Robert G. FRANKLIN; Glenn E. Steele; Edward W. Lewis; Darryl Bailey; Leon Booker, Appellants, v. LOCAL 2 OF THE SHEET METAL WORKERS INTERNATIONAL ASSOCIATION, Appellee, 565 F.3d 508

RILEY, Circuit Judge. Appellants Robert G. Franklin (Franklin), Glenn E. Steele (Steele), Edward W. Lewis (Lewis), Darryl Bailey (Bailey), and Leon Booker (Booker) (collectively, Appellants) are African-American current or former members of Local 2 of the Sheet Metal Workers International Association (Local 2). Appellants filed a civil rights action against Local 2, alleging (1) racial discrimination in job placement under theories of disparate impact and disparate motive, and (2) retaliation. The district court granted

summary judgment to Local 2 on the disparate motive and retaliation claims, but denied

summary judgment on the disparate impact claim. Following a bench trial on the disparate impact claim, the district court found the evidence was insufficient to establish Local 2’s referral procedures had a disparate impact on African-American members. Appellants challenge the district court’s grant of

summary judgment to Local 2 on the retaliation claim and the district court’s denial of relief on the disparate impact claim. We affirm the district court’s judgment with respect to the disparate impact claim, but we reverse the district court’s

summary judgment for Local 2 on the retaliation claim. I. BACKGROUND A. Parties Local 2 is an affiliate of the Sheet Metal Workers International Association. Local 2’s jurisdiction includes 100 counties in Kansas and Missouri, and its principal office is in Kansas City, Missouri. Local 2 has approximately 1,600 members, with 1,200 members working in the area of building trades. Of Local 2’s 1,200 members in building trades, approximately sixty-five members are African-American, and of these sixty-five, approximately twenty-five are apprentices. Franklin, Lewis, and Bailey are current members of Local 2.

Steele is a current member of Local 2; however, Steele retired from the trade in February 2005. Booker was a member of Local 2 until he retired in April 2004 due to a disability. B. Local 2’s Referral Procedures The primary function of Local 2 is to negotiate the terms and conditions of its members’ employment with contractors who are either signatories to a collective bargaining agreement (CBA) with Local 2 or who recognize Local 2 as the bargaining representative for sheet metal workers within Local 2’s jurisdiction.

Under Local 2’s “Referral Procedures,” signatory contractors are required to provide Local 2 the “first opportunity ... to provide suitable journeymen sheet metal workers.” In the event Local 2 “fails to supply journeymen sheet metal workers within forty-eight hours,” the contractors are free to hire workers outside of Local 2’s membership. Contractors can comply with the CBA’s “first opportunity” provision by hiring Local 2 members through one of three procedures: (1) solicitation, (2) request, or (3) referral. A solicitation occurs when a Local 2 member applies directly to a contractor for a job.

With a solicitation, the hiring decision is in the sole discretion of the contractor. Local 2 does not participate in the hiring decision, but Local 2 requires the contractor to send paperwork to Local 2 containing the hired member’s name and the date on which the member was hired. Solicitation is the most common procedure under which members obtain employment. A request hiring occurs when a contractor contacts Local 2 and asks for a particular member by name. Local 2’s only participation in a request hiring is to receive the request and to record the date and the name of the member.

A referral occurs when a contractor contacts Local 2 and asks Local 2 to refer members with specific skills, rather than requesting particular members by name. To fill the needs of its members and the contractors who ask for referrals, Local 2 maintains an out-of-work list (OWL). Journeymen members of Local 2 who are available for employment sign the OWL, and when a contractor contacts Local 2 for a referral, Local 2 uses the OWL to refer members. Generally, members on the OWL are referred on a “first in, first out” basis, meaning those who have been on the list the longest are referred first.

There are exceptions. First, if a contractor asks Local 2 to refer a member with a particular skill, Local 2 goes down the OWL until it finds a member with the relevant skill. Local 2 reviews the members’ experience and qualification cards to determine which skills each of the members on the list possess. Second, if the first qualified member on the OWL cannot be reached at the number he provided to Local 2 after two calls, Local 2 moves down the list and contacts the next member. If Local 2’s call is answered, but the member is not present, the member has one hour from the time of the call to contact Local 2.

