BALDOCK, Circuit Judge.
Monfort, Inc. and United Food and Commercial Workers Union, AFL-CIO, Local [1540]*1540Union No. 7-R (“Union”) petition for review of an order of the National Labor Relations Board (“Board”). Monfort challenges the Board’s finding that Monfort, when it reopened its Greeley, Colorado plant, unlawfully discriminated against former employees in hiring because of the former employees’ Union activity and membership. 29 U.S.C. §§ 158(a)(1), (a)(3). The Union seeks review of the Board’s remedy for these unfair labor practices. Additionally, Monfort seeks review of the Board’s finding that Monfort unlawfully terminated James Little because of his Union activity. Id. § 158(a)(4). Finally, Monfort seeks review of the Board’s remedy for Monfort’s unfair labor practices relating to its activities leading up to the Union election that followed the reopening of the Greeley plant. Id. §§ 158(a)(1), (a)(3). General Counsel seeks enforcement of the Board’s order in its entirety.1 Our jurisdiction arises under 29 U.S.C. §§ 160(e), (f).
I.
Initially, we address Monfort’s challenge to the Board’s finding that Monfort’s hiring practices when it reopened its Greeley plant were unlawful. “[A]n employer who declines to hire employees solely because they are members of a union commits a [29 U.S.C. § 158(a)(3) ] unfair labor practice.” NLRB v. Burns Int’l Sec. Servs., Inc., 406 U.S. 272, 280-81 n. 5, 92 S.Ct. 1571, 1578-79 n. 5, 32 L.Ed.2d 61 (1972). Accordingly, an employer may not discriminate against union members in its hiring decisions. See United Food & Commercial Workers v. NLRB, 768 F.2d 1463, 1475 (D.C.Cir.1985), enf'g, Spencer Foods, 268 NLRB 1483 (1984). In NLRB v. United States Postal Serv., 906 F.2d 482 (10th Cir.1990), we stated:
A violation of [29 U.S.C. § 158(a)(3)] is established where General Counsel demonstrates that an employer’s opposition to protected union activity was a motivating factor in a decision to take adverse action against an employee and the employer is unable to demonstrate that the adverse action would have been taken even absent the protected activity.
Id. at 486 (citing NLRB v. Transportation Management Corp., 462 U.S. 393, 403, 103 S.Ct. 2469, 2475, 76 L.Ed.2d 667 (1983)). Accord Wright Line, 251 NLRB 1083 (1980), enf'd, 662 F.2d 899 (1st Cir.1981), cert. denied, 455 U.S. 989, 102 S.Ct. 1612, 71 L.Ed.2d 848 (1982). Both the ALJ and the Board purported to apply this test, commonly known as the Wright Line test, and all the parties agree that this is the governing standard. Conceding the finding by both the Board and the ALJ that General Counsel established a prima facie case, see Pet’r Monfort Brief at 13, Mon-fort’s challenge relates solely to the second part of the analysis—i.e. whether Monfort rebutted the prima facie case.
Our review of the Board’s finding that Monfort failed to rebut the prima facie case is limited. We must uphold the Board’s factual findings if they are supported by substantial evidence in the record considered as a whole.2 Universal Cam[1541]*1541era Corp. v. NLRB, 340 U.S. 474, 488, 71 S.Ct. 456, 464, 95 L.Ed. 456 (1951); United States Postal Serv., 906 F.2d at 486. While we must consider the findings of both the Board and the AU in our review of the record as a whole, our “standard of review is not altered in cases in which the ALJ and the Board reached contrary conclusions.” Glaziers Local Union 558 v. NLRB, 787 F.2d 1406, 1411-12 (10th Cir.1986) (citations omitted). Nonetheless, “[ejvidence may properly be considered less substantial when the NLRB’s administrative law judge, ‘who has observed the witnesses and lived with the case,’ has drawn conclusions different from those reached by the NLRB.”3 Cartwright Hardware Co., Inc. v. NLRB, 600 F.2d 268, 270 (10th Cir.1979) (quoting Universal Camera, 340 U.S. at 496, 71 S.Ct. at 468). See also NLRB v. First Nat’l Bank of Pueblo, 623 F.2d 686, 693 (10th Cir.1980).
