merchandise.1 The individual Plaintiffs recount similar stories.
On separate occasions, Plaintiffs Crawford, Kornegay-Belton and Tate attempted to purchase gifts by check at the KB Toy Store at Iverson Mall in Temple Hills, but were told that the store did not accept checks. In July 1999, Crawford asked to see the manager about the no-check policy, as she previously had paid for items by check at KB Toys in Waldorf, Maryland and Falls Church, Virginia. In response to her inquiry concerning the no-check policy, the manager, an African-American, allegedly told her “you know how we are; we write bad checks.” Paper no. 45, ¶ 31. Upset, Crawford left without buying the items.
In December 1997, Kornegay-Belton was similarly told by the cashier that the store did not accept checks. She bought the items she wanted anyway. In December 1999 while working for the Equal Rights Center (“ERC”), Kornegay-Belton learned that ERC was investigating KB’s no-eheck policy and told her supervisor that she had earlier been a victim of the policy.
In November 1998, Tate also was told by a cashier at the Iverson Mall store that she could not pay for her merchandise by check. Tate, who nevertheless completed her purchase, claims that she previously had paid by cheek at KB Toys in Bowie, Maryland and Arlington, Virginia and suspected that the policy was due to the store’s largely African-American clientele. A day after Tate’s incident at the Iverson Mall store, she purchased items by check at the KB Toys at Pentagon City in Virginia.
Dr. Conley attempted to purchase a video game for his sons in July 1999 at KB Toys in Prince George’s Plaza and was told by a cashier that the store did not accept checks. Concerned about the matter, he asked to see the manager, who explained to him that because the store had been receiving bad checks, his superiors decided that checks would no longer be accepted there. The manager also told him that stores in Land-over and Silver Spring did not accept checks either. Dr. Conley purchased the video game by credit card.
In November 1999, Buchanan attempted to make a purchase by check at the KB Toys in Forest Village Park Mall in Forestville, Maryland and was told by the cashier that that particular store did not accept checks. Buchanan used her credit card to buy the items she wanted. Suspecting that she had been the victim of unlawful discrimination, Buchanan called the ERC the next day and asked the center to investigate KB’s no-check policy. Following Buchanan’s call, ERC conducted tests by telephone and in person and claims it uncovered “a pattern and practice of discrimination against African Americans____” Id. at ¶ 46. The tests allegedly uncovered that KB Toy Stores located in areas with a predominately African-American population did not accept checks from any of its customers while stores located in predominantly white areas would accept checks.
It is undisputed that between 1992 and 2000, some KB Toy Stores did not accept personal checks as a form of payment.2 As of December 15, 1999, the date this suit was filed, cheeks were not accepted at the following KB Toy Stores in the Washington-Baltimore area: (1) Forest Village Park Mall in Forestville (Store 180); (2) Prince George’s Plaza in Hyattsville (Store 190); (3) Laurel Center in Laurel (Store 119); (4) Iverson Mall in Temple Hills (Store 7875); (5) Belt*183way Plaza in Greenbelt (Store 2453); (6) Mondawmin Mall in Baltimore (Store 7720); (7) Reisterstown Road Plaza in Baltimore (Store 7722); and (8) City Place in Silver Spring (Store 1199). See Paper no. 45, ¶ 47.
Prior to the placement of stores on a no-check list, KB’s overall “check return expense rate” exceeded the rate for other retail stores owned by the Melville Corporation, KB’s parent company at the time.3 See Paper no. 77, Ex. T, Deposition of Charlie Stengl (“Stengl Dep.”), Ex. 7. In 1991, KB Assistant Controller and Treasurer Charlie Stengl, together with Diane Strong, compiled the first proposal analyzing options to reduce losses on check sales, including a recommendation to stop accepting checks in 37 stores across the country where the cheek return expense rate was greater than ten percent. See Stengl Dep. at 102-103, Ex. 4. Instead of adopting the no-check recommendation, KB adopted other methods to reduce check expense which were somewhat successful, but did not bring the check return expense rate down to the level of other Melville companies. In 1992, Stengl and Strong compiled their second “Check Return Proposal” for a group of KB upper management employees, known as the “Executive Committee,” which, among other options, recommended that KB discontinue cheek acceptance at the 25 stores having a check return expense rate of over ten percent. See id., Ex. 5.4 The Executive Committee discontinued accepting checks at 19 of the 25 stores, eighteen of which had check return expense rates in excess of ten percent. See id., Ex. 3. Of these nineteen stores, two were in the Washington-Baltimore area (Mondawmin Mall and Reisters-town Road Plaza).
Between 1992 and 1996, Stengl and his staff at KB’s Pittsfield, Massachusetts headquarters annually provided the Executive Committee with a proposal to reduce KB’s check return expense rate, which contained recommendations for stores to stop accepting checks. See id., Exs. 5, 7, 8. These proposals primarily contained data on check return expense and check usage for KB stores and other Melville stores, but did not contain any reference to race or ethnicity. According to Stengl, no-check decisions during this period were not determined by any bright-line rule, but rather by the gap between the worst performing check return expense stores and the next level of stores. See id. at 60-62. In October 1995, KB implemented a pilot program to test the use of debit cards as a replacement for checks in stores with high check return expense rates. As part of the program, KB discontinued accepting cheeks at the Forest Village, Prince George’s Plaza, and Iverson Mall stores. See id. at 307-311. After 1996, KB headquarter’s decisions regarding no-cheek stores were made by Finance Director Timothy Lee on an ad hoc basis, rather than as part of a regular review. See id. at 352-59, 375, Ex. 20; Deposition of Timothy Lee (“Lee Dep.”) at 76-77, 218. In addition to looking at check return expense and check usage when evaluating whether a store should be put on the no-eheck list, Lee considered other factors such as input from a Regional Vice President or a store’s District Manager. See Lee Dep. at 55-57.
In June 1997, the store in Beltway Plaza in Greenbelt (Store 2453) opened as a no-check store, making it the only KB Toy Store in the Washington-Baltimore area to refuse checks upon opening. District Manager Tammie Studebaker and Store Manager David Luton have differing recollections of the rationale for opening it as a no-check store, but neither has any recollection of race being involved. Studebaker testified in her deposition that Luton, who is African-American, expressed to her his opinion that the store should not accept checks because the store was in close proximity to two other stores that already did not accept- checks, Prince George’s Plaza (190) and Capital Plaza (466). See Paper no. 77, Ex. U, Deposition of Tammie Studebaker (“Studebaker Dep.”) at 69-*18474. Studebaker recalls that Luton was concerned that the Beltway Plaza store would likely attract many of the same customers as these stores, especially since the Capital Plaza store was closing. Id. Luton does not recall approaching Studebaker with this concern, but rather recalls that, prior to the opening of the store, she asked him whether or not the debit card system from the Capital Plaza store should be brought over to the new Beltway Plaza store. See id., Ex. Q, Deposition of David Luton (“Luton Dep.”) at 26-29. Luton responded in the affirmative because he believed the system to be more convenient than checks for both customers and employees. See id.
The Silver Spring (City Place) store (1199) opened in June 1997 around the same time as Beltway Plaza and, at first, accepted cheeks. Sometime during 1998, the City Place store stopped accepting checks after experiencing a rise in check return expense. See Lee Dep. at 209-210. KB also discontinued accepting checks at its Laurel store (119) in 1998 after District Manager Brian Metcalf requested that Lee substitute debit cards for cheeks due to high check return expense. See Paper no. 77, Ex. S, Deposition of Brian Metcalf (“Metcalf Dep.”) at 13-15; Metcalf Dep., Ex. 2.
At the time this lawsuit was filed, KB refused to accept checks at eight out of its fifty-two stores in the Washington-Baltimore area. Plaintiffs allege that the eight no-check stores served market areas with African-American populations significantly higher than the national average and included the KB stores with the highest percentages of African-American residents in the area. See Paper no. 83, Ex. 7, Charles A. Calhoun, Demographic and Statistical Analysis of KB Toy Stores (June 28, 2001) (“Calhoun Rep.”) at 2 and Tables 1.1 and 1.2. However, at least three stores located in predominately African-American areas continued accepting checks (Stores 1100,1080, and 1119). See id. Additionally, 49 of KB’s 66 no-check stores nationwide were located in majority white markets. See id., Ex. 9, Charles A. Calhoun, Demographic and Statistical Analysis of KB Toy Stores, Rebuttal Analysis (Aug. 9, 2001) (“Calhoun Rebuttal”) at 2.
Plaintiffs filed this action alleging deprivation of their right, and that of the members of the class they seek to represent, to make and enforce contracts on the same terms as those enjoyed by white persons, in violation of 42 U.S.C. § 1981. Specifically, Plaintiffs contended in their complaint that Defendants deliberately instituted a “no checks” policy at stores in the Washington-Baltimore area where the customers are most likely to be African-Americans. See Paper no. 45, ¶ 50. They contended that Defendants instituted this policy with the intent to deny African-Americans the right to pay by check, based on their belief that African-Americans are more likely to default on payment or tender a bad check. See id. Plaintiffs seek class-wide injunctive and declaratory relief, as well as class-wide punitive damages and individual compensatory damages. The compensatory damages sought are meant to compensate individual Plaintiffs and the members of the class they seek to represent for the economic loss, humiliation, embarrassment, and mental and emotional distress they have allegedly suffered. Plaintiffs seek class action status, pursuant to Fed. R. Crv. P. 23(a) and (b)(2). On January 4, 2001, the court denied Defendants’ motion to dismiss the case with respect to the named individual Plaintiffs, finding that they had stated a claim under 42 U.S.C. § 1981. Defendants filed the pending motion for summary judgment on October 29, 2002.