If the member does not return Local 2’s call within the hour, Local 2 moves down the OWL to the next member. Third, when a contractor contacts Local 2 requesting a referral for a foreman, Local 2 is not required to use the OWL to refer a foreman. A member on the OWL may request to be placed in “will call” status, meaning the member wishes to retain his position on the OWL, but does not wish to be considered for referrals until he contacts Local 2 and changes his status. Similarly, a member who has been referred by Local 2 to a job outside of Local 2’s jurisdiction can retain his position on the OWL.

Local 2 will not consider the absent member for referrals until he is again available for work within the jurisdiction. A member’s name is removed from the OWL when the member (1) requests to have his name removed, or (2) has accumulated eighty hours of employment. If a member turns down three job referrals in areas in which the member indicated on his experience and qualification card that he would work, and before accumulating eighty hours of work, the member loses his place on the OWL and is placed at the bottom of the OWL. C.

EEOC Proceedings Between 2001 and 2002, Appellants each filed charges of discrimination with the Equal Employment Opportunity Commission (EEOC) against Local 2. Appellants alleged Local 2’s referral procedures discriminated against African-American members. Appellants filed subsequent charges of discrimination alleging Local 2 retaliated against them for filing their charges of discrimination by posting documents which identified Appellants by name and listed Local 2’s legal bills associated with defending against Appellants’ EEOC charges.

After investigating Appellants’ charges, the EEOC issued an Amended Determination on September 19, 2003, finding there was reason to believe Local 2 engaged in intentional discrimination against African-American members and subsequently retaliated against Appellants for filing EEOC charges. After the EEOC advised Local 2 to discontinue posting the legal bills which identified Appellants by name, Local 2 continued posting them. Local 2 eventually blacked out the Appellants’ names on the postings. However, Local 2 continued to read the bills and names at union meetings.

The EEOC issued right to sue letters to Appellants on October 6, 2005. D. District Court Proceedings On January 3, 2006, Appellants filed a complaint against Local 2 in the United States District Court for the Western District of Missouri. Appellants alleged Local 2’s referral procedures discriminated against Appellants on the basis of race, in violation of 42 U.S.C. § 2000e-2(c). Appellants alleged racial discrimination Under two theories:

(1) Local 2 intentionally discriminated against African-American members on the basis of race, and, in the alternative,

(2) Local 2’s referral procedures had a disparate impact on African-American members. Appellants next alleged Local 2 interfered with Appellants’ right to contract in violation of 42 U.S.C. § 1981, by referring white members over qualified African-American members who were ahead of the white members on the OWL. Appellants also alleged Local 2, in violation of 42 U.S.C. § § 1981 and 2000e-3(a), retaliated against Appellants for filing charges of discrimination with the EEOC.

Appellants contended Local 2 retaliated against them by revealing to contractors that Appellants had filed charges of discrimination, and by posting in the union hall monthly statements which identified Appellants and the legal bills associated with Appellants’ EEOC proceedings. Local 2 moved for

summary judgment on each of Appellants’ claims on August 15, 2007. Five days later, Appellants filed a cross motion for partial

summary judgment on Appellants’ retaliation claims, arguing Local 2’s posting of the legal bills constituted a per se violation of Title VII. On February 21, 2008, the district court denied Appellants’ partial

summary judgment motion and granted in part, and denied in part, Local 2’s

summary judgment motion. The district court denied Local 2’s

summary judgment motion with respect to Appellants’ discrimination claim based on disparate impact. The district court determined Appellants’ stated a prima facie case of disparate impact and also met their burden of showing there were alternative referral procedures available that would have a less adverse impact on African-American members. The district court, however, found there were no material issues of fact precluding

summary judgment on Appellants’ discrimination claim based on disparate treatment, and granted

summary judgment to Local 2. The district court also granted

summary judgment to Local 2, and denied partial

summary judgment to Appellants, on Appellants’ retaliation claim. The district court determined Appellants met their burden to show

