This is a collective action suit to collect unpaid wages and overtime. Roy Maynor sued his former employer, the Dow Chemical Company, under the Fair Labor Standards Act ("FLSA"), 29 U.S.C. § 201, et seq. Maynor alleged that Dow failed to pay him and similarly situated employees for time spent training for and taking skills-assessment tests to comply with a requirement that the employees advance two levels per year in the areas tested. This court conditionally certified a class of current and former members of the International Union of Operating Engineers Local No. 564 employed at Dow's Freeport, Texas facility, hired before May 14, 2003, who had to undergo training and testing to comply with the skills-assessment section of the collective bargaining agreement that applied between October 9, 2004 and November 30, 2006. (Docket Entry No. 42). Maynor has been joined in this collective action by 129 or 130 other current or former Dow employees.

The following motions are pending:

• The plaintiffs have moved for partial summary judgment on their claim that the hours they spent participating in the skills-assessment program are compensable and that their acceptance of payments from Dow made after an investigation by the Department of Labor into Dow's failure to pay for this time did not waive the unpaid wage and overtime claims raised in this suit. (Docket Entry No. 188). Dow has responded. (Docket Entry No. 194). Dow agrees that accepting the partial payments it made after the Department of Labor investigation did not waive the plaintiffs' claims in this case. Dow vigorously disputes that the studying and training time is compensable. The plaintiffs have replied to Dow's response. (Docket Entry No. 198).

• Dow has moved for partial summary judgment on Maynor's individual claim for improper denial of overtime compensation for time spent studying to take the skills assessment tests and his individual claim that decision to fire him in February 2006 was in retaliation for his complaints about the failure to pay overtime. (Docket Entry No. 189). Maynor has responded, (Docket Entry No. 193), and Dow has replied, (Docket Entry No. 195).

• Dow has moved to decertify the class, (Docket Entry No. 192), arguing that the plaintiffs are not similarly situated because there was no common policy or plan of required training for the skills-assessment tests. Dow argues that the only common policy or plan was that the employees advance two skill levels per year. Dow also argues that there are individualized defenses to liability that preclude collective action treatment. The plaintiffs have responded, arguing that the skills-assessment program, including the training, is a common policy that gives rise to the FLSA violation; Dow's defenses to liability are common to the class; and that subclassing will adequately address the differences among the plaintiffs' claims. (Docket Entry No. 197).

Based on a careful review of the complaint; the motions, responses, and replies; the summary judgment record; and the applicable law, this court:

• grants the plaintiffs' motion for partial summary judgment that the acceptance of Dow's payment did not waive the claims asserted in this case;

• grants in part and denies in part the motion for a judgment that as a matter of law, all the time spent studying or training for the tests was compensable;

• denies Dow's motion for summary judgment as to Maynor's unpaid wages and overtime claim and his retaliatory discharge claim; and

• denies Dow's motion to decertify the class.

The reasons for these rulings are explained in detail below. A status conference is set for Monday, November 30, 2009 at 8:30 a.m. in Courtroom 11B.

I. Background

The parties have engaged in significant discovery that provides the basis for the factual background set out below. Dow hired Maynor in 1974 to work as an operator at the Freeport facility, which has over 5,000 employees. Maynor belonged to Local 564 of the International Union of Operating Engineers (the "Union"), which represents over 900 hourly nonexempt employees at the Freeport facility. (Docket Entry No. 188, Ex. A).