II. Class-Related Motions
A. Defendants’ Motion to Strike Class Allegations
Defendants have moved to strike from the second amended complaint the allegations which Plaintiffs purport to assert as representatives of a class.5 The gravamen of this motion is that Plaintiffs’ delay in filing a *185motion for class certification violates Fed. R. Civ. P. 23(c)(1) which states that “[a]s soon as practicable after the commencement of an action brought as a class action, the court shall determine by order whether it is to be so maintained.” Plaintiffs filed this action in December 1999 and alleged class action claims for the first time in their first amended complaint, filed in March 2000. On September 30, 2002, after discovery had largely been completed, the parties filed a joint status report which indicated that Plaintiffs would file a motion for class certification within a month. The motion for class certification was filed on November 1, 2002, nearly three years after commencement of the suit.
In their motion, Defendants rely primarily on McCarthy v. Kleindienst, 741 F.2d 1406 (D.C.Cir.1984), for support for the proposition that a delay in seeking class certification, combined with resulting prejudice to the defendants, warrants denial of class certification. Defendants claim to have been prejudiced because they were forced to decide whether to file a summary judgment motion prior to class certification, thereby risking a ruling that the outcome of the litigation will not be binding on the plaintiff class. They assert that because of the delay in class certification, the notice and opt-out requirements of Rule 23(b) have not gone into effect, thereby allowing potential class members to wait to see how the litigation proceeds and decide accordingly whether to opt out. Defendants also argue that they have been prejudiced because the tardiness in the filing of the motion for class certification has delayed them from conducting discovery as to the damages that allegedly have been suffered by the unidentified members of the class. In addition, Defendants argue, citing East Texas Motor Freight System, Inc. v. Rodriguez, 431 U.S. 395, 404-05, 97 S.Ct. 1891, 52 L.Ed.2d 453 (1977), that Plaintiffs’ delay in filing for class certification demonstrates that they would not adequately represent and protect the interests of the class, as required by Rule 23(a)(4).
Plaintiffs contend that they were entitled to and needed full discovery on class certification issues prior to the filing of the motion, and that Defendants have not been prejudiced by the timing of the class certification motion. They note that Defendants explored class issues during discovery, including during depositions of named Plaintiffs and putative class members, and that an earlier class certification motion would not have changed the way Defendants conducted discovery. They also note, and Defendants do not dispute, that the motion for class certification contains no information about putative class members that had not already been provided to Defendants in the early stages of discovery.6 Additionally, Plaintiffs note that unlike the jurisdiction in McCarthy, the main case cited by Defendants, this court does not have a local rule requiring that motions for class certification be filed within a specific time period.
Defendants have not shown that they have been prejudiced by the timing of the filing of the class certification motion. They have had notice of the class allegations since March 2000, and they submitted to this court a joint status report in September 2002 with Plaintiffs which stated that Plaintiffs would file a motion for class certification within a month. At no point during the course of the litigation did Defendants raise any concerns about the timing of the class certification motion. Defendants were made aware of all known potential class members in March and June of 2001, when Plaintiffs disclosed them in response to Defendants’ document requests. Defendants conducted discovery of class related issues, including taking depositions of many potential class members, and have provided no evidence that an earlier class certification motion would have changed the way they conducted discovery. Moreover, Plaintiffs were entitled to pre-certification discovery to establish the record the court needs to determine whether the requirements for a class action suit have been met. See, e.g., Miller v. Baltimore Gas & *186Elec. Co., 202 F.R.D. 195, 201-02 (D.Md.2001).
In addition, the argument that potential class members will be able to await the outcome of the litigation and then make a decision about whether to opt-out is inapplicable here because Plaintiffs are seeking certification under Rule 23(b)(2), which, unlike 23(b)(3), does not require notice to class members and the opportunity to opt-out.7 See Fed. R. Civ. P. 23(c). As discussed further below, Plaintiffs are not seeking to certify a damages class under Rule 23(b)(3), but rather to pursue compensatory damages in a non-class second litigation phase. Thus, Defendants’ concern about potential class members making calculated decisions to opt-out of the suit based on the outcome of the summary judgment motion is moot.
Lastly, the argument that the timing of the motion renders Plaintiffs inadequate class representatives is without merit, particularly since, unlike in the McCarthy case, there is no local rule in this district requiring a class certification motion to be filed within a specific period of time. This case is also easily distinguished from Rodriguez because the plaintiffs in that case failed entirely to file a motion for class certification prior to trial. See Rodriguez, 431 U.S. at 404-05, 97 S.Ct. 1891. In any event, Defendants have failed to identify any harm suffered by class members due to the timing of the motion which would warrant the conclusion that the named Plaintiffs are inadequate class representatives.
For the foregoing reasons, the court does not agree with Defendants that Rule 23(c)(1) requires that the class allegations be stricken, and Defendants’ motion will be denied.
B. Motion for Class Certification
Plaintiffs have moved for an order certifying them as representatives of a class of African-Americans who were prevented from paying by check because of a no-check policy at certain KB Toy Stores in the Washington-Baltimore metropolitan area between 1992 and 2000. They seek certification for class-wide liability, injunctive and declaratory relief and punitive damages under Rule 23(b)(2), but request that the court sever the determination of compensatory damages for individual class members into a separate phase of the litigation.8
To maintain a class action under Fed. R. Civ. P. 23, Plaintiffs must satisfy the four prerequisites of Rule 23(a), as well as at least one of the three requirements of Rule 23(b). The four requirements of Rule 23(a) are commonly referred to as (1) numerosity, (2) commonality, (3) typicality, and (4) adequacy of representation. The burden of establishing class status is on the Plaintiffs. See Hewlett v. Premier Salons Int’l, Inc., 185 F.R.D. 211, 215 (D.Md.1997), citing International Woodworkers of America, AFL-CIO, CLC v. Chesapeake Bay Plywood Corp., 659 F.2d 1259, 1267 (4th Cir.1981).
Defendants argue that the motion for class certification should be denied because Plaintiffs cannot fulfill the commonality and typicality requirements for two primary reasons.9 First, Defendants claim that the *187evolution of the decision-making process regarding check acceptance between 1992 and 2000 essentially vitiates Plaintiffs’ ability to fulfill the commonality and typicality requirements. They assert that the no-eheck decisions concerning the stores for which Plaintiffs seek class certification occurred during three distinct time frames, with different decision-makers involved in each time frame. Defendants argue that if the court certifies the ease for class action treatment, the class should be limited to only those African-American customers who, like the named Plaintiffs, were unable to pay by check at stores 180,190, and 7875.10
Second, Defendants argue that in contrast to other class action cases, like Hewlett, where there was a uniform, centralized policy at issue, no such governing policy exists in this ease. Defendants characterize this case as one involving a series of individual decisions over the course of several years concerning whether to accept checks at particular stores. Defendants note that only 67 out of over 1200 stores were ever placed on the no-check list, and that checks were accepted in 119 of the 141 KB stores in areas where African-Americans constituted the majority of the population. Defendants conclude that because there was no overarching policy that affected all class members, Plaintiffs cannot fulfill the commonality and typicality requirements.
Rule 23(a)(2) requires that class members share common legal or factual issues, and Rule 23(a)(3) requires that the claims or defenses of the representative parties be “typical” of the claims or defenses of the class. The commonality requirement is relatively easy to satisfy, and very few cases have been dismissed for failing to meet it. See Hewlett, 185 F.R.D. at 216. This court recently noted that
the commonality “inquiry is not whether common questions of law or fact predominate, but only whether such questions exist.” Hewlett, 185 F.R.D. at 216 (citations omitted). Commonality does not require class members to share all issues in the suit, but simply a single common issue. Central Wesleyan College v. W.R. Grace & Co., 143 F.R.D. 628, 636 (D.S.C.1992), aff'd, 6 F.3d 177 (4th Cir.1993); Holsey[ v. Armour & Co.], 743 F.2d [199]at 216-17 [(4th Cir.1984)]. Thus, factual differences among the class members’ cases will not preclude certification if the class members share the same legal theory. Peoples v. Wendover Funding, Inc., 179 F.R.D. 492, 498 (D.Md.1998) (citing Brown v. Eckerd Drugs, Inc., 663 F.2d 1268, 1275 (4th Cir.1981), vacated on other grounds, 457 U.S. 1128, 102 S.Ct. 2952, 73 L.Ed.2d 1345 (1982)); Hewlett, 185 F.R.D. at 216.