(1) Appellants engaged in protected conduct when they filed charges of discrimination, (2) a reasonable member would view Local 2’s actions as materially adverse, and (3) there was a temporal link between Local 2’s actions and Appellants’ protected activity. The district court then found Local 2’s proffered explanations that Local 2 was required to provide full disclosure of its legal bills to its members, and that it was Local 2’s longstanding practice to post legal bills with names listed, satisfied Local 2’s burden to show there were non-retaliatory reasons for the adverse employment action. The district court granted

summary judgment to Local 2 because the court found Appellants failed “to raise a question of fact as to pretext or motive for Local 2’s conduct.” Thus, the only claim that proceeded to trial was Appellants’ disparate impact claim. On June 23, 2008, the district court began a seven-day non-jury trial on Appellants’ disparate impact claim. Dr. Arthur Gutman (Dr. Gutman) testified as an expert witness for Appellants. Dr. Gutman prepared an expert report upon which Appellants relied to support their claim. In the report, Dr.

Gutman concluded white members of Local 2 worked significantly more hours than their African-American counterparts in the years 2000 to 2006. Dr. Gutman’s data indicated white members worked on average 1,566.26 hours per year during this period, while black members worked on average 1,338.75 hours per year. Dr. Gutman opined, “race-neutral factors [could not], statistically, account for the differences in hours worked.” On July 8, 2008, the district court issued an order, finding in favor of Local 2.

The district court determined the evidence was insufficient to establish Local 2’s referral procedures had a disparate impact on African-American members. The district court found “Dr. Gutman’s report [was] not reliable because of the assumptions he m[ade], unsupported conclusions he dr[ew], and variables he fail[ed] to consider in rendering his opinion.” One reason the district court gave for finding Dr. Gutman’s report unreliable was “Dr. Gutman failed to consider the individual characteristics of [Appellants] to determine if there were race neutral reasons” to explain the racial disparity in hours worked.

The district court noted the five Appellants made up approximately 13.5% of Local 2’s African-American sheet metal journeymen. The district court further noted, of the approximately thirty-seven African-American journeymen, fifteen to twenty were core employees, meaning they were steadily employed. Appellants made up 23% to 29% of the non-core African-American employees. The district court observed, “The record reflects significant impediments to each of the [Appellants’] employability which Dr.

Gutman’s report does not persuasively address.” The district court then discussed each of the five Appellants’ impediments, making the following findings. Franklin entered into settlement agreements with five separate sheet metal contractors to resolve complaints of discrimination. In-exchange for agreeing to no longer work for those particular contractors, Franklin received “nuisance value compensation.” Another contractor designated Franklin as ineligible for rehire.

The district court noted, “Franklin did not want to work outside of Kansas City, he was not willing to work as a boilermaker and he did not possess a cell phone in order to make himself more accessible for calls from Local 2 for work referrals.” The district court observed that Franklin did not come across as an honest person when he testified at trial. The district court pointed out that Lewis also entered into settlement agreements with two contractors to resolve discrimination complaints, accepting nuisance value compensation and agreeing not to seek work from those contractors in the future.

Like Franklin, Lewis was designated ineligible for rehire by -Local 2’s largest contractor after the contractor fired Lewis. The district court stated Lewis was willing to work in only one of Local 2’s six zones, did not possess a cell phone, and was demanding with respect to working conditions and with whom he was willing to work. The district court observed Lewis presented himself as unreasonable in his request for back pay and not fully committed to his position because he stopped appearing at the trial after he testified.

The district court noted Steele entered settlement agreements to resolve discrimination complaints with seven separate contractors, accepting nuisance value compensation and agreeing not to work for those contractors. Steele was fired from one job after he allegedly displayed aggressive behavior, and he did not have a cell phone.