The summary judgment record includes: the July 30, 2009 Declaration of Charlie Singletary (the Union's Business Manager) and attached exhibits, (Docket Entry No. 188, Ex. A); portions of the oral deposition of Robin Campbell, Dow's Labor Relations Manager, (Id., Ex. B, Docket Entry Nos. 194, Ex. I, 189, Ex. B, 197, Ex. D, 192, Ex. A); the May 14, 2003 Collective Bargaining Agreement, (Docket Entry No. 188, Ex. C); the January 31, 2008 affidavit of Robin Campbell and attached exhibit, (Id., Ex. D); covers of study manuals obtained from Brazosport College, (Id., Ex. E); the July 31, 2009 declaration of plaintiffs' counsel Patricia Haylon and attached exhibit, (Id., Ex. F); portions of the oral depositions — including any attached exhibits — of plaintiffs Kenneth Nugent, (Id., Ex. G, Docket Entry Nos. 192, Ex. P, 194, Ex. D), Clifton Butler, (Docket Entry Nos. 188, Ex. H, 189 Ex. D, 192 Ex. E), Cuney Woodward, (Docket Entry Nos. 192, Ex. K, 194, Ex. A), Terry Benge, (Docket Entry Nos. 192, Ex. L, 194, Ex. B), Hipolito Nieto, (Docket Entry Nos. 192, Ex. N, 194, Ex. C), Roy Maynor, (Docket Entry Nos. 189, Ex. A, 192, Ex. O, 193, Ex. A, 194, Ex. E), Jesse Garza, (Docket Entry Nos. 192, Ex. J, 194, Ex. F), Anthony Moody, (Docket Entry Nos. 192, Ex. J, 194 Ex. G), Robert Roy, (Docket Entry Nos. 192, Ex. I, 194 Ex. H), Gary Phillips, (Docket Entry Nos. 189, Ex. C, 192, Ex. D), and Randell Joe Gilbert, (Docket Entry No. 192, Ex. F); the January 31, 2008 affidavit of Larry Kirby, the Safety Technical Coordinator at Brazosport College, (Id., Ex. G); the January 31, 2008 affidavit of Jamie Gay, a computing coordinator at Brazosport College (Id., Ex. M); the March 18, 2003 proposal for skills achievement, (Docket Entry No. 198, Ex. C); the August 17, 2009 declaration of Roy Maynor and attached exhibits (Docket Entry No. 193, Ex. B); a January 2006 e-mail chain involving Roy Maynor, his supervisor Cheryl Weinberger, and Robin Campbell (Id., Ex. F); the September 6, 2009 declaration of Patricia Haylon and attached exhibits, (Docket Entry No. 197, Ex. E); the August 14, 2009 declaration of Robin Campbell, (Docket Entry No. 192, Ex. C); and the August 14, 2009 declaration of Heather McConnell, a technology analyst at the law firm representing Dow, including supporting exhibits, ( Id., Ex. Q).

On May 14, 2003, a collective bargaining agreement ("CBA") between Dow and the Union became effective. Article XXXIV of the CBA established a "Skills Initiative Program." Under Section 1 of Article XXXIV, Union members hired before May 14, 2003 had to comply with the "Site Foundational Skills Program." Section 1 stated:

Foundational Skills Assessment will be required for employees hired prior to May 14, 2003. Employees required to obtain an associates degree will take the assessment upon completion of the degree. The Foundational Skills include the areas of Reading, Applied Mathematics, Locating Information, Teamwork, Applied Technology and Observation (the "Skill Category(s)"). In each Skill Category there are different Skill Levels, 3, 4, 5, 6 and 7 ("Skill Levels"). Each employee will be required to take a skills assessment to determine the level they are at, paid for by the COMPANY. Once the Skill Level has been determined, an employee will be expected to move up at least 2 Skill Levels per year until the employee has achieved the required level in all the Skill Categories. Employees must reach a Skill Level of 5 in Locating Information and Teamwork and a Skill Level of 6 in the other four categories.

(Docket Entry No. 197, Ex. B). Dow's Labor Relations Manager, Robin Campbell, testified that Dow intended to "upgrade and maintain . . . skills to increase competitiveness of [operators]" and to "insure that all [operators] were given opportunities and assistance in achieving the various skill levels." (Docket Entry No. 192, Ex. B, Campbell Aff., ¶ 5). New hires had to have at least an associate's degree. Present employees had to show that they had basic skills by complying with the Site Foundational Skills Program. Campbell testified that without the foundational skills the assessments measured, some employees "may not have the skill and ability to understand the training" in technical areas that would be important in many plant jobs. (Id., Ex. A, Campbell Depo. at 40-41).

A. The Site Foundational Skills Program

Under the Site Foundational Skills Program ("FSP") set out in Article XXXIV, Section 1 of the CBA, an employee had to take an initial assessment in each of six skill areas — reading, applied mathematics, locating information, teamwork, applied technology, and observation. There were five to seven skill levels within each of these six areas. Once each operator's skill level was determined by this mandatory initial assessment test, the operators had to take reassessment tests and advance two skill levels per year until the operator met the level required for each area. The assessment tests were standardized tests administered by ACT, a national testing service, at Brazosport College, a nearby school.