Bullock v. Board of Ed. of Montgomery County, 210 F.R.D. 556, 560 (D.Md.2002). In the discrimination context, however, it has been said that ‘“the Supreme Court has pulled in the reins on across-the-board attacks. Conclusory allegations of discrimination on a class basis are not enough.’ ” Hewlett, 185 F.R.D. at 216, quoting Zapata v. IBP, Inc., 167 F.R.D. 147, 158 (D.Kan.1996). In Hewlett, the court concluded that “the key to the commonality requirement in discrimination cases seems to be whether there is evidence of centralized decisionmaking.” Id. (citations omitted).
Addressing the typicality requirement in Bullock, the court stated that
[t]he typicality requirement, which has been observed to be a redundant criterion, determines “whether a sufficient relationship exists between the injury to the named plaintiff and the conduct affecting the class, so that the court may properly attribute a collective nature to the challenged conduct.” Hewlett, 185 F.R.D. at 217 (citing Zapata v. IBP, Inc., 167 F.R.D. 147, 160 (D.Kan.1996)). This court has previously held that a plaintiffs claim may differ factually and still be “typical” of the claims of class members if “ ‘it arises from *188the same event or practice or course of conduct that gives rise to the claims of other class members, and if his or her claims are based on the same legal theory.’ ” Id. (citations omitted).
Bullock, 210 F.R.D. at 560.
In the instant case, there are clearly questions of law or fact common to the named Plaintiffs and proposed class members, most importantly whether racial stereotyping affected Defendants’ decision making concerning which stores should not accept checks. The court finds unpersuasive Defendants’ argument that commonality is lacking because there were several people involved in the decision making process at different points in time. The fact that the individual employees involved in the decisions may have changed over the years does not negate the existence of common questions regarding the motives of the Defendant corporations and does not establish a lack of a centralized policy.11 The individuals who made the no-check decisions did so from Defendants’ central headquarters in Pittsfield, Massachusetts, on behalf of their employer. The evidence of centralized decisionmaking is, therefore, sufficient for the court to conclude that Plaintiffs have fulfilled the commonality requirement.
The typicality element is similarly fulfilled. The claims of the named Plaintiffs in this case arise from the same alleged practice that gives rise to the claims of other proposed class members. Like the named Plaintiffs, the class members are African-Americans who were allegedly prevented from paying by check because of a no-check policy at a KB Toy Store in the Washington-Baltimore metropolitan area, which Plaintiffs allege was based on discrimination. Additionally, the claims of the named Plaintiffs are based on the same legal theory as those of the proposed class members, namely that Defendants violated their rights under 42 U.S.C. § 1981. The court finds unpersuasive Defendants’ suggestion that the named Plaintiffs may only be deemed “typical” of class members who attempted to pay by check at the same three stores as the named Plaintiffs, and not the other five no-cheek stores in the area. The typicality requirement is met if the claims of the named plaintiffs “arise[] from the same event or practice or course of conduct that gives rise to the claims of other class members____” Hewlett, 185 F.R.D. at 217 (citation omitted). Because the injury to class members at the five stores not visited by the named Plaintiffs arises from the same alleged conduct as the injury caused to the named Plaintiffs, as well as from the same legal theory, the typicality requirement is fulfilled as to all eight stores.
Plaintiffs must also fulfill one of the requirements of Rule 28(b) in order to be certified as a class. Certification under Rule 23(b)(2), which Plaintiffs here purport to have fulfilled, is appropriate where “the party opposing the class has acted or refused to act on grounds generally applicable to the class, thereby making appropriate final injunctive relief or corresponding declaratory relief with respect to the class as a whole.” The fact that Plaintiffs seek compensatory and punitive damages “does not preclude certification under this subsection; however, Rule 23(b)(2) certification is appropriate only where the relief sought is primarily injunctive or declaratory.” Hewlett, 185 F.R.D. at 218-19 (citations omitted). In other words, certification under 23(b)(2) turns on whether the claims for injunctive and declaratory relief “predominate” over the claims for compensatory and punitive damages. Id. at 219.
In Hewlett, the court bifurcated the suit so that issues of liability and class-wide relief would be tried first, with individual compensatory relief to be determined later. The court then ruled that, with the issues separated, the request for declaratory and injunctive relief predominated over the request for punitive damages, making certification under 23(b)(2) proper. See id., citing Kernan, 1990 WL 289505 at *5-6.
Plaintiffs here have proposed a similar bifurcation of the case in order to ensure that their request for damages will not defeat Rule 23(b)(2)’s requirement that declaratory *189and injunctive relief predominate.12 Under Plaintiffs’ proposal, the first stage would focus on class-wide liability, injunctive and declaratory relief, and punitive damages. The determination of individual compensatory damages, if any, would be conducted in a second, non-class phase, where there would be a presumption that Defendants had discriminated against each class member. Defendants would bear the burden of showing that a class member did not fall within the class or was not entitled to damages. Defendants have not objected to the proposed bifurcation.13
As in Hewlett and other public accommodations cases, the court agrees that bifurcation of the class-wide and individual claims would promote efficiency and would be the “soundest approach” to take. See id., citing Kemcm, 1990 WL 289505 at *5-6. With the issues thus separated, the court must consider whether injunctive and declaratory relief sought predominate over the monetary relief sought in this first stage. At least one of the named Plaintiffs has indicated that the main goal of this litigation is a public acknowledgment of discrimination from Defendants. See Paper no. 78, Ex. V, Deposition of Carolyn Kornegay-Belton at 149-151. It is clear that injunctive and declaratory relief lie at the heart of this ease and predominate over any damages sought.14 Because injunctive and declaratory relief predominate in this case, class certification is appropriate under Rule 23(b)(2). Accordingly, the court will grant Plaintiffs’ motion for class certification pursuant to Rules 23(a) and 23(b)(2), with the individual compensatory damages claims severed.15
III. Defendants’ Motion for Summary Judgment and Motions to Exclude Testimony and Evidence from Plaintiffs’ Expert Witnesses16
A. Standard of Review
It is well established that a motion for summary judgment will be granted only if there exists no genuine issue as to any material fact and the moving party is entitled to judgment as a matter of law. Fed. R. Civ. P. 56(c); Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 250, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986); Celotex Corp. v. Catrett, 477 U.S. 317, 322, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986). In other words, if there clearly exist factual issues “that properly can be resolved only by a finder of fact because they may reasonably *190be resolved in favor of either party,” then summary judgment is inappropriate. Anderson, 477 U.S. at 250, 106 S.Ct. 2505; see also Pulliam Inv. Co. v. Cameo Properties, 810 F.2d 1282, 1286 (4th Cir.1987); Morrison v. Nissan Motor Co., 601 F.2d 139, 141 (4th Cir.1979); Stevens v. Howard D. Johnson Co., 181 F.2d 390, 394 (4th Cir.1950). The moving party bears the burden of showing that there is no genuine issue as to any material fact. Fed. R. Civ. P. 56(c); Pulliam Inv. Co., 810 F.2d at 1282, citing Charbonnages de France v. Smith, 597 F.2d 406, 414 (4th Cir.1979).
When ruling on a motion for summary judgment, the court must construe the facts alleged in the light most favorable to the party opposing the motion. United States v. Diebold, Inc., 369 U.S. 654, 655, 82 S.Ct. 993, 8 L.Ed.2d 176 (1962); Gill v. Rollins Protective Servs. Co., 773 F.2d 592, 595 (4th Cir.1985). A party who bears the burden of proof on a particular claim must factually support each element of his or her claim. “[A] complete failure of proof concerning an essential element ... necessarily renders all other facts immaterial.” Celotex Corp., 477 U.S. at 323, 106 S.Ct. 2548. Thus, on those issues on which the nonmoving party will have the burden of proof, it is his or her responsibility to confront the motion for summary judgment with an affidavit or other similar evidence. Anderson, 477 U.S. at 256, 106 S.Ct. 2505.
In Celotex Corp., the Supreme Court stated:
In cases like the instant one, where the nonmoving party will bear the burden of proof at trial on a dispositive issue, a summary judgment motion may properly be made in reliance solely on the “pleadings, depositions, answers to interrogatories, and admissions on file.” Such a motion, whether or not accompanied by affidavits, will be “made and supported as provided in this rule,” and Rule 56(e) therefore requires the nonmoving party to go beyond the pleadings and by her own affidavits, or by the “depositions, answers to interrogatories, and admissions on file,” designate “specific facts showing that there is a genuine issue for trial.”
Celotex Corp., 477 U.S. at 324, 106 S.Ct. 2548. However, “ ‘a mere scintilla of evidence is not enough to create a fact issue.’ ” Barwick v. Celotex Corp., 736 F.2d 946, 958-59 (4th Cir.1984), quoting Seago v. North Carolina Theatres, Inc., 42 F.R.D. 627, 632 (E.D.N.C.1966), aff'd, 388 F.2d 987 (4th Cir.1967). There must be “sufficient evidence favoring the nonmoving party for a jury to return a verdict for that party. If the evidence is merely colorable, or is not significantly probative, summary judgment may be granted.” Anderson, 477 U.S. at 249-50, 106 S.Ct. 2505 (citations omitted).