On March 31, 1980, Monfort closed its Greeley plant resulting in the permanent layoff of all of its more than 800 production employees who were all represented by the Union.4 By January 1982, Monfort had decided to reopen the plant.5 In preparation thereof, Monfort developed hiring criteria for applicants which considered absenteeism, discipline, medical condition, accidents, interest and ability, days, and hours of work, and attitude.6 Significantly, the specific criteria relating to absenteeism, medical condition and discipline were based on objective factors and could only disqualify an applicant from being hired.7
Monfort solicited employment applications from the general public for two days in January 1982, from former employees [1542]*1542for two days in March 1982, and then from the general public continually beginning in May 1982. In total from January 1982 to November 1983, Monfort received 7,287 applications, of which 352 were from former employees.8 The written application form did not require the applicant to provide the information necessary to determine whether the applicant met Monfort’s objective hiring criteria.
When Monfort received an application from a former employee, it reviewed the former employee’s personnel file. If the file disclosed that the former employee failed to meet the hiring criteria for absenteeism, discipline or medical condition, the former-employee applicant was disqualified without a personal interview. By contrast, Monfort had no way to prescreen new applicants (nonformer employees) based on Monfort’s objective hiring criteria because the application did not elicit the necessary information.
During interviews, applicants were disqualified if they were unwilling to accept the terms and conditions of employment, or unwilling to work a particular job. Mon-fort also determined, at the interview, whether the applicant met the hiring criterion relating to attitude.9 Monfort attempted to determine whether the applicant met the remaining hiring criteria through an authorization release form that Monfort would send to the applicant’s former employers. Approximately 50% of the forms were returned by former employers. The applicants whose former employers did not return the form were presumed to meet the objective hiring criteria as long as Monfort did not have specific disqualifying information.
Of the approximately 50% of the forms that were returned by applicants former employers, only 25-50% had complete information. Monfort made no effort to follow up on the information submitted by new applicants’ past employers. Thus, less that one quarter of the new applicants were evaluated based on a comparable amount of information as the former-employee applicants. Like the applicants whose former employers failed to return the forms, applicants whose former employers returned but did not complete the form were presumed to meet the hiring criteria unless Monfort had specific disqualifying information.
Even if the applicant’s former employer returned the form completed, Monfort could not equally evaluate a new applicant based on its objective criteria. The form asked the former employer to rate the applicant in various job related categories and to provide information on absences, lost-time accidents, discharge or other discipline within the previous thirty-six months. “These forms sought only generalized information from past employers about an applicant’s work history, attendance, discipline, and accident records, and they gave no guidance to employers about how to set forth such information in a manner permitting meaningful evaluation under [Mon-fort's] hiring criteria." 298 NLRB No. 16 at 22. Notably, the form did not request the specific information necessary for Mon-fort to apply its own objective hiring criteria. Like the other new applicants, Mon-fort presumed that applicants for whom it had received a completed form met the objective hiring criteria unless it received specific disqualifying information. Because of the lack of the form’s specificity, however, the specific disqualifying information essentially had to be volunteered by the former employer.
Not only was Monfort’s procedure substantially skewed in favor of new applicants by failing to evaluate them based on a comparable level of information that Monfort retained for former-employee applicants, even when Monfort had comparable information on a new applicant, it did not apply its objective criteria equally as [1543]*1543compared to former-employee applicants. For example, because the form requested only the total number of absences for a three year period, Monfort could not accurately apply its “absences-per-annum” standard to new applicants. See supra note 7. Accordingly, Monfort averaged the total absences over the period worked. Moreover, in computing the average annual number of absent days for new applicants, Monfort treated fractions of a year as full years resulting in a correspondingly better attendance rate than was actually the case. By contrast, Monfort did not give former-employee applicants the benefit of averaging; they were disqualified if their absences exceeded the standard set for any one year within the review period. Moreover, unlike new applicants for whom Monfort counted fractions of a year as a full year, Monfort did not consider interim employment records of former-employee applicants unless they had worked a full year.