The district court characterized Steele’s demeanor while testifying as “disrespectful” and “shocking.” The district court also discussed Booker’s impediments to employment, including his lack of a driver’s license and cell phone, his refusal to accept short-term work, and his failure to demonstrate personal initiative. The district court noted Booker’s absence on the final day of trial due to feeling ill.

According to the district court, “Bailey’s employment experience for the past five years [did] not support his claim.” Bailey had been fully employed since 2003, and in 2001 and 2002, when he was not fully employed, there were large numbers of members on the OWL. Bailey accepted nuisance value compensation and agreed not to seek work from one contractor as part of the settlement of his discrimination complaint, did not possess a cell phone, and refused to accept work as a boilermaker.

The district court observed Bailey did not present himself as a reasonable person because he would not accept work with less than a two-month duration, and he sought back pay for periods during which he was employed. Bailey was also absent from the trial for one full day without explanation. After discussing Dr. Gutman’s failure to consider Appellants’ impediments to employability, the district court next addressed Dr. Gutman’s failure to account for the impact of “core employees” on the total number of hours worked.

According to the district court, approximately 200 members were working as core employees during the relevant time frame. Dr. Gut-man testified he was not familiar with the concept of core employees, and thus, the district court concluded Dr. Gutman’s analysis and opinions were incomplete and speculative. The district court observed the referral rate of African-American members was equal to the referral rate of white members, and Appellants failed to prove a single instance of the union giving preference to a member contrary to the referral procedures. The district court declared, [E]ven if the Court assumes Dr.

Gut-man’s testimony meets the Daubert standard, the statistical evidence is overwhelmingly undermined by the circumstances and conduct of the individual [Appellants] (who represent a significant portion of the data sample) and the limited and narrow scope of data from which Dr. Gutman draws his conclusions.

The district court maintained the evidence demonstrated Local 2 could not control “the number of hours members worked, the conduct or motivation of the individual members, the number of qualified African-American members available on the OWL, when or whether a member was terminated or laid off, or to what extent contractors relied on core employees.” Finally, the district court found Local 2 demonstrated its current referral procedure was maintained in good faith and created incentives for Local 2, Local 2 members, and Local 2 contractors.

In contrast, the district court determined Appellants failed to propose or support an alternative referral procedure which would remedy the alleged disparate impact. Appellants raise two issues on appeal. First, Appellants argue the district court erred in dismissing Appellants’ disparate impact claim following the bench trial. Second, Appellants contend the district court erred in granting

summary judgment to Local 2, and denying partial

summary judgment to Appellants, on Appellants’ retaliation claim. II. DISCUSSION A. Disparate Impact Claim Appellants assert the district court erred by (1) failing to find Appellants established a prima facie case of disparate impact, (2) finding Dr. Gutman’s analysis unreliable, (8) imposing an intent requirement in the prima facie case, (4) making clearly erroneous factual findings regarding Dr.

Gutman’s consideration of “core” employees and Appellants’ employability, (5) using an incorrect legal standard to evaluate Local 2’s purported justification for maintaining its current referral procedures, and (6) finding Appellants failed to propose or support an alternative referral process which would have less of a disparate impact on African-American members. “Following a bench trial, ‘we review the trial court’s factual findings for clear error and its legal conclusions de novo.’ ” King v. United States, 558 F.3d 1156, 1160 (8th Cir.2009) (quoting Richardson v.

Sugg, 448 F.3d 1046, 1052 (8th Cir.2006)). “Using this standard, we will overturn a factual finding only if it is not supported by substantial evidence in the record, if it is based on an erroneous view of the law, or if we are left with the definite and firm conviction that an error was made.” Richardson, 448 F.3d at 1052 (citing Tadlock v. Powell, 291 F.3d 541, 546 (8th Cir.2002)). “A factual finding supported by substantial evidence, as well as a district court’s choice between two permissible views of the evidence, are not clearly erroneous.” Id. (citing Tadlock, 291 F.3d at 546).

A disparate impact claim challenges an employment, or union, practice that on its face is neutral, “but that in fact fall[s] more harshly on one group than another and cannot be justified by business necessity.” Mems v. City of St. Paul, 224 F.3d 735, 740 (8th Cir.2000) (quoting Hazen Paper Co. v.