B. Analysis
Plaintiffs’ claims are based entirely on 42 U.S.C. § 1981. Section 1981, in pertinent part, states:
All persons within the jurisdiction of the United States shall have the same right in every State and Territory to make and enforce contracts, ... as is enjoyed by white citizens____
(b) ... the term “make and enforce contracts” includes the making, performance, modification, and termination of contracts, and the enjoyment of all benefits, privileges, terms, and conditions of the contractual relationship.
To prevail on a § 1981 action like the one presently before the court, a plaintiff must prove that: (1) he or she is a member of a racial minority; (2) the defendant intended to discriminate on the basis of race; and (3) the discrimination concerned one or more of the activities protected by the statute. See Harris v. Allstate Insurance Co., 83 F.Supp.2d 423, 430 (S.D.N.Y.2000); Hill v. Shell Oil Co., 78 F.Supp.2d 764, 776 (N.D.Ill.1999), citing Morris v. Office Max, Inc., 89 F.3d 411, 413 (7th Cir.1996); Bobbitt v. Rage Inc., 19 F.Supp.2d 512, 517 (W.D.N.C.1998). The critical element of a § 1981 claim is the showing of intentional discrimination, not merely that the defendant adopted a policy or practice that had a disparate impact upon minorities. See General Building Contractors Ass’n, Inc. v. Pennsylvania, 458 U.S. 375, 102 S.Ct. 3141, 73 L.Ed.2d 835 (1982); Powell v. Super 8 Motels, Inc., 181 F.Supp.2d 561, 564 (E.D.N.C. *1912000). A plaintiff may establish a prima facie case of discrimination through either direct evidence of intentional discrimination or circumstantial evidence which creates a reasonable inference of discrimination. If a plaintiff establishes a prima facie case of discrimination through circumstantial evidence, the burden of production then shifts to the defendant to provide a legitimate, nondiscriminatory reason for its behavior. See Reeves v. Sanderson Plumbing Products, Inc., 530 U.S. 133, 142, 120 S.Ct. 2097, 147 L.Ed.2d 105 (2000); Texas Dept. of Community Affairs v. Burdine, 450 U.S. 248, 252-53, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981); McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973). The plaintiff must then “ ‘prove by a preponderance of the evidence that the legitimate reasons offered by the defendant were not its true reasons, but were a pretext for discrimination.’ ” Reeves, 530 U.S. at 142, 120 S.Ct. 2097, citing Burdine, 450 U.S. at 253, 101 S.Ct. 1089.17
1. Prima Facie Case of Intentional Discrimination
In the instant case, Defendants contend in their motion for summary judgment that Plaintiffs have offered neither direct nor circumstantial evidence that Defendants’ no-eheek decisions were motivated by a desire to prevent African-Americans from paying by personal check. With respect to direct evidence, Plaintiffs respond that even though they need not present direct evidence to survive a summary judgment motion or prevail at trial, they have produced anecdotal direct evidence that KB’s own employees, even though not themselves responsible for implementing the no-cheek policy, knew that the policy was based on race and on stereotypes about the creditworthiness of African-Americans. Plaintiffs cite the deposition testimony of Plaintiff Ardelia Crawford that when she asked the African-American manager of the Iverson Mall store why the store would not accept her check, he replied, “You know how we are; we write bad checks.” See Paper no. 83, Ex. G, Deposition of Ardelia Crawford at 87-88. Additionally, declarant Latrell Duncan testified that an unidentified KB employee at the Greenbelt store, most likely a sales associate, told her that KB does not accept cheeks in “black areas.” See id., Ex. J, Deposition of Latrell Duncan at 35. The comments of two unidentified, low-level field personnel who were not involved in any way in check acceptance decisions cannot be viewed as direct evidence of discrimination. Numerous KB executives who were actually involved in selecting the no-eheck stores testified that financial considerations, not race, determined which stores would stop accepting cheeks. See Stengl Dep. at 88, 145, 289-90, 306-311; Lewis Dep. at 35; Lee Dep. at 171-72, 209-210; Watanabe Dep. at 28-31, 99-105; Johnson Dep. at 19-22,'32-34, 42; Winn Dep. at 36, 39-41. Moreover, the comments by unidentified KB employees about KB’s purported motives for not accepting checks in certain stores are inadmissible hearsay. See, e.g., Gbenoba v. Montgomery Co. Dept. of Health and Human Services, 209 F.Supp.2d 572, 579 (D.Md.2002) (employees’ hearsay statements that white employees were paid more than minority employees were not admissible under hearsay exception for admission by party opponent). Only admissible evidence can be considered in opposition to summary judgment. See id., citing Fed. R. Civ. P. 56(e). Plaintiffs have not forecast direct evidence of intentional discrimination sufficient to survive a motion for summary judgment.
Circumstantial evidence of alleged discrimination in a case brought pursuant to § 1981 usually consists of a showing that members of a protected class were deprived, at a particular location, of services that were provided to similarly situated persons outside the class. See, e.g., Murrell v. Ocean Mecca Motel, Inc., 262 F.3d 253, 257 (4th Cir.2001); Callwood v. Dave & Buster’s, Inc., 98 F.Supp.2d 694, 705-07 (D.Md.2000); Hill, 78 *192F.Supp.2d at 777 (requirement to pre-pay for gasoline applied only to African-Americans). However, in this case, the undisputed evidence shows that KB’s no-eheck stores refused to accept payment by check from all customers regardless of race, not just African-Americans. Plaintiffs have not alleged and do not attempt to prove otherwise. Rather, Plaintiffs allege and must provide evidence that gives rise to the inference that Defendants attempted to prevent African-Americans from being able to pay by check by purposefully designating stores in predominately African-American areas as no-check stores.18
Plaintiffs assert that they have amassed several types of circumstantial evidence that, when viewed together, support their allegations that KB relied upon the race of a store’s clientele in deciding whether to add it to the no-check list. The burden of presenting a circumstantial prima facie case of intent is “not onerous.” Burdine, 450 U.S. at 253, 101 S.Ct. 1089. The function of this initial step was described by the Court as follows:
The prima facie case serves an important function in the litigation: it eliminates the most common nondiseriminatory reasons for the plaintiffs rejection. See Teamsters v. United States, 431 U.S. 324, 358, and n. 44, 97 S.Ct. 1843, 1866, n. 44, 52 L.Ed.2d 396 (1977). As the Court explained in Furnco Construction Corp. v. Waters, 438 U.S. 567, 577, 98 S.Ct. 2943, 2949, 57 L.Ed.2d 957 (1978), the prima facie case “raises an inference of discrimination only because we presume these acts, if otherwise unexplained, are more likely than not based on the consideration of impermissible factors.” Establishment of the prima facie case in effect creates a presumption that the employer unlawfully discriminated against the employee. If the trier of fact believes the plaintiffs evidence, and if the employer is silent in the face of the presumption, the court must enter judgment for the plaintiff because no issue of fact remains in the case.
Id., 450 U.S. at 253-54,101 S.Ct. 1089. It is, as recognized by Judge Davis, “essentially a ‘channeling device’ which ‘is not a difficult requirement to satisfy.’ ” Callwood, 98 F.Supp.2d at 705-06, citing Gibson v. Old Town Trolley Tours of Washington, D.C., Inc., 160 F.3d 177, 181 (4th Cir.1998). He concluded, in a widely cited decision, that a prima facie case in a restaurant setting consists of proof that (1) plaintiff is a member of a protected class, (2) plaintiff made himself or herself available to receive and pay for services ordinarily provided by the defendant to all members of the public in the manner in which they are ordinarily provided, and (3) plaintiff did not enjoy the privileges and benefits of the contracted for experience under factual circumstances which rationally support an inference of unlawful discrimination in that (a) plaintiff was deprived of services while similarly-situated persons outside the protected class were not deprived of those services and/or (b) plaintiff received services in a markedly hostile manner and in a manner which a reasonable person would find objectively unreasonable. See id. at 707. As the court noted:
[Sjubpart 3(a) of the ... test invokes the traditional “similarly situated” analysis, and recognizes that the point of comparisons between plaintiffs within the protected class and similarly situated persons outside the class “is this: because the classes are similarly situated in most relevant respects except their protected status (e.g., gender or race), there arises a rational *193inference of discrimination on the basis of that status.” Myers v. Hose, 50 F.3d 278, 284 (4th Cir.1995).