The FSP was implemented in May 2003. Operators had until December 31, 2004 to advance the initial two skill levels. If an employee failed to advance, disciplinary action could be imposed. An employee who demonstrated that he was working to achieve the skills needed to perform successfully on the assessment tests but could not do it within the time limits could get additional time and chances to take the tests.

Dow did not require that employees prepare for the assessment tests in any specific way or that they prepare at all. The only requirement was to advance two skill levels per year on the test results until a certain level for each of the six areas was reached. Dow took the position when the CBA was negotiated that the time spent on studying or preparing for the assessment tests was not compensable. Dow did provide options for studying and training for employees who chose to do so. Those options were provided through Brazosport College. The options included accessing online training programs, attending classes covering the materials in the assessment tests, attending one-on-one tutoring, visiting computer labs, and obtaining materials, including workbooks and CDs, for self-study. Dow paid ACT and Brazosport College for the study materials, the online training portal, and the classroom programs, which were all offered to covered Dow employees through Brazosport College. Although Dow did not require any studying or studying in a particular way or amount, Brazosport College only allowed employees to take reassessment tests every 90 days unless they had engaged in some form of training. If an employee failed to advance a skill level after taking the assessment test twice, the College required the employee to meet with a tutor before taking the test a third time. (Docket Entry No. 192, Ex. B, Campbell Aff., ¶ 14, Ex. G, Kirby Aff., ¶¶ 7-8).

Dow required employees subject to the Foundational Skills Assessment requirement to sign releases that would allow Dow to get their test results from Brazosport College. (Id., ¶ 12). Failure to execute a release was an offense subject to discipline. (Id., ¶ 18). The releases also allowed Dow to get information on the hours an employee spent studying if that employee did so by accessing online materials through Brazosport College or attended classroom sessions, one-on-one tutoring, or went to computer labs at the College. The College reported those hours to Dow. Each month, Dow received a spreadsheet from the Brazosport College identifying the assessment tests taken and the results, as well as the number of hours the employees spent in training through the College. (Id., Ex. A, Campbell Depo. at 67-71). If, however, an employee used self-study to prepare for the assessments, the only information Dow had about those hours was from the employee reporting them to Dow. The record does not show that employees were required to report their self-study hours, and at least some employees did not do so.

Some employees were able to meet the two-skill-level-per-year advancement requirement without studying or training at all. (Id., Ex. C, Campbell Decl. ¶ 6). Most employees, however, prepared through the College classes, tutoring, computer labs, or online materials. Some employees used self-study. (See, e.g., id., Ex. E, Butler Depo. at 45-47, Ex. F, Gilbert Depo. at 49-50, Ex. Q, McConnell Decl, Ex. 2). As of the end of 2004, employees who had satisfied the FSP requirement had spent an average of 18.5 hours of training time to advance one skill level. ( Id., ¶¶ 16-17). Because the record does not show that employees were required to report their self-study hours, it appears that this 18.5-hour number was based only on hours tracked by Brazosport College and reported to Dow on the spreadsheets.

B. The February 2005 Agreement

In the fall of 2004, the Union asked Dow for an amendment to the Site Foundational Skills program to allow employees more time to advance two skill levels before the December 31, 2004 deadline. While over 500 covered employees had satisfied the requirement, 370 had not. In February 2005, Dow and the Union entered into an Agreement addressing the appropriate discipline for employees who failed to achieve the required skill-level advancement. (Docket Entry No. 188, Ex. D, Ex.1-B at 1). The Agreement, entitled "Skills Initiative — Addressing Non-Compliances," was "created as a guideline for site consistency in addressing non-compliances as it relates to the contractual requirements on Foundational I II skills initiatives." The Agreement provided "examples of different situations and the guidelines on how to address them" grouped into three "Tiers." (Id. at 2). "Tier I" scenarios, including failing to sign the privacy release form, signing the release form but not taking any Assessments, and signing the release form but not having any activity after that, were to be addressed through a written warning or through progressive discipline "[i]f discipline above the level exists." (Id.). Signing the release form, taking the Foundational Skills Assessments, and enrolling in training, but failing to advance two skill levels per year, was classified as a "Tier II" scenario if the employee had less than forty hours of training and as a "Tier III" scenario if the employee had forty or more hours of training. A Tier II scenario warranted progressive discipline up to time off without pay or a more severe penalty (but not termination) if the employee's last disciplinary action was time off without pay. The Tier II approach was to allow employees who had made "reasonable efforts" to improve the assessment test results — by taking at least 40 hours of class training or online training reported to Dow — a chance to avoid job termination despite unsatisfactory test results. A Tier III scenario would lead to an employee receiving a "letter of expectations" establishing a time frame for completing the required skill levels. (Id.).