Similarly, Monfort’s application of its medical condition criterion favored new applicants over former-employee applicants. The form sent to an applicant’s former employer did not request information regarding the applicant’s medical condition. Thus, Monfort had to rely on the applicant to provide the information necessary to apply its medical condition criteria. Even if Monfort received information that a new applicant had a history of a specific ailment that would disqualify him or her based on Monfort’s objective criteria, a new applicant was given the opportunity to undergo a current medical examination that could relieve him or her from disqualification. By contrast, a majority of former-employee applicants who were disqualified based on a medical condition were not given the benefit of a contemporaneous medical examination.
Finally, Monfort disparately applied its discipline hiring criterion as between former-employee applicants and new applicants. The criterion disqualified any applicant who had received a written disciplinary action within the previous three years. However, even if a former employer indicated that the new applicant had been disciplined within the past three years, Monfort did not disqualify the new applicant unless the former employer volunteered the reason for the discipline. The form sent to former employers did not request that the former employer list the reason for the disciplinary action. In cases in which the reason was volunteered, whether it would disqualify the new applicant was dependent entirely on the subjective judgment of Mon-fort’s personnel director. By contrast, Monfort treated documents in former employees’ personnel files less favorably than the forms returned by new applicants’ former employers documenting substantially similar misconduct.
Monfort reopened the Greeley plant on March 1, 1982, with 228 employees, 57 of whom were former employees. By March 25, Monfort employed 410 employees, 61 of whom were former employees.10 From the period when the plant reopened to November 1983, Monfort hired 3,150 production employees, 160 of whom were former employees.11 The new employees subsequently voted on whether to be represented by the Union. Following an election marred by numerous unfair labor practices by Monfort, see supra note 1, the production employees voted down the Union.
The Board, agreeing with the AU, found that “General Counsel ... made a strong prima facie showing in support of the allegation that [Monfort] engaged in an overall pattern of discrimination against former employees by disparately applying its hiring criteria.” 298 NLRB No. 16 at 28-29. See also AU Decision at 87-89. The Board inferred an anti-Union motivation in the disparate application of the hiring criteria based on
[Monfort’s] knowledge of its former employees’ union membership, participation in the 1979-1980 strike, and poststrike [1544]*1544grievance and other protest activities; [Monfort’s] open hostility to the employees’ union activities and to the possibility of unionization of the reopened Greeley plant work force; and the numerous unfair labor practices found here and in previously cited cases involving these parties at the Grand Island plant.
298 NLRB No. 16 at 21. See also AU Decision at 87; supra note 1. As noted above, Monfort concedes these findings.
However, the Board and the AU disagreed over whether Monfort had rebutted the prima facie case. The AU found that Monfort rebutted the prima facie case of discrimination by showing that the former-employee applicants who were not hired did not meet Monfort’s objective hiring criteria. AU Decision at 93. The AU supported his conclusion by relying on a statistical summary prepared by Monfort showing that the percentage of former-employee applicants who were hired was relatively equal to the percentage of new applicants who were hired. Monfort’s summary also purported to show that a significantly greater percentage of former-employee applicants who did not meet the objective hiring criteria were hired as compared to the percentage of new applicants who were hired but did not meet the criteria. The AU concluded that these statistics made it less probable that Monfort applied its hiring criteria in a disparate manner between former employees and nonformer employees. Id. at 90.
The Board disagreed with the AU as to whether Monfort had rebutted the prima facie case. The Board framed the issue as whether Monfort had “met its burden ... of proving that it would not have hired the alleged discriminatees who filed formal Monfort applications even in the absence of their union and other protected activities.” 298 NLRB No. 16 at 29. Reasoning that the AU “greatly misconceived the significance of statistical hiring comparisons,” id., the Board found that Monfort “failed to show that it would not have hired any of the alleged discriminatees who filed formal Monfort applications even in the absence of their union membership and activities.” Id. at 30. Consequently, the Board stated
we find that .the General Counsel has proven that [Monfort] discriminated against former employees who sought reemployment at the reopened Greeley plant by disparately applying its facially neutral hiring criteria to the detriment of those individuals. Accordingly, we further find that [Monfort] violated [29 U.S.C. §§ 158(a)(3), (a)(1) ] when it failed to rehire or delayed rehiring those alleged discriminatees who filed formal applications for reemployment at the Greeley plant.