Id. The court in O’Neill v. Gourmet Systems of Minnesota, Inc., 213 F.Supp.2d 1012, 1019 (W.D.Wis.2002), similarly remarked, quoting Smerkiewicz, v. Sorema N.A., 534 U.S. 506, 122 S.Ct. 992, 152 L.Ed.2d 1 (2002) and Furnco Constr. Corp. v. Waters, 438 U.S. 567, 577, 98 S.Ct. 2943, 57 L.Ed.2d 957 (1978), that “[t]he precise requirements of a prima facie case can vary depending on the context and were ‘never intended to be rigid, mechanized, or ritualistic.’ ” The O’Neill decision involved a corporate policy to accept only state-issued drivers licenses, and not tribal identification cards, as proof of age for purchasing alcohol. The court, while noting that 3(b) was not directly implicated, observed:
Because plaintiff in this case is not complaining of hostile treatment from a particular restaurant server or manager, subpart 3(b)’s “markedly hostile manner” formulation may appear inapt for purposes of challenging a uniform corporate policy. However, this concern is unwarranted because “[fjactors relevant to subpart 3(b)’s ‘markedly hostile’ component include whether the conduct ‘is (1) so profoundly contrary to the manifest financial interests of the merchant and/or her employees’; (2) so far outside of widely-accepted business norms; and (3) so arbitrary on its face, that the conduct supports a rational inference of discrimination.” [Christian v. Wal-Mart Stores, Inc., 252 F.3d 862, 871 (6th Cir.2001).] (quoting Callwood, 98 F.Supp.2d at 708). These factors are as well-suited to analyzing a restaurant’s adoption of a uniform alcohol policy as they are to examining individual interactions between retail staff and their customers.
Id. at 1020.
In this case, then, a prima facie case will consist of proof that (1) plaintiffs are members of a protected class; (2) they sought to pay for an item in a manner ordinarily accepted by defendant; and (3) they were denied the opportunity to make the purchase in that manner under circumstances which rationally support an inference of unlawful discrimination in that either (a) those customers frequenting stores in African-American communities were prevented from paying by check while similarly-situated persons visiting stores in other communities were allowed to do so or (b) the no-check policy applicable to some, but not all stores, was objectively unreasonable.
Without question, Plaintiffs are members of a protected class. They have also established that paying by check was an ordinarily accepted method of paying for items at Defendants’ stores. The question, then, is whether the -evidence presented is sufficient to permit an inference of unlawful discrimination. In light of the minimal standard for presenting a prima facie ease of discrimination, it is not necessary at this stage to look beyond the statistical information in Tables 1.1 and 1.2 attached to the report of Dr. Calhoun, Plaintiffs expert witness. These tables show that when the 52 stores in the Washington Metropolitan region are ranked according to the percentage of African-Americans in the surrounding market area, all of the no-eheck stores were in the top-third. See Calhoun Rep., Tables 1.1, 1.2. Additionally, one half (4) of the 8 no-eheck stores in the area were located in the market areas with the highest concentration of African-Americans, well over 50% of each market area. See id. Looking solely at that information, without an explanation from Defendants, it is permissible to infer a racially discriminatory motive. Thus, the court finds that Plaintiffs have met the burden of establishing a prima facie case.
2. Legitimate, Non-discriminatory Reason
Defendants, however, have met their burden of producing a legitimate, non-discriminatory reason for the no-cheek policy, namely that their no-eheck decisions were made in order to reduce check losses among KB Toy Stores. As described above in the Background section, Defendants contend that their check acceptance policy was part of an ongoing effort to reduce KB’s overall check return expense rate.
*1943. Pretext
In order to prevail, Plaintiffs are required to forecast evidence demonstrating that the legitimate reason offered by Defendants was not the true reason for the no-check decisions, but was rather a pretext for discrimination. See Reeves, 530 U.S. at 143, 120 S.Ct. 2097. Plaintiffs contend that there is substantial evidence that Defendants’ purported legitimate business reason is a pretext for intentional discrimination.19 Plaintiffs now clearly perceive this case as a mixed motive situation where check expense rates were considered, but that “among KB Toys stores with comparable check return and usage rates, race was often the deciding factor in determining whether the store would accept checks.” Paper no. 83 at 1. As such, Plaintiffs still must prove that race was a motivating factor in that it made a difference in KB’s no-cheek decisions. See Desert Palace, Inc. v. Costa, — U.S. -, 123 S.Ct. 2148, 2152, 156 L.Ed.2d 84 (2003).
a) Statistical Evidence
Plaintiffs’ first claim is that statistical data demonstrate that the no-check decisions cannot be explained by KB’s check expense rationale. The Supreme Court has held that “‘(s)tatistical analyses have served and will continue to serve an important role’ in eases in which the existence of discrimination is a disputed issue.” International Broth. of Teamsters v. United States, 431 U.S. 324, 339, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977) (internal citations omitted). However, the Court went on to note that “statistics are not irrefutable; they come in infinite variety and, like any other kind of evidence, they may be rebutted.” Id. In the instant case, Plaintiffs rely chiefly on statistical analyses by their expert witness, Dr. Calhoun, in their effort to offer circumstantial evidence that Defendants used race as a factor in designating no-check stores and to overcome the stated non-discriminatory rationale.
i. Location of No-Check Stores
Citing Calhoun’s report, Plaintiffs assert that no-check stores were located in African-American communities at a very disproportionate rate, including in the Washington, D.C. area. That report notes that all eight no-check stores in the area served communities with “significantly higher than the national average” population of African-American customers and that 75% of KB’s no-check stores nationwide were located in areas with higher than average African-American populations. See Calhoun Rep. at 2, 4-11.20 Plaintiffs conclude that Calhoun’s analysis demonstrates that, in many instances, placement on KB’s no-check list was more closely related to the racial composition of the market area than to check history.
As noted in footnote 20, this approach is misleading and illogical because it proceeds from the premise that “higher than the national average” denotes majority African-American. It does not. While in the Washington, D.C. Metropolitan area, many stores do, in fact, serve majority African-American populations, that is not as true nationwide. Of the 66 no-eheck stores nationwide, only 12 were located in majority African-American communities. See Calhoun Rep., Table 2.1. Even Calhoun recognizes this in his rebuttal report at p. 6:
Majority black market areas are simply not a likely outcome for a national concern operating in an overwhelming majority white population. What is relevant is that KB discontinued accepting checks at 12 of the 17 stores located in majority black areas nationwide.
Thus, while the simple analysis of demographics may supply a prima facie case, it does not overcome the rationale stated by Defendants as the basis for their decisions.
Even if the analysis by Plaintiffs’ expert shows that there was a correlation between the racial composition of store mar-*195két areas and cheek acceptance at stores in those areas, it does not establish that race caused Defendants to stop accepting checks at any particular store. As noted by Defendants’ expert, Dr. Jonathan L. Walker, the correlation would only indicate race as a factor if race was not correlated with the legitimate business factors upon which KB claims to have based its decisions. However, Walker’s analysis found, and Plaintiffs did not provide evidence to dispute, that there was indeed a correlation between race and check return expense. See Paper no. 38, Ex. 14 (Walker Decl.) at 9-10, ¶ 16.21 In other words, his research determined that stores in market areas with greater proportions of African-Americans tended to have higher check return expense rates and lower check usage rates. See Walker Decl. at 9-10, ¶ 16. Thus, given that check performance and racial composition of the market area are correlated, it logically follows that KB, in an effort to control losses on checks, would discontinue accepting checks in areas with higher than average African-American populations. Therefore, the findings of correlation between the location of no-check stores and the racial composition of the surrounding market areas do not show that discrimination was the real reason for the no-check decisions, but rather reinforce KB’s primary stated reason for placing stores on the no-eheck list.22
Furthermore, it is undisputed that while KB stopped accepting cheeks at eight stores in areas with higher than average African-American populations, it continued to accept checks at some stores located in African-American areas, two of which (Stores 1100 and 1080) had low cheek return expense rates and one of which had a high cheek return expense rate (Store 1119). See Paper no. 83, Ex. 12. Plaintiffs try to explain this away in their opposition by shifting their argument to assert that the race of stores’ customers only played a role in no-check decisions for “mid-level” stores. In other words, Plaintiffs now contend that some stores had such high cheek return expense rates that KB stopped accepting checks regardless of the race of the store’s customers, and that other had such low check return expense rates that KB continued accepting checks regardless of race. They argue that it is only in a third group of stores, where the check return expense rates fell in a middle range, that KB considered the race of stores’ customers in deciding whether to accept checks.23
This claim, however, retreats from the categorical position taken in Plaintiffs’ complaint and indicates an attempt by Plaintiffs to make their theory of discrimination fit the statistical findings. Moreover, this theory does not explain why Defendants continued to accept checks at Store 1119, a store with undisputedly high check return expense, which Plaintiffs describe as serving an African-American community. The analysis of the location of no-check stores conducted by Plaintiffs’ expert witness does not support Plaintiffs’ claim that Defendants relied on race intentionally to discriminate against African-Americans to the effect that they refused to accept checks at stores in areas with higher than average African-American populations.
ii. “Exceptions” Stores
Plaintiffs argue, again relying on Dr. Calhoun’s report, that while stores serving higher percentage African-American communities were over-represented on the no-check list, they were under-represented in stores Calhoun dubbed “exception” stores. Dr. Calhoun performed the “exceptions” analysis by using the so-called 10/5 Rule to identify 62 KB Toy Stores which he claims would have been designated as no-eheek stores if KB had been making decisions solely by select*196ing stores with a check return expense rate of over 10% and a check usage rate of less than 5% of total sales. See Calhoun Rep. at 2-4, 11-14. Dr. Calhoun compiled the list of stores that were exceptions to the 10/5 Rule by applying the 10/5 Rule to monthly bad check data for eleven non-consecutive months between June 1996 and August 1999. See id. He found that for 4 of the 11 months of bounced check data provided by Defendants, the median check return expense rate for the 62 “exception” stores was higher than the check return expense rates for stores on the no-check list. Id. at 13. He concluded from this that placement of certain stores on the no-check list could, therefore, not be explained by those stores’ check histories. Instead, Calhoun’s demographic analyses showed that exception stores were closely correlated to race, with 77% of the exception stores located in market areas with white populations higher than the national average. Id. at 2-4,11.