Biggins, 507 U.S. 604, 609, 113 S.Ct. 1701, 123 L.Ed.2d 338 (1993)). “To establish a prima facie case for disparate impact, the plaintiff must show: (1) an identifiable, facially-neutral personnel policy or practice; (2) a disparate effect on members of a protected class; and (3) a causal connection between the two.” Id. (citation omitted) Appellants first argue the district court improperly rejected Dr. Gutman’s statistical analysis. Dr. Gutman concluded there was a disparity in hours worked between white members and African-American members and race-neutral factors could not account for the disparity.

The district court declared, “Dr. Gutman’s report is not reliable because of assumptions he makes, unsupported conclusions he draws, and variables he fails to consider in rendering his opinion.” Appellants claim Local 2 failed to rebut Dr. Gutman’s analysis, and, thus, the district court discounted the analysis based on speculation.

Statistical evidence which fails to include every measurable variable can be used to prove a plaintiffs case by a preponderance of the evidence; however, whether such evidence actually “carrfies] the plaintiffs’ ultimate burden will depend in a given case on the factual context of each case in light of all the evidence presented by both the plaintiff and the defendant.” Bazemore v. Friday, 478 U.S. 385, 400, 106 S.Ct. 3000, 92 L.Ed.2d 315 (1986) (per curiam) (Brennan, J., concurring and joined by all members of the Court). Appellants, citing Coble v. Hot Springs School Dist.

No. 6, 682 F.2d 721 (8th Cir.1982), suggest Local 2 was required to present its own expert testimony or statistical analysis to rebut Dr. Gutman’s analysis or to prove Dr. Gutman failed to consider relevant factors. In Coble, a class of female teachers sued a school district for sex discrimination in pay, job assignments, and promotions. See id. at 723. The district court granted judgment to the school district. Id. On appeal, the court addressed whether the teachers’ expert’s analysis was sufficient to establish the teachers’ discrimination claims. See id. at 729-30.

The court stated, The burden is on the opposing party to clearly rebut statistical evidence; hypotheses or conjecture will not suffice. When a plaintiff submits accurate statistical data, and a defendant alleges that relevant variables are excluded, defendant may not rely on hypothesis to lessen the probative value of plaintiffs statistical proof. Rather, defendant, in his rebuttal presentation, must either rework plaintiffs statistics incorporating the omitted factors or present other proof undermining plaintiffs claims. Id. at 730 (quoting Segar v. Civiletti, 508 F.Supp. 690, 712 (D.D.C.1981)).

Appellants quote the above portion of Coble in their brief, but fail to address the Coble court’s ultimate conclusion in that case that the statistical evidence was insufficient to state a prima facie case because the' expert’s methodology considered the teachers’ education and experience separately, rather than considering these two factors together. See id. at 731. The court reached that conclusion even though the court also determined the school district’s rebuttal evidence would have been insufficient had the teachers successfully established a prima facie case of discrimination.

Id. at 732 (“Assuming for the purpose of argument that appellants had established a prima facie case using multiple regression analysis, we think that the school district’s rebuttal evidence would have been inadequate to undermine the accuracy or significance of appellants’ statistical evidence.”). Appellants’ argument that Coble mandates reversal is faulty. Contrary to Appellants’ assertion, the district court here did not err in finding Dr. Gutman’s report unreliable and finding Appellants failed to establish a prima facie case of disparate impact.

Local 2 did not produce its own expert or statistical analysis accounting for the factors allegedly missing from Dr. Gut-man’s report. See id. Local 2 followed a trial strategy of attacking Dr. Gutman’s report and opinions as erroneous and unreliable, thereby defeating Appellants’ prima facie case. The district court found Dr. Gutman’s report unreliable after addressing, in detail, the deficiencies in Dr. Gutman’s analysis. For instance, Dr.

Gutman’s report focused solely on the average annual number of hours worked by white members compared to the average annual number of hours worked by African-American members. As the district court found, Local 2 had no “control over the number of hours members worked.” Dr.