C. Discipline For Failing to Comply with the Foundational Skills Program; Maynor's Job Termination

Beginning in 2005, covered employees received disciplinary warnings for failing to comply with the program. (Docket Entry No. 188, Ex. A, Singletary Decl. at 2). Approximately four covered employees were suspended, which led to the loss of a performance award for 2005. (Id.). By September 2006, over 40 covered employees had been disciplined. (Id.).

Maynor refused to sign the privacy release form, a basic requirement of the FSP. On February 3, 2005, Dow sent Maynor a letter stating that he had failed to meet the Site Foundational Skills program annual requirement; he would have until June 30, 2005 to satisfy the two-level requirement for the prior year; and that he would have until December 31 to satisfy the two-level requirement for 2005. (Docket Entry No. 189, Ex. A, Maynor Depo., Ex. 4, Ex. A-2). Under the February 2005 Agreement between Dow and the Union, an employee's refusal to sign the release was a Tier I scenario. An initial written warning would issue and continued failure would be addressed under Dow's progressive discipline policy. (Docket Entry No. 192, Ex. B, Campbell Aff., ¶ 18). On August 10, 2005, after warnings, Dow sent Maynor a letter suspending him without pay for one day for failing to meet the requirements. (Docket Entry No. 189, Ex. A, Maynor Depo., Ex. 4, Ex. A-3). Maynor was the only employee out of more than 700 covered by the program who refused to sign the release. (Docket Entry No. 189, Ex. B, Campbell Depo. at 63, 148).

In early January 2005, Maynor gave his supervisor, Cheryl Weinberger and Dow's Labor Relations Manager, Robin Campbell, a packet from Brazosport College containing a paper copy of his training hours report. The record shows that Campbell and Weinberger both knew that the Brazosport College transcript was included in the packet. Weinberger told Maynor in an e-mail that Campbell would be meeting with the legal department and that "[a]t that time, they will consider your request to accept the paper-copies of your college information," until which time she would "keep the packet in my office." (Docket Entry No. 193, Ex. F). The following month, in February 2006, Dow terminated Maynor's employment. The stated reason was his failure to sign the privacy release. Dow took the position that as a result, it could not verify whether Maynor had engaged in training under the FSP or performed satisfactorily on the assessment tests. (Docket Entry No. 192, Ex. B, Campbell Depo. at 115). Maynor filed three grievances in connection with the discipline and the job termination. The termination was upheld in a grievance hearing. (Docket Entry No. 193, Ex. B, Maynor Decl. at 2).

D. The DOL Investigation and the November 2006 Contract Modification

On September 6, 2006, after approximately 40 covered employees had been disciplined in connection with the FSP, the Union filed a complaint with the DOL. The Union alleged that Dow had violated the FLSA by failing to compensate employees for time spent training for and taking the Foundational Skills Assessments. (Docket Entry No. 188, Ex. A-1 at 19).

On November 30, 2006, Dow and the Union entered into a Mid-Term Agreement modifying Article XXXIV of the CBA to eliminate the two-skill-level-advancement-per-year requirement. The Mid-Term Agreement eliminated the clause stating that "[f]ailure of the employee to attempt to reach these requirements could result in corrective disciplinary action, however employees clearly working toward achieving the skills but simply needing additional time in achieving required levels may be given additional time and opportunities to succeed," and the clause stating that "[o]nce the Skill Level has been determined, an employee will be expected to move up at least 2 Skill Levels per year until the employee has achieved the required level in all the Skill Categories." (Docket Entry No. 32, Ex. 1-E). Although Dow retained the Foundational Skills Assessment Program, it was revised to provide that the only consequence of failing to complete the requirements was the inability to bid into a new job. (Docket Entry No. 192, Ex. B, Campbell Aff, ¶ 23). No discipline would result from any failure to comply with the program. (Id.). Covered employees were also compensated for the time spent taking reassessment tests. (Id.).