Separate from their motion for summary judgment, Defendants have filed a motion to exclude Dr. Calhoun’s testimony and evidence pursuant to Fed. R. Evid. 702 because it is factually and methodologically flawed, and therefore, irrelevant. Defendants’ primary contention is that the assumption underlying Calhoun’s entire analysis is incorrect because they never used a 10/5 Rule to make no-cheek decisions. They assert, citing the deposition of Charlie Stengl, that the 10% check return expense rate was merely a general threshold level in 1991 at which a store would be assessed potentially to become a no-check store. See Stengl. Dep. at 122-23. The 5% check usage mark, they claim, was just the average check usage rate for those stores that exceeded the 10% check return expense level in 1991. Id. Thus, Calhoun’s reliance on the non-existent 10/5 Rule to analyze no-check decisions between 1996 and 1999, Defendants contend, renders the conclusions about the “exception” stores meaningless.
Federal Rule of Evidence 702, which governs the circumstances under which a witness can offer expert testimony, provides:
If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert ... may testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case.
Under Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 589, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), the court must assess whether the proffered expert testimony is relevant and reliable. The Fourth Circuit recently explained the court’s rule under Rule 702 as follows:
The first prong of this inquiry necessitates an examination of whether the reasoning or methodology underlying the expert’s proffered opinion is reliable—that is, whether it is supported by adequate validation to render it trustworthy. See [Daubert], at 590 & n. 9, 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469. The second prong of the inquiry requires an analysis of whether the opinion is relevant to the facts at issue. See id. at 591-92, 509 U.S. 579, 113 S.Ct. 2786,125 L.Ed.2d 469. Thus, an expert’s testimony is admissible under Rule 702 if it “rests on a reliable foundation and is relevant.” Kumho Tire Co. v. Carmichael, 526 U.S. 137, 119 S.Ct. 1167, 1171, 143 L.Ed.2d 238 (1999) (internal quotation marks omitted).
Westberry v. Gislaved Gummi AB, 178 F.3d 257, 261 (4th Cir.1999).
The record shows that even if the 10/5 Rule was ever used as a bright-line test, it was not in use during the time period during which Dr. Calhoun conducted his exceptions analysis. Thus, Dr. Calhoun’s analysis is grounded in incorrect assumptions about the process used by KB to determine which stores would stop accepting checks. An expert opinion that analyzes exceptions to a rule that was never actually used during the relevant time frame, if at all, does not provide relevant or reliable information and undoubtedly cannot support an inference that the exceptions were the result of intentional discrimination. Plaintiffs argue that Dr. Calhoun’s study is still probative even if the 10/5 *197Rule was not in effect at the time period for which he conducted the analysis because it demonstrates that high check return expense rates and low check usage do not explain placement on the no-check list. This argument is unpersuasive because it assumes that Defendants made their no-check decisions after 1992 from some mechanical formula of high cheek return expense and low check usage, but Plaintiffs have not provided evidence that decisions were made, in such a fashion.24 Because the assumptions underlying Dr. Calhoun’s “exceptions” analysis, which Defendants have proven through un-eontroverted deposition testimony to be faulty, render the analysis irrelevant, the court will grant ’Defendants’ motion to exclude Dr. Calhoun’s exceptions analysis testimony.
b) Designation of Prince George’s Plaza, Forest Village, Iverson Mall, City Place, and Laurel Center as No-Check Stores
Plaintiffs also offer as evidence of discrimination Defendants’ decision to stop accepting checks at the Prince George’s Plaza (190), Forest Village (180), Iverson Mall (7875), City Place (1199), and Laurel Center (119) stores, while allegedly continuing to accept checks at stores with worse check return expense histories located in less African-American areas.
The Forest Village, Prince George’s Plaza, and Iverson Mall stores stopped taking checks as part of the debit card pilot program in October 1995.25 In arguing that these no-check decisions are evidence of discrimination, Plaintiffs point to KB’s list of “Store Risk Groups” from the third quarter of 1995, which categorized the Iverson Mall store as “medium risk.” See Paper no. 77, Ex. 12. They note that Defendants stopped accepting checks at “medium risk” Iverson, while continuing to accept checks at the Tyson’s Corner Mall in Northern Virginia (Store 1085), which was categorized as “super high risk,” and the Springfield, Virginia store (Store 8060), which was categorized as “high risk.” See id. They further note that the areas surrounding the Tyson’s Corner and Springfield stores have a much lower percentage of African-Americans than the three Maryland stores. See id., Ex. 16.
This evidence is insufficient to raise an inference of discrimination or to show pretext for several reasons. First, although the Iv-erson Mall store was categorized as “medium risk,” the Forest Village and Prince George’s Plaza stores were categorized as “super high risk.” See id., Ex. 12. More importantly, these three stores were on the list of eighteen stores identified in the 1995 return check proposal as having cheek return expense in excess of eight percent, while the Tyson’s Corner and Springfield stores were not. See Paper no. 83, Ex. 12; Stengl Dep., Ex. 8.26 Thus, the record indicates that, regardless of the risk categories, the Iverson Mall, Forest Village, and Prince George’s Plaza stores all had worse check histories than the stores with lower African-American populations identified by Plaintiffs. Additionally, the District Sales Manager of the three Maryland stores, Phil Lewis, testified in his deposition that he was concerned about the high check return expense rates at those stores and consulted with Stengl for ways to reduce them, apparently resulting in the no-check policy at those stores. See Paper no.77, Ex. P., Deposition of Phil Lewis (“Lewis Dep.”) at 41-42. Lewis, who is African-American, testified in his deposition that the switch from checks to debit cards at the problem stores in his district was based on bad check performance and not on race. Id. at 42.
*198Similarly unpersuasive is Plaintiffs’ argument that the substitution of the debit card system for checks at the stores at issue is pretextual because checks and debit cards need not be mutually exclusive. Plaintiffs argue that a jury could easily infer that the debit card program was a cover for deciding to stop accepting checks at these stores in areas with a higher than average African-American population. They argue that the evidence shows that there is no explanation for why the debit card program was implemented primarily in stores in African-American areas. However, the fact that Defendants chose to accept debit cards in some stores instead of checks does not demonstrate that their proffered rationale of reducing cheek return expense is false. Indeed, these stores appear to have installed the debit card pilot system at least in part for the purpose of eliminating check return expense, and continuing to accept checks would render the debit card system somewhat useless. Additionally, contrary to Plaintiffs’ assertion, the record shows that the Forest Village, Prince George’s Plaza, and Iverson Mall stores were chosen for the pilot program because they were all identified in the 1995 check return expense reduction proposal as having check return expense rates in excess of eight percent and because of the initiative of their District Sales Manager, Phil Lewis.
The Laurel (119) and City Place (1199) stores stopped accepting checks in 1998. Plaintiffs argue that these stores stopped accepting checks while stores with comparable or less favorable check histories, but located in lower percentage African-American communities, continued to accept checks. The Laurel store had a 9.45% check return expense rate and was located in a market area that was 21% African-American. City Place had a check return expense rate of 7.84% and was located in an area that was nearly .40% African-American. See Calhoun Rebuttal at Table CC-5. Plaintiffs contend that Store 16, which continued to accept cheeks, had a check return expense rate of 12.75% and an African-American population of only 7.5%. See id.
However, even if the check return expense rates at the Laurel and City Place stores were lower than that of Store 16, the record shows that race was not the determining factor behind the decision to stop accepting checks at those stores. Defendants have offered evidence that the Laurel store stopped taking checks at the request of District Sales Manager Brian Metcalf, who was concerned with bad check return expense rates, not race. See Metcalf Dep. at 13-15. E-mail communication between Metcalf and Stengl clearly indicates that Metcalf wanted to stop accepting checks in, order to remedy what he perceived to be a bad check problem at that store. See Metcalf Dep., Ex. 2. Similarly, District Sales Manager Lewis testified in his deposition that he noted an increase in the check return expense rate at the City Place store after the Forest Village, Iverson Mall and Prince George’s Plaza stores stopped accepting checks, and that this increase was the reason for the subsequent refusal to accept checks.27 See Lewis Dep. at 53-56. Plaintiffs have not forecast any evidence to overcome this testimony that race was not the reason for placing the Laurel and City Place stores on the no-check list.
c) Opening of Greenbelt Store as a No-Check Store
Plaintiffs contend that the June 1997 opening of the Beltway Plaza store in Greenbelt (Store 2453) as a no-check store is further evidence from which a jury could infer that KB’s no-check policy was discriminatory because, as a new store, there is no check history to which Defendants may attribute the decision. Rather, Plaintiffs argue, the fact that KB researched and recorded the racial demographics of the area surrounding Beltway Plaza, including creating a detailed map of the area, demonstrates that they considered the race of the store’s customers in making the no-check acceptance decision. See Paper no. 83, Ex. 21; Ex. K, Deposition of Julie Evans (“Evans Dep.”) at 101. They also note that the Columbia, Maryland store, *199which opened the same day and serves a predominately white population, accepted checks upon opening.