Gutman admitted during cross examination at trial that he had assumed in his analysis that Local 2 was responsible for the number of hours accumulated by its members, and he conceded this assumption during the following exchange: Q If the union had no responsibility to control the duration of employment by the workers, then your analysis would not be valid with respect to the union? A Once again, anything’s possible in a hypothetical, but I find that to be a red herring. Q My question is, if the union does not control the duration of employment, is your analysis a valid measure of the union’s responsibility?

A You’re asking me to agree with a hypothetical that I disagree with based upon what I’ve read and understand. Q You have to assume that the union is responsible for the accumulation of all these hours, correct? A I do assume that for reasons stated. Q So if your assumption is wrong, then your numbers are wrong? A No, my numbers are right. Q If your assumption is wrong, your conclusion is wrong? A I don’t believe my assumptions are wrong. Q Would you answer the question that I asked? A If you phrase it that way, I’m forced to agree, but I disagree with what I’m forced to agree with.

We agree with the district court that Dr. Gutman’s admission that his analysis assumed Local 2 was responsible for the hours accumulated by its members fatally undermines the reliability of Dr. Gutman’s analysis and conclusions. Another reason the district court determined Dr. Gutman’s analysis was flawed was because Dr. Gutman failed to take into account the concept of “core employees.” Dr. Gutman testified he “would be shocked if you could find even one [member] that worked for the same employer” during the entire period of 2000 to 2006.

Yet, the evidence revealed approximately 200 members were continuously employed with the same contractor over that period. Other members had changed employers only once or twice over that period. Still others had remained continuously employed in some fashion over that time period. Finally, the district court noted Dr. Gut-man’s analysis failed to take into account the personal traits and conduct of the individual Appellants and how those characteristics might have affected the analysis and disparity in hours worked. Appellants assert this was an improper consideration.

We agree with the district court that the personal traits of Appellants may be relevant to Dr. Gutman’s analysis because there was a small number of African-American journeymen, and Appellants made up a significant portion of this group. We also agree with Appellants that there is no evidence Appellants’ individual conduct differed from the conduct of other Local 2 members and, if so, whether this would account for the disparity in hours worked. We further recognize the impact of the personal traits of Appellants would be a difficult factor for Dr.

Gutman to consider in his analysis, and it is unlikely Local 2 maintained meaningful data on the personal traits of individual members. However, any error on the part of the district court in faulting Dr. Gutman for failing to consider Appellants’ personal traits in Dr. Gutman’s analysis is harmless because we conclude the district court provided sufficient explanation and consideration of other factors which were lacking in Dr. Gutman’s analysis and which made Dr. Gutman’s report unreliable as a measure of disparate impact in this case.

As evidenced by Local 2’s Referral Procedures pamphlet, the only procedure under which the contractor does not have the sole authority to make the hiring decision is referral. With referrals, Local 2 is involved in the hiring decision, but Local 2’s discretion is limited by the strict “first in, first out” policy. As the district court noted, and Dr. Gutman conceded, African-American members were sent out to work by referral in proportion to their representation of the workforce. Dr.

Gutman’s analysis, which measured only the disparity in number of hours worked by African-American members compared to white members, was insufficient to satisfy Appellants’ burden to show Local 2’s referral procedures create a disparate impact in job placement. Appellants’ other arguments likewise do not warrant reversal of the district court’s judgment.

Appellants suggest the district court improperly imposed a duty on Appellants to prove discriminatory intent when the district court made the following statement: “[Appellants] did not contend or attempt to prove a single instance where the union gave anyone a preference contrary to the referral process.” When taken in context, we conclude this statement, rather than imposing an intent requirement, was an illustration of the shortcomings of Dr. Gutman’s analysis and conclusions and was the district court’s rebuttal of Dr.