Id. at 31.
Monfort points to its statistical summary as requiring a finding that it did not discriminate against former employees as a class. While we have little doubt that statistics evidencing equal treatment between two classes of employees may be used to rebut a prima facie case of discrimination, Monfort’s statistics do not rebut the particular pattern of discrimination in this case.
First, Monfort’s statistics indicate that 160 out of 352, or 45% of former-employee applicants were hired through November 1983; and, 2,990 out of 6,935, or 43% of new applicants were hired during this same period. Monfort contends that because experience in the meat packing industry was not one of the hiring criteria, it is logical that the percentage of former-employee applicants hired and the percentage of new applicants hired should be relatively equal. Nonetheless, these statistics are “not directly responsive to the nonstatistical evidence of specific discriminatory treatment revealed by the prima facie case.” Id. at 30. The prima facie case was based on Monfort’s disparate application of its hiring criteria. Id. at 28-29. Thus, the discriminated class is former-employee applicants. Given that Monfort’s application of its hiring criteria was far more rigorous for former-employee applicants than new applicants, and that in many cases, new applicants were not even subject to the hiring criteria because Monfort did not have the necessary information, Monfort’s comparison of the hiring rates between the two groups simply does not address the discrimination found by the Board.
[1545]*1545Monfort also relies on statistics showing that its hiring rate for former non-Union employees — i.e. supervisors and clerical— was less than the hiring rate of former Union employees. According to Monfort, of the 79 supervisory employees hired when the Greeley plant reopened, 22, or 28%, were former employees; furthermore, of the 80 clerical employees, 6, or 8%, were former employees. Monfort argues that if it had discriminated against former Union employees, it would have hired former nonUnion employees at rates exceeding the rate of hire for former Union employees. Notwithstanding that Monfort’s hiring practices regarding non-Union employees are simply not responsive to the discrimination found by the Board, Monfort’s own statistics refute its contention. Of the 3,150 production employees hired from when the plant reopened to November 1983, 160, or only 5%, were former employees. Thus, the hiring rate of former nonUnion employees exceeds the hiring rate of former Union employees.
Monfort’s final statistical argument relates to “mistakenly hired” employees. Monfort defines this group of employees as persons who were hired despite the fact that Monfort possessed information which should have disqualified them under Mon-fort’s objective hiring criteria. According to Monfort, 79 of the 3,150 employees were hired by mistake. Of these applicants, 25 were former employees while 54 were new employees. Monfort argues that because 25 out of 352, or 6.8%, of former-employee applicants were hired by mistake, whereas only 54 out of 6,918, or 0.8%, of new applicants were hired by mistake, unqualified former-employee applicants were hired nine times as often as unqualified new applicants. This argument overlooks the substantial evidence that, due to Monfort’s application of its hiring criteria, it had no way to assess the qualifications of new applicants based on its own objective hiring criteria. Thus, the comparison simply is not valid given that Monfort failed to obtain comparable information between the two groups.
Monfort also contends that because it reasonably could not have known at the time it received most former employee applications that it would not receive comparable information on new applicants, its actions would have been the same even if the former-employee applicants had not been Union members. According to the argument, because Monfort could not have foreseen the low return rate on the forms from former employees, it could not have intended to discriminate against former-employee applicants when it relied on their personnel files to disqualify them from consideration. Monfort contends that it has plainly shown that its actions would have been the same irrespective of past Union activity, and it should therefore be fully exonerated of liability. Monfort’s argument does not account for the substantial evidence of the prima facie case that Mon-fort’s own release form did not even request that former employers provide comparable information on new applicants. Moreover, even when a former employer would provide the information requested, Monfort applied its hiring criteria relating to absenteeism, discipline and medical condition more strictly to former-employee applicants than to new applicants. Thus, we find this argument to be without merit.