The DOL investigated the Union's complaint and concluded that because the training and assessments were "mandatory," Dow was required to compensate employees for the time spent in the FSP. Dow provided the DOL a spreadsheet showing the number of hours each operator had spent on training for the assessments, according to the Brazosport College records. (Docket Entry No. 188, Ex. B, Campbell Depo. at 144). The number of hours and the hourly rates on the spreadsheet led to a total of $2.67 million.

Dow disputed the DOL's conclusions. Without admitting liability, Dow agreed to pay covered employees who participated in the FSP for up to 40 hours per year of training from mid-2005 through November 30, 2006. If an employee had spent less than 40 hours of training during that period, Dow paid for the time actually spent in training. Dow also paid for the time spent taking the skills assessments. Dow agreed to expunge disciplinary records related to the FSP. (Docket Entry No. 192, Ex. B, Campbell Aff., ¶¶ 21-22).

On January 14, 2008, the DOL sent Dow a letter stating that under this approach, Dow had underpaid $640,033.28 to 637 employees between July 1, 2005 and June 20, 2007. The letter confirmed that Dow had agreed to pay these amounts. (Docket Entry No. 32, Ex. 4). The payments were made in August and November 2007. (Docket Entry No. 192, Ex. B, Campbell Aff., ¶ 21). On November 9, 2007, Dow sent employees a letter explaining that they would be receiving the payments. The letter stated:

You will be receiving a payment from The Dow Chemical Company ("Dow") and we want to inform you of the purpose of the payment and give you some background on how the payment came to be made.

. . . .

U.S. Department of Labor ("DOL") recently completed a review of how Dow administered certain skills initiative training in the past. As a result of this review, Dow and the DOL have reached an agreement addressing the question of compensability of time that some employees spent on this training.

The skills initiative practice which was in question had been negotiated and agreed to by the Union and the Company in 2003. However, the DOL took the position that some of the employees' time in the training program was compensable under federal law. After numerous discussions, Dow has voluntarily reached an agreement with the DOL on this issue. While our company does not necessarily agree with the position taken by the DOL, we felt that it was in the best interests of all parties involved to reach an agreement as quickly as possible, in order to bring this matter to resolution.

As of November 30, 2006, ratification of the mid-term agreement, there are no 2 level requirements or potential discipline related to the skills initiative. This means that there are no ongoing skills initiative activities that would be compensable under the DOL's position. Employees will continue to be paid for certain assessment activities.

The payment you will receive is based on an assessment of the time that you spent on the compensable activities. Any questions related to this should be directed to Labor Relations.

(Docket Entry No. 33, Ex. A). Dow did not make payments to Maynor or expunge his disciplinary records because he had not signed a release and because he was no longer a Dow employee. (Docket Entry No. 192, Ex. B, Campbell Aff., ¶¶ 21-22). The DOL investigator told the Union that operators would not waive their rights to pursue future claims by accepting Dow's partial payment. (Docket Entry No. 188, Ex. A, Singletary Decl. at 2).

E. This Suit and the Conditional Collective Action Certification

Maynor filed this lawsuit on October 9, 2007. (Docket Entry No. 1). On May 28, 2008, this court conditionally certified a class made up of the following:

Current and former members of the International Union of Operating Engineers Local No. 564 employed by Dow at the Dow facility in Freeport, Texas and hired before May 14, 2003 who were required to undergo training and testing in order to comply with the Site Foundational Skills section of the collective bargaining agreement between Dow and the Union between October 9, 2004 and November 30, 2006.

(Docket Entry No. 42 at 26). When the class was conditionally certified, sixteen current or former covered employees had filed consents to join. The notice sent to putative class members included the following statement:

Again, Dow denies any liability in this case. The lawsuit is in its early stages and no right to recovery has been determined by the Court. If you choose to join the lawsuit, your continued right to participate in the lawsuit may depend on a later decision by the court that you and the Plaintiffs are "similarly situated" employees in accordance with federal law.

If you choose to join the lawsuit, you will be bound by the judgment of the Court on all issues of the case. While the lawsuit is proceeding, you may be required to respond to written questions, produce documents, give a deposition and/or testify in Court. If you choose not to join the lawsuit, you will not be affected by the decision rendered, whether favorable or unfavorable.