Gutman’s statement that the referral process was somehow “fixed.” Appellants’ final two arguments concern the last paragraph of the district court’s order where the district court stated: As a final consideration, Plaintiffs do not propose or support a specific alternative process by which to remedy the disparate impact they claim exists in the current process. On the other hand, Local 2 posits a good faith basis for maintaining the referral process as it currently exists by supporting claims that it creates incentive and rewards for Local 2, Local 2 members, and Local 2 contractors.

Appellants maintain the correct legal standard was whether a “business necessity” justified Local 2’s maintenance of the process, see Hazen Paper, 507 U.S. at 609, 113 S.Ct. 1701, and the district court erred when it considered whether Local 2 had a “good faith basis” for maintaining its process and whether an alternative process would put contractors at a “competitive advantage” if Local 2 changed the process. Appellants also argue the district court clearly erred in finding Appellants had not proposed or supported a specific, less discriminatory alternative to the current referral process.

We need not address the validity of these arguments because we conclude the district court did not err in finding Appellants’ evidence, specifically Dr. Gutman’s report, was insufficient to establish a prima facie case of disparate impact., Once the district court made that finding, there was no need to analyze whether a business necessity justified maintenance of the process or whether there were less discriminatory alternatives. The district court merely provided an alternative basis for why it believed Appellants were not entitled to relief on their disparate impact claim. B.

Retaliation Claim Appellants also claim the district court erred in granting

summary judgment to Local 2, and in denying partial

summary judgment to Appellants, on Appellants’ retaliation claim. “We review a grant of

summary judgment de novo, viewing the facts in the light most favorable to the non-moving party.” Smith v. Int’l Paper Co., 523 F.3d 845, 848 (8th Cir.2008) (citation omitted). “Summary judgment is appropriate if there is no genuine issue of material fact and the moving party is entitled to judgment as a matter of law.” Id. (citations omitted).

To demonstrate a prima facie case of retaliation, Appellants must show: “1) [they] engaged in protected conduct; 2) a reasonable [member] would have found the challenged retaliatory action materially adverse; and 3) the materially adverse action was causally linked to the protected conduct.” Higgins v. Gonzales, 481 F.3d 578, 589 (8th Cir.2007) (citing Burlington Northern & Santa Fe Ry. Co. v. White, 548 U.S. 53, 68, 126 S.Ct. 2405, 165 L.Ed.2d 345 (2006)).

To meet their burden at the second prong of the prima facie case, Appellants “must demonstrate that a reasonable [member] would have found [Local 2’s] actions ‘materially adverse,’ meaning they ‘well might have dissuaded a reasonable worker from making or supporting a charge of discrimination.’ ” Jackson v. United Parcel Serv., Inc., 548 F.3d 1137, 1142 (8th Cir.2008) (citing Higgins, 481 F.3d at 589). “Once a plaintiff has shown a prima facie case of retaliation, the burden shifts to the employer to proffer a legitimate, non-retaliatory basis for the adverse employment action.” McCullough v.

Univ. of Arkansas for Med. Sciences, 559 F.3d 855, 864 (8th Cir.2009) (citation omitted). “The plaintiff then bears the burden of proving that the employer’s proffered reason was a mere pretext for a retaliatory motive.” Id. In the

summary judgment order, the district court determined Appellants established all three prongs of a prima facie case of retaliation. The district court found

(1) Appellants engaged in protected activity by filing EEOC charges,

(2) Local 2’s posting and announcing of the legal bills identifying Appellants by name may dissuade a reasonable member from filing EEOC charges, and (3) the temporal connection between the protected activity and the adverse employment action satisfied the causal link requirement. The district court then found Local 2 satisfied its burden to show a non-retaliatory reason for its actions by claiming it was Local 2’s longstanding policy to post legal bills, and Local 2 was required to obtain member approval for the legal bills.