In addition to challenging the Board’s finding that Monfort failed to rebut the prima facie case, Monfort argues that the Board erred in finding discrimination against the entire class of former-employee applicants who were not hired based solely on what it terms “anecdotal evidence.” This argument completely ignores the critical statistic that Monfort did not have comparable information on at least 75% of new applicants. Given that the information could only disqualify an applicant, the pri-ma facie case of discrimination against former-employee applicants is well supported.
Finally, Monfort alleges legal error in the Board’s failure to consider Monfort’s evidence that fifty former-employee applicants took themselves out of consideration and therefore, can not be considered as part of the class discriminated against. Contrary to Monfort’s claim, the Board did not fail to consider this evidence, rather it merely found that
[1546]*1546in light of the overwhelming evidence of an overall pattern of discrimination in the hiring process and in the absence of any specific evidence concerning the reasons for disqualifying nonformer employee applications, we are not persuaded by other hearsay and subjectively interpreted reasons for not hiring alleged discri-minatees, such as the alleged receipt of information that an individual was no longer interested in employment.
298 NLRB No. 16 at 81. We recognize that the fifty former-employee applicants whom Monfort contends removed themselves from consideration were included in the class of former-employee applicants against whom the Board found discrimination. Nonetheless, after reviewing the evidence which Monfort contends was erroneously excluded, we find no error in the Board’s finding that such evidence was unpersuasive in rebutting the prima facie case.
II.
The Union challenges the Board’s remedy for Monfort’s unfair labor practices relating to the application of its hiring criteria.12 The remedy arguably limits reinstatement and backpay to former-employee applicants who would have been hired (or would have been hired earlier) if Monfort had applied its hiring criteria equally to all applicants. The Union contends that the Board’s remedy erroneously gives Monfort a second chance to apply its hiring criteria to former-employee applicants, and essentially permits Monfort to relitigate the second prong of the Wright Line test.13
In reviewing a remedy imposed by the Board for a violation of the National Labor Relations Act, we must give “special respect” to the Board’s choice. NLRB v. Gissel Packing Co., 395 U.S. 575, 612 n. 32, 89 S.Ct. 1918, 1939 n. 32, 23 L.Ed.2d 547 (1969). Indeed, the Board’s remedial power is “a broad, discretionary one, subject to limited judicial review.” Fibreboard Paper Products Corp. v. NLRB, 379 U.S. 203, 216, 85 S.Ct. 398, 406, 13 L.Ed.2d 233 (1964). See also Sure-Tan, Inc. v. NLRB, 467 U.S. 883, 898-99, 104 S.Ct. 2803, 2812, 81 L.Ed.2d 732 (1984). However, our deference to the Board’s remedial power ends at the point that the Board’s order “is a patent attempt to achieve ends other than those which can fairly be said to effectuate the policies of the Act.” Virginia Elec. & Power Co. v. NLRB, 319 U.S. 533, 540, 63 S.Ct. 1214, 1218, 87 L.Ed. 1568 (1943).
The Board clearly and appropriately intended to tailor the remedy to the unfair labor practice it was intended to redress. See Sure-Tan, 467 U.S. at 900,104 S.Ct. at [1547]*15472813. The discrimination found by the Board was based on Monfort’s application of the hiring criteria. Nevertheless, Mon-fort’s disparate application of its hiring criteria inevitably resulted in Monfort’s failure to hire the former-employee applicants against whom it discriminated.
Monfort attempted to rebut the pri-ma facie case by showing that it “would not have hired the alleged discriminatees who filed formal Monfort applications even in the absence of their union and other protected activities.” 298 NLRB No. 16 at 29. The Board rejected Monfort’s proffered evidence, finding that Monfort “failed to show that it would not have hired any of the alleged discriminatees who filed formal Monfort applications even in the absence of their union membership and activities.” Id. at 30. Monfort did not have to raise this issue, as it could have attempted to rebut the prima facie case with proof that it would have applied its hiring criteria disparately as between former-employee applicants and new applicants even in the absence of the former-employee applicants’ Union membership and activities. However, by choosing to litigate the issue of whether Monfort would not have hired the former employee applicants absent their Union membership at the proceeding on the merits, Monfort is precluded from relitigat-ing this issue at the compliance phase.