Plaintiffs, however, have not provided any evidence that any racial demographic information concerning the Beltway Plaza store was used to determine its check policy, as opposed to merchandise selection or some other legitimate purpose. They have provided no basis for the inference that District Sales Manager Studebaker, Tim Lee, or any other KB employee involved in cheek acceptance decisions used the demographic information from the real estate file in connection with the decision to refuse checks at the store. The deposition of Julie Evans, the market research analyst who researched the demographic information, also revealed no connection between her work and check acceptance policies.
Moreover, Defendants have provided deposition testimony that the decision to open the Greenbelt store as a no-check store was based on a factor other than race. Studebaker testified in her deposition that Store Manager Luton, who is African-American, requested that the store not accept checks because the store was in close proximity to two other stores that already did not accept checks, Prince George’s Plaza (190) and Capital Plaza (466). See Studebaker Dep. at 69-74. Luton does not recall specifically requesting that Greenbelt open as a no-check store, but rather that he responded affirmatively when Studebaker asked him whether the debit card system from the closing Capital Plaza store should be brought to the Greenbelt Store. See Luton Dep. at 26-29. He believed that the debit card system would be more convenient than checks for both customers and store employees. See id. Additionally, Ruth Stewart, Store Manager for Store 190, who was assisting with the opening of the Greenbelt store, testified that when Studebaker asked her if she thought the Greenbelt store should accept checks, she replied no because she thought it should be consistent with the other stores in the area, which did not accept checks. See Paper no. 89, Ex. R, Deposition of Ruth Stewart at 26-27. Stewart explained that she recommended to Studebaker that the Greenbelt store should use the debit card system instead for the sake of customer convenience and because it was an advancement over checks. Id. at 27.
Finally, the fact that the Columbia Mall store opened around the same time as the Greenbelt store and accepted checks does not raise an inference of intentional discrimination because, unlike the Greenbelt store, the Columbia Mall store was not located near the Prince George’s Plaza and Capital Plaza stores, which used the debit card system. As discussed above, proximity to those stores was a factor in deciding to use the debit card system at the Greenbelt store—a factor not present with respect to the Columbia Mall store. Thus, the fact that the Columbia Mall store accepted checks is irrelevant and does not raise an inference of discrimination.
d) Defendants’ Collection of Racial Demographic Information and Subjective Implementation of No-Check Policy by Employees With Access to Such Information
Plaintiffs also argue that the fact that check acceptance decisions were highly subjective and that KB managers, including those who had input into no-check decisions, possessed racial demographic data for stores’ customers provides further evidence of pretext. It is undisputed that KB’s real estate files contained census tract demographic information, including the racial demographics of the areas where stores are located, and that certain designations were given to stores for inventory replenishment purposes. However, Plaintiffs have failed to offer evidence of any connection between the racial information collected or classifications made and KB’s no-eheck decisions sufficient to raise an inference of intentional discrimination. For example, there was no evidence offered that any KB executives or managers involved in real estate or inventory replenishment decisions with access to the information at issue were also involved in no-cheek decisions. Plaintiffs offered no evidence to counter witness testimony that classifications such as “Black A” and “Black B” were used strictly to determine the allocation of ethnic merchandise, specifically dolls, among stores. See Paper no. 89, Ex. D, Deposition of Ste*200ven Fruehterman at 73, 80; Ex. 0, Deposition of Robert Muller at 44-47, 99-100.
Plaintiffs’ contention that field managers were aware of the racial composition of their stores’ customers is irrelevant because Plaintiffs failed to offer any evidence that KB executives considering stores for the no-check list ever contacted field managers to inquire about racial demographics. Even if some KB executives themselves visited stores in the Washington-Baltimore area and were generally aware of the racial composition of their customers, Plaintiffs have still not offered anything to permit the logical leap from general awareness of racial demographics to use of the demographics for the purpose of discrimination. Mere collection and availability of racial demographic information, without more, does not support the inference of intentional discrimination or prove pretext. See Powell, 181 F.Supp.2d at 564.
e) Additional Purported Non-Discriminatory Reasons
Finally, Plaintiffs also assert that a jury may infer pretext because of a lack of evidence supporting Defendants’ purported business rationales that the survival of area stores was threatened by check return expenses, that a check guarantee service was unavailable, or that Defendants’ no-cheek practices were in keeping with comparable retailers. The problem with these arguments, however, is that Defendants never actually claimed to have stopped taking cheeks for any of these reasons. KB officials testified repeatedly in depositions that the rationale for placing stores on the no-check list was reducing check return expenses, not concern for profitability. See Stengl Dep. at 432-433. No deponents claimed that the survival of a store was at risk if it was not placed on the no-check list. Similarly, KB never claimed that it was forced to stop accepting checks in certain stores because a check guarantee service was prohibitively expensive. Thus, the fact that KB decided in August 2000 to begin using a check guarantee service is irrelevant to the question of pretext. Lastly, Defendants never maintained that they stopped accepting checks at certain stores because this was the industry practice. Therefore, whether or not other comparable retailers in malls where KB had no-check stores also refused payment by check is irrelevant for the purpose of pretext. Similarly, whether comparable retailers accepted payment by check is irrelevant because departure from industry practice does not constitute proof of intentional discrimination.28
Despite the extensive discovery conducted by the parties, including the statistical analy-ses and the multitude of depositions, Plaintiffs have failed to forecast evidence sufficient to show that Defendants’ legitimate business reason for placing any store on the no-check list was a pretext for discrimination and intended to prevent African-Americans from being able to pay by check. The quantity of evidence forecast does not overcome the deficiencies in the quality of the evidence. For these reasons, even though Plaintiffs established a prima facie case of intentional discrimination, they do not prevail on the motion for summary judgment because they have not forecast sufficient evidence that the legitimate, non-discriminatory reason proffered by Defendants is a pretext for discrimination.
IV. Conclusion
For the foregoing reasons, the court shall deny the motion to strike the class allegations, grant the motion for class certification, grant the motion to waive the local rule limiting the length of memoranda, grant the motion to exclude the testimony of Calhoun only as to the “exceptions” analysis, grant the motion to exclude the testimony of Shan-*201kar and Feinberg, and grant the motion for summary judgment. A separate Order will be entered.
ORDER
For the reasons stated in the foregoing Memorandum Opinion, it is this 21st day of July, 2003, by the United States District Court for the District of Maryland, ORDERED that:
1. Defendants’ motion to strike the class action allegations (paper no. 74) BE, and the same hereby IS, DENIED;
2. Plaintiffs’ motion for class certification (paper no. 78) BE, and the same hereby IS, GRANTED;
3. Plaintiffs Avis E. Buchanan, Dr. Albert R. Conley, Ardelia Crawford, Carolyn Kornegay-Belton and Yvette D. Tate are hereby CERTIFIED pursuant to Fed. R. Crv. P. 23(b)(2), for purposes of determining class-wide liability, injunctive and declaratory relief and class-wide punitive damages, as representatives of a class comprised of African-Americans who were prevented from paying by check at the following KB Toy Stores in the Washington-Baltimore metropolitan area between 1992 and August 2000:(1) Forest Village Park Mall in Forest-ville, Maryland; (2) Prince George’s Plaza in Hyattsville, Maryland; (3) Laurel Center in Laurel, Maryland; (4) Iverson Mall in Temple Hills, Maryland; (5) Beltway Plaza in Greenbelt, Maryland; (6) Mondawmin Mall in Baltimore, Maryland; (7) Reisterstown Road Plaza in Baltimore, Maryland; and (8) City Place in Silver Spring, Maryland;
4. Defendants’ motion to exclude all testimony and evidence from Plaintiffs’ expert witness Charles Calhoun (paper no. 76) BE, and the same hereby IS, GRANTED as to the “exceptions” analysis;
5. Defendants’ motion to exclude all testimony and evidence from Plaintiffs’ expert witnesses Venkatesh Shankar and Richard Feinberg (paper no. 75) BE, and the same hereby IS, GRANTED;
6. Defendants’ motion to waive the local rule limiting the length of memoranda (paper no. 90) BE, and the same hereby IS, GRANTED;
7. Defendants’ motion for summary judgment (paper no. 77) BE, and the same hereby IS, GRANTED;
8. JUDGMENT BE, and the same hereby IS, ENTERED in favor of Defendants Consolidated Stores Corp. and KB Consolidated, Inc. and against the certified plaintiff class;
9. It is hereby DECLARED that Defendants’ check acceptance policy between 1992 and August 2000 did not violate the right of the plaintiff class under 42 U.S.C. § 1981 to make and enforce contracts; and
10. The Clerk will transmit copies of the Memorandum Opinion and this Order to counsel for the parties and CLOSE this ease.
. In this opinion, Consolidated Stores and KB Consolidated will be referred to collectively as "Defendants” or simply "KB”.