(Docket Entry No. 43 at 3). Since the notice issued, many more plaintiffs have filed consents and some have voluntarily terminated their claims. At present, Maynor has been joined by either 129 or 130 opt-in plaintiffs.

There is a conflict between the docket sheet and the number reported by the plaintiffs in their response to Dow's motion to decertify. The plaintiffs report that 129 employees have joined Maynor's suit. (Docket Entry No. 197 at 1). On the docket sheet, there are 130 nonterminated plaintiffs aside from Maynor. It appears that the disparity might be the result of Docket Entry No. 97. The docket entry describes that document as a consent by Jesse Garza and Rick Morphew. Only Morphew's consent form is part of the document itself. Garza's consent form does not appear to have been filed. There are also more consent forms in the record than there are plaintiffs. Docket Entries 46 and 50 are both consent forms for Tom N. Balridge. One plaintiff, Cornell Porter is now proceeding pro se. (Docket Entry No. 200).

In the conditional certification order, this court clarified that the two-step Lusardi approach would govern this collective action. (Docket Entry No. 42 at 11-12). Under that framework, after certification, discovery proceeded on a collective basis. Twelve depositions have been taken and both parties have conducted written discovery.

Discovery has revealed details about how the FSP and the reporting of hours spent in preparation and training for the assessments worked in practice. One training option available to covered employees was to access online programs through the ACT servers and the Brazosport College. Each covered employee received a user ID and password to access the online training materials remotely, including from a laptop computer at the employee's home or work. ACT tracked and recorded how long each employee was logged onto the system and what training materials the employee accessed while online. Brazosport College obtained its data about the employees' online training from the ACT servers. Brazosport College submitted that data to Dow. (Docket Entry No. 192, Ex. A, Campbell Depo. at 69-71). Although Dow apparently did not realize it when the FSP began, the company at some point discovered that the only information ACT obtained about how long a covered employee spent in online training was how long an employee remained logged on to the computer, as opposed to how long the employee actually engaged in online training. Once a Dow employee had logged on, the ACT server would record how much time elapsed before the employee logged off. (Docket Entry No. 192, Ex. M, Gay Decl., ¶¶ 3-5). One plaintiff, Gary Phillips, testified that, on occasion, he left the online training portal open while not actually engaged in training activity. He testified that the longest period he spent training after logging on was 90 minutes, but his training records recorded him as engaged in online training for as long as 11 straight hours. (Id., Ex. D, Phillips Depo. at 28-29). Another plaintiff, Clifton Butler, would log onto the computer for online training and simply stay logged on for up to 30 minutes at a sitting. (Id., Ex. E, Butler Depo. at 41-42). Hipolito Nieto would log on at work and stay logged on for his entire eight-hour shift, spending only a few minutes each day actually looking at the online materials. (Id., Ex. N, Nieto Depo. at 26-27). According to Larry Kirby, the Safety Technical Coordinator for Brazosport College, some Dow employees logged onto online training and stayed logged on for extended periods, some so extended that the system recorded over twenty-four hours of training in a single day. (Id., Ex. G, Kirby Decl., ¶ 9). Maynor's training records include days in which he spent 16.31, 16.32, 20.39, 20.95, 45.47, and 53.98 hours logged onto the online training system. (Id., Ex. O, Maynor Depo., Ex. 4, Ex. A-4). Maynor testified that each of these entries was incorrect as a record of how much time he actually spent training. (Id., Ex. O at 150-54).

The record does not reveal how the computer would record more than twenty-four hours of online training time during a single day, even if the covered employee remained logged on during that entire day.

Not every employee who did online training stayed logged on for extended periods with only intermittent or no participation in training. Cluney Woodward testified that he never got up and walked away from an online training session after logging on and that he would participate in training for as many as seven consecutive hours in a session. (Id., Ex. K, Woodward Depo. at 52).

Some employees engaged in online and other study methods during their work shifts at Dow. Nieto was not the only employee who logged online training hours while at work. Phillips testified that all the time he spent in online training was while he was on the clock at Dow during his regular shift. Woodward testified that only around five percent of his online training hours occurred while working at Dow; the rest were when he was off duty. (Id., Ex. K, Woodward Depo. at 32-33).