After shifting the burden back to Appellants, the district court found Appellants failed “to raise a question of fact as to pretext or motive for Local 2’s conduct.” Thus, the district court granted

summary judgment to Local 2, and denied partial

summary judgment to Appellants. Appellants suggest the district court erred in denying Appellants’ motion for partial

summary judgment because Local 2’s actions constituted per se retaliation. Appellants claim the policy of posting and reading legal bills at meetings, and identilying by name members who have filed EEOC charges, is retaliatory on its face. Appellants rely on EEOC v. Bd. of Governors of State Colleges and Univs., 957 F.2d 424 (7th Cir.1992). In Bd. of Governors, the board had a policy in its CBA that once an employee filed an EEOC charge or initiated a lawsuit, the board was “authorized to halt existing or future grievance proceedings.” Id. at 425-26.

The Seventh Circuit determined this policy was retaliatory on its face, stating, “When an employee’s participation in statutorily protected activity is the determining factor in an employer’s decision to take adverse employment action, that action is invalid regardless of the employer’s intent.” Id. at 428. We disagree with Appellants that the district court erred in denying partial

summary judgment to Appellants. Bd. of Governors is distinguishable from Appellants’ case. Local 2 claimed it was required to obtain member approval for expenditure on all legal bills, and in that sense, Local 2, unlike the board in Bd. of Governors, was not singling out members who filed EEOC charges. Local 2’s actions do not constitute per se retaliation in this case. In the alternative, Appellants contend there were questions of material fact precluding

summary judgment on the issue of whether Local 2’s proffered non-retaliatory explanations for its actions were pretextual. We agree. “Courts have [long] recognized the difficulty in disposing of issues of discriminatory or retaliatory intent at the

summary judgment stage. ‘Summary judgments should be sparingly used and then only in those rare instances where there is no dispute of fact and where there exists only one conclusion.’ ” Davis v. Fleming Companies, Inc., 55 F.3d 1369, 1371 (8th Cir.1995) (quoting Johnson v. Minnesota Historical Soc’y, 931 F.2d 1239, 1244 (8th Cir.1991)). We conclude Appellants’ retaliation claim was inappropriate for

summary disposition. Based on the

summary judgment record, there were material questions of fact regarding whether Local 2’s proffered explanations were pretextual. The evidence raises a possible reasonable inference Local 2 was aware of a negative impact on Appellants based upon the listing and, during meetings, reading of Appellants’ names, claims, and related costs. Local 2 initially refused to end these practices after the EEOC requested a halt, but later blacked out Appellants’ names on the postings. However, Local 2 continued to read Appellants’ names at union meetings.

Although Local 2’s prior practice and obligation to disclose expenses may justify what Local 2 did, the degree of Local 2’s disclosures raises credibility issues and a potential reasonable inference of retaliation. Appellants’ retaliation claim cannot be resolved as a matter of law. Thus, Appellants meet the minimum requirement on their retaliation claim to show a genuine issue as to a material fact, see Fed.R.Civ.P. 56(c), and we reverse the district court’s

summary judgment for Local 2 on Appellants’ retaliation claim. III. CONCLUSION For the reasons stated in this opinion, we affirm the district court’s judgment on Appellants’ disparate impact claim and the district court’s ruling on Appellants’ partial

summary judgment motion, and we reverse the district court’s

summary judgment for Local 2 on Appellants’ retaliation claim. . As an initial matter, we note it is not clear from the district court’s opinion whether the district court excluded Dr. Gutman’s report as unreliable under Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993) or whether the district court merely considered reliability when assessing what weight to give Dr. Gutman's report. The district court cited Rule 702 of the Federal Rules of Evidence and the criteria to consider under Daubert immediately before concluding Dr.

Gutman's report was not reliable, but the court later declared, "even if the Court assumes Dr. Gutman’s testimony meets the Daubert standard, the statistical evidence is overwhelmingly undermined by the circumstances and conduct of the individual Plaintiffs ... and the limited and narrow scope of data from which Dr. Gutman draws his conclusions.” We find it unnecessary to analyze whether the district court made a Daubert ruling, and if so, whether it was correct, because Appellants do not raise a Daubert argument in their challenge to the district court’s reliability determination.

Moreover, this was a non-jury trial, the district court permitted Dr. Gutman to testify, and our review of the district court’s opinion makes clear the district court carefully considered the methodology and conclusions in Dr. Gutman’s report.