. As of August 2000, all KB Toy Stores, including the ones at issue in this case, allowed customers to pay by check and used a check guaranty program. See Paper no. 83, Ex. 3.
. A store’s check return expense rate consists of the store’s bad checks received as a percentage of the store’s total check sales. In other words, it is the percentage of payments by check received that bounced.
. These stores' check return expense rates ranged from 10.02% to 38.56% and averaged 14.09%. See id. at CSC 0105. The check usage as a percentage of total sales ("check usage rate") at these 25 stores ranged from 0.93% to 14.35% and averaged 4.7%. See id.
. Defendants assert that the court’s authority to strike allegedly improper class action allegations from a complaint without dismissing the case in its entirety stems from Fed. R. Civ. P. 23(d)(4) which states that the court "may make appropriate orders ... requiring that the pleadings be amended to eliminate therefrom allegations as to representation of absent persons, and that the action proceed accordingly ....’’
. On March 15, 2001, Plaintiffs provided to Defendants the declarations of eighteen known class members, and on June 12, 2001, they produced the declarations of two additional known class members. Defendants deposed all five named Plaintiffs and eleven of the twenty known potential class members.
. The court has the discretion to require notice in the case of a class action brought under Rule 23(b)(2), but declines to do so here. See Fed. R. Civ. P. 23(d).
. Both parties correctly noted that it is inappropriate to consider the merits of the claims at stake when deciding whether to certify a class, see Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 177, 94 S.Ct. 2140, 40 L.Ed.2d 732 (1974), yet both disregarded this and spent considerable time and space arguing the merits of their case. For the purpose of the class certification motion, the court, however, will only address the class action requirements.
. Defendants do not challenge Plaintiffs' assertions that the numerosity and, except as noted in the previous section, adequacy of representation requirements are met. Plaintiffs estimate, based on deposition testimony, that the proposed class consists of over 3,000 members. See Paper no. 78 at 30-31. Based on this estimate, it is clear that the numerosity requirement is met because the potential class members are "so numerous that joinder of all members is impracticable.” Fed. R. Civ. P. 23(a)(1). The adequacy of representation element is also clearly met because there are no conflicts of interest between the class representatives and the proposed class members, and Plaintiffs have demonstrated their intent to pursue this action vigorously on behalf of the class by retaining qualified counsel. See Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 626, n. 20, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997); General Tel. Co. v. Falcon, 457 U.S. 147, *187157, n. 13, 102 S.Ct. 2364, 72 L.Ed.2d 740 (1982) (holding that the key issues with respect to adequacy of representation are "the competency of class counsel and conflicts of interest”).
. These three stores were placed on the no-check list in October 1995 when KB implemented a debit card pilot program in the area. The other five stores which Plaintiffs seek to include in their class claims were placed on the no-check list at other times.
. Indeed, in Kernan v. Holiday Universal, Inc., 1990 WL 289505 (D.Md., Aug.14, 1990), a case fairly similar to the instant case, the court found the fact that the allegations at issue covered a "long time span”, to be probative of commonality-
. Plaintiffs' motion to sever the individual damages claims from the claims for class-wide liability and injunctive relief was made pursuant to Rule 42(b), which states that "[tjhe court, in furtherance of convenience or to avoid prejudice, or when separate trials will be conducive to expedition or economy, may order a separate trial of any claim, ... or of any separate issues
. A similar bifurcation was granted by the district court in Gratz v. Bollinger, - U.S. -, 123 S.Ct. 2411, 2418, 156 L.Ed.2d 257, 2003 WL 21434002 *7, 71 USLW 4480 (2003).
. Plaintiffs contend that the fact that the no-check policy was abolished in August 2000 does not render injunctive relief moot. The court agrees. Several courts have certified classes under Rule 23(b)(2) where the defendants have taken steps following the filing of a lawsuit that effectively moot the requested injunctive relief. See, e.g. Wetzel v. Liberty Mutual Ins. Co., 508 F.2d 239, 251 (3d Cir.1975); Arnold v. United Artists Theatre Circuit, Inc., 158 F.R.D. 439, 456 (N.D.Cal.1994). See also Friends of the Earth, Inc. v. Laidlaw Environmental Services, Inc., 528 U.S. 167, 189, 120 S.Ct. 693, 145 L.Ed.2d 610 (2000) ("a defendant's voluntary cessation of a challenged practice does not deprive a federal court of its power to determine the legality of the practice ... If it did, the courts would be compelled to leave the defendant 'free to return to his old ways.'"). A request for injunctive relief is only mooted by the defendant’s voluntary conduct if " 'subsequent events made it absolutely clear that the allegedly wrongful behavior could not reasonably be expected to recur.' ” Id. (internal citation omitted). In this case, Defendants' commitment to the new check acceptance policy is questionable, having instituted it just several months after the filing of this lawsuit. See Mack v. Suffolk County, 191 F.R.D. 16, 21 (D.Mass.2000).
. Plaintiffs have also requested that the court order notice to all members of the class, which is discretionary when a class is certified pursuant to Rule 23(b)(2) rather than 23(b)(3). See Fed. R. Civ. P. 23(d). As already noted in footnote 7, the court declines to do so.
. Defendants have filed a motion to waive the local rule limiting the length of memoranda, for the purpose of their reply memorandum. Because of the length of Plaintiffs’ opposition and the numerous issues involved in this case, that motion will be granted.
. Although this framework, first set forth in McDonnell Douglas, originally involved employment discrimination claims under Title VII, the Supreme Court has held that the burden-shifting scheme is equally applicable to § 1981 claims. See Patterson v. McLean Credit Union, 491 U.S. 164, 185-87, 109 S.Ct. 2363, 105 L.Ed.2d 132 (1989); Hawkins v. PepsiCo, 203 F.3d 274, 278 (4th Cir.2000).
. Defendants once again argue that Plaintiffs cannot sustain a claim because, unlike some other cases brought under § 1981, they allege disparate treatment among different locations rather then disparate treatment of African-Americans and whites at any particular location. See, e.g., Murrell, 262 F.3d at 257 (hotel allegedly treated inter-racial party differently than it would have treated white party at same location); Callwood, 98 F.Supp.2d at 705-07 (African-Americans allegedly treated differently than whites at same restaurant location). However, this court has already considered and rejected this argument at the motion to dismiss stage. Plaintiffs’ claim of "retail redlining,” where discrimination is based on the racial composition of an area rather than the race of an individual person, is similar to claims brought in the fair lending context, which have been held to state a claim under § 1981 and other civil rights statutes. See, e.g., Old West End Ass’n v. Buckeye Fed. Sav. & Loan, 675 F.Supp. 1100, 1102-03 (N.D.Ohio 1987). Their claim alleges disparate treatment based on race, just simply in a different form.
. The parties address the bulk of the evidence in the context of whether Plaintiffs have presented a prima facie case of intentional discrimination. The court finds it more logical to assess the evidence in the context of pretext and the ultimate burden.
. "Higher than the national average African-American population" does not mean, however, majority African-American. The average African-American population is 12.3%. See Calhoun Rebuttal at 1. Thus, stores with higher than average African-American clientele could still serve a predominately non-African-American population.
. Instead, Dr. Calhoun dismisses this observation as irrelevant. See Calhoun Rebuttal at 8.
. While the statistical analyses may show that the no-check list had a disparate impact on areas with higher than average African-American populations, disparate impact is insufficient for a claim under § 1981. See, e.g., Powell, 181 F.Supp.2d at 564.
. Under this theory, Plaintiffs must abandon their claims with regard to the Mondawmin Mall and Reisterstown Road Plaza stores because these stores, both serving predominately African-American clientele, were among those with the worst check return expense rate when they were placed on the no-check list in 1992. See Stengl Dep., Exs. 4, 5.
. To the extent an exceptions analysis based on the 10/5 Rule could provide relevant information, it could only do so for the time period in which the rule actually may have been in effect, namely around 1991-92.
. These Washington area stores were involved in the pilot program, as well as stores in Southern California and Puerto Rico. While the parties dispute whether the debit card program was instituted in the Washington area at the recommendation of KB's bank, or whether it was chosen for some other reason, there is no dispute that a pilot program was indeed implemented.
. The Prince George’s Plaza store had a check return expense rate of 13.10%, Forest Village had a rate of 10.73%, and Iverson Mall had a rate of 10.87%. See id.
. The accounts by the two District Sales Managers about their role in deciding to place a store on the no-check list is consistent with Tim Lee’s testimony that the ad hoc process for determining check acceptance after 1996 included input from field personnel. See Lee Dep. at 55-57.
. Defendants have filed a motion to exclude the testimony of expert witnesses Venkatesh Shankar and Richard Feinberg, who conducted surveys concluding that KB’s practice of refusing checks at certain stores was not in keeping with industry practice (both witnesses) and that KB had available alternative measures for reducing check expense (Feinberg). See Paper no. 86, Exs. 3, 10. Having concluded that industry practice and the availability of a check guarantee service are irrelevant to the determination of pretext, the court will grant the motions to exclude the testimony of these two witnesses on relevance grounds. The court need not address the Daubert challenges to the reliability of their studies.