Some employees testified that they performed self-study away from work. Phillips also reported fifty hours of self-study, which he spent looking at printed materials provided through Brazosport College. Of those fifty hours, forty were while Phillips was on the clock at Dow on his regular shift. (Id. at 29-32). Other employees only engaged in self-study away from their work at Dow, during their off-duty hours. (Id., Ex. E, Butler Depo. at 46-47). Most of this self-study was similar to what Phillips reported he did; reading manuals and other printed material supplied by ACT and the College. There were, however, some variations in the types of self-study material used. There were also variations in whether the employee studied while engaging in other activities. Maynor testified that a large portion of his self-study hours involved reading the printed materials obtained from the Brazosport College while he was driving his car to and from work. (Id., Ex. O, Maynor Depo. at 25-26). And some employees self-studied by reading materials that he routinely read apart from the FSP. Anthony Moody testified that between 100 and 150 of his 864 self-study hours was spent reading newspaper articles to test his ability to retain information. Moody regularly read newspapers before Dow began the skills initiative. (Id., Ex H., Moody Depo. at 37, 42-43).

The covers of the training manuals are part of Exhibit C to the plaintiff's response to Dow's decertification motion. (See Docket Entry No. 197, Ex. C).

The plaintiffs moved for a summary judgment ruling that their hours of participating in the FSP by studying for and taking the assessment tests from October 2004 to November 2006 were compensable. The plaintiffs argue that the undisputed facts show that the studying and training was directly related to their jobs and that their participation was not voluntary. As a result, the plaintiffs argue, the program did not meet all the criteria necessary to exclude the hours from compensable work time under 29 C.F.R. §§ 785.27-785.29. (Docket Entry No. 188). Dow responded by arguing that there are genuine issues of disputed fact material to determining whether the assessment and training time was directly related to the plaintiffs' jobs and whether the training was voluntary. (Docket Entry No. 194). The plaintiffs replied. (Docket Entry No. 198).

Dow also moved for partial summary judgment dismissing Maynor's claims. (Docket Entry No. 189, 191). The plaintiffs responded, (Docket Entry No. 193), and Dow replied, (Docket Entry No. 195).

Finally, Dow moved to decertify the class. (Docket Entry No. 192). Dow argues, in part, that discovery revealed individualized defenses to the FLSA claims. Many employees are claiming compensation for self-study training hours. Brazosport College did not report self-study hours to Dow. Instead, Brazosport College reported to Dow only the hours logged online and spent in classroom studying, in tutoring sessions. As a result, Dow asserts that it did not and should not have known about most or all of the self-study hours claimed. Dow argues that there will have to be individualized inquiries to determine what each plaintiff communicated to Dow about his self-study hours. (Docket Entry No. 192 at 15). Dow also argues that the flaw in the system for tracking online hours spent in training (as opposed to merely being logged on) means that individualized inquiries will be required to determine how many hours the plaintiffs who used online training actually spent in training. (Id. at 16-17). In response, the plaintiffs emphasize that Dow had reliable information about the number of hours the plaintiffs spent in classroom study, computer labs, and tutoring. (Docket Entry No. 197). These hours were reported to Dow by Brazosport College and are not challenged. The plaintiffs argue that the issues Dow raises relate to only a portion of the hours claimed and involve two questions: whether the hours all the plaintiffs claimed for online study were accurate; and whether Dow knew or should have known about most or all of the hours 85 of the plaintiffs claimed for self-study. The plaintiffs argue that these issues go to damages, not liability, and cannot defeat otherwise appropriate collective action treatment; that determining how many online hours were spent in actual training can be resolved by representative testimony or by bifurcating liability and damages; and that determining the number of hours spent by the 85 plaintiffs who are claiming compensable time for self-study can be managed by subclassing and, if necessary, bifurcation.

Each of these motions is considered below.

II. The Plaintiffs' Motion for Summary Judgment

A. The Applicable Law 1. Summary Judgment

Summary judgment is appropriate if no genuine issue of material fact exists and the moving party is entitled to judgment as a matter of law. FED. R. CIV. P. 56(c). "The movant bears the burden of identifying those portions of the record it believes demonstrate the absence of a genuine issue of material fact." Triple Tee Golf, Inc. v. Nike, Inc., 485 F.3d 253, 261 (5th Cir. 2007) (citing Celotex Corp. v. Catrett, 477 U.S. 317, 322-25, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986)).