Additionally, we believe that permitting Monfort to relitigate at the compliance hearing whether it would have hired a particular former-employee applicant is simply unworkable in light of the unfair labor practices found in this case. Monfort failed to apply its objective hiring criteria to at least 75% of the new applicants. Thus, in order to treat former-employee applicants equally, Monfort should not be permitted to rely on its objective hiring criteria at the compliance hearing.14 Given that 43% of the new applicants were hired, Monfort’s hiring decisions relating to new applicants were largely subjective. To permit Monfort to assert defenses at the compliance hearing based on its subjective evaluation of former-employee applicants for whom the Board has already determined that Monfort harbored anti-Union animus would afford Monfort an additional and unwarranted opportunity to justify its unfair labor practices. See United Food & Commercial Workers, 768 F.2d at 1476. Based on the Board’s findings, this is an impossible burden for Monfort to meet, and the Board’s order invites additional unnecessary and protracted litigation.
Therefore, we remand to the Board with instructions to clarify the reinstatement and backpay remedy for former-employee applicants so that it is consistent with the Board’s finding relating to Monfort’s failure to rebut the prima facie case and so as to preclude Monfort from relitigating issues that it chose to litigate at the proceeding on the merits.
III.
Monfort contends that the Board erred in finding that Monfort terminated James Little because of his Union activity. An employer violates 29 U.S.C. §§ 158(a)(1), (a)(3) by discharging an employee for engaging in Union activity. See NLRB v. Wilhow Corp., 666 F.2d 1294, 1301-02 (10th Cir.1981); NLRB v. Dillon Stores, Inc., 643 F.2d 687, 692 (10th Cir.1981). After reviewing the record as a whole, we find that the Board’s determination that Little was terminated because of his activity on behalf of the Union is supported by substantial evidence and that Monfort’s justification for Little’s termination was pretextual.
IV.
Finally, Monfort contends that the Board erred in imposing an “extraordinary [1548]*1548remedy” for Monfort’s unfair labor practices relating to the Union election following the reopening of the plant. The Board found numerous unlawful acts committed by Monfort leading up to the Union election. See supra note 1. The Board ordered that a new election be held, and that Monfort notify all of the former employees who were discriminated against when the plant reopened, as well as all current employees by the mailing of a notice, publication in a local newspaper, posting the notice at the plant, and convening all current employees and reading the notice out loud. Additionally, the remedy requires Monfort to provide the Union with a list of all current employees; grant the Union access to the plant bulletin board and any other place notices are posted; grant the Union access to nonworking areas of the plant including canteens, cafeterias, rest areas, and parking lots; allow two Union representatives to be present whenever Monfort discusses unionization with a gathering of its employees and afford the Union equal time to respond at such gatherings; and grant the Union access to the plant and a facility to assemble the employees on paid work time to listen to a thirty minute speech delivered by a Union representative. 298 NLRB No. 16 at 45-47.
Notwithstanding Monfort's characterization of this remedy as “Orwellian,” we believe that it is well within the Board’s broad remedial power.15 Monfort’s contention that these remedies are unwarranted because no unfair labor practices against Monfort have been documented since 1983 are unwarranted is similarly without merit. “[T]he Board is not required to place the consequences of its own delay, even if inordinate, upon wronged employees to the benefit of wrongdoing employers.” NLRB v. J.H. Rutter-Rex Mfg. Co., 396 U.S. 258, 265, 90 S.Ct. 417, 421, 24 L.Ed.2d 405 (1969). We similarly are not persuaded by Monfort’s other factual contentions relating to subsequent elections at other Mon-fort plants or at the Greeley plant relating to different groups of employees even though these elections have withstood unfair labor practices charges.
Monfort’s petition for review of the order by the National Labor Relations Board is DENIED. The Union’s petition for review of the order by the National Labor Relations Board is GRANTED and the case is REMANDED to the National Labor Relations Board to clarify the remedy for Mon-fort’s unfair labor practices in applying its hiring criteria. General Counsel’s application for enforcement is GRANTED IN PART and DENIED IN PART as is consistent with this opinion.