If the burden of proof at trial lies with the nonmoving party, the movant may satisfy its initial burden by "'showing' — that is, pointing out to the district court — that there is an absence of evidence to support the nonmoving party's case." See Celotex, 477 U.S. at 325,106 S.Ct. 2548. While the party moving for summary judgment must demonstrate the absence of a genuine issue of material fact, it does not need to negate the elements of the nonmovant's case. Boudreaux v. Swift Transp. Co., 402 F.3d 536, 540 (5th Cir. 2005) (citation omitted). "A fact is 'material' if its resolution in favor of one party might affect the outcome of the lawsuit under governing law." Sossamon v. Lone Star State of Texas, 560 F.3d 316, 326 (5th Cir. 2009) (quotation omitted). "If the moving party fails to meet [its] initial burden, the motion [for summary judgment] must be denied, regardless of the nonmovant's response." United States v. $92,203.00 in U.S. Currency, 537 F.3d 504, 507 (5th Cir. 2008) (quoting Little v. Liquid Air Corp., 37 F.3d 1069, 1075 (5th Cir. 1994) (en banc)).

When the moving party has met its Rule 56(c) burden, the nonmoving party cannot survive a summary judgment motion by resting on the mere allegations of its pleadings. The nonmovant must identify specific evidence in the record and articulate how that evidence supports that party's claim. Baranowski v. Hart, 486 F.3d 112, 119 (5th Cir. 2007). "This burden will not be satisfied by 'some metaphysical doubt as to the material facts, by conclusory allegations, by unsubstantiated assertions, or by only a scintilla of evidence.'" Boudreaux, 402 F.3d at 540 (quoting Little, 37 F.3d at 1075). In deciding a summary judgment motion, the court draws all reasonable inferences in the light most favorable to the nonmoving party. Connors v. Graves, 538 F.3d 373, 376 (5th Cir. 2008).

The moving party bears a heavier burden when seeking summary judgment on a claim or defense on which it would bear the burden of proof at trial. Fontenot v. Upjohn Co., 780 F.2d 1190, 1194 (5th Cir. 1986). "Thus, if the movant bears the burden of proof on an issue, either because he is the plaintiff or as a defendant he is asserting an affirmative defense, he must establish beyond peradventure all of the essential elements of the claim or defense to warrant judgment in his favor." Id.; see also Meecorp Capital Markets LLC v. Tex-Wave Industries LP, 265 Fed. App'x 155, 157 (5th Cir. 2008) (per curiam) (unpublished) (quoting Fontenot). 2. Compensation for Training Hours

Two statutes must be analyzed to decide whether the undisputed facts establish that as a matter of law, the time they spent studying for and taking the skills assessments was compensable: the FLSA, which requires the employer to pay employees for time that the employer controls and requires, 29 U.S.C. § 207; and the Portal-to-Portal Act, which limits that obligation by excluding from compensable time hours spent on activities that are preliminary or "postliminary" to an employee's principal work activity. 29 U.S.C. § 254(a)(2). An employer must compensate an employee for time spent on all "principal activities" performed during the work day, including time for activities before or after the regular work shift if those activities are an "integral and indispensable part" of the employee's principal activities for which the employee is employed. The Portal-to-Portal Act exempts employers from liability under the FLSA for failing to compensate for time spent on "activities which are preliminary to or postliminary to said principal activity or activities." Id. In Dunlop v. City Electric, Inc., 527 F.2d 394 (5th Cir. 1976), the Fifth Circuit held that time spent on activities performed by electricians before their 8:00 a.m. start time, such as filling out daily time, material, and requisition sheets, checking job locations for the day, and preparing for that day's job, were compensable as "principal activities." Id. at 401. Dunlop defines principal activities as those "performed as part of the regular work of the employees in the ordinary course of business" and "at the employer's behest and for the benefit of the business." Id. "The only activities excluded from FLSA coverage are those undertaken 'for [the employees'] own convenience, not being required by the employer and not being necessary for the performance of their duties for the employer.'" Barrentine v. Arkansas-Best Freight Sys., Inc., 750 F.2d 47, 50 (8th Cir. 1984) (quoting Dunlop, 527 F.2d at 398).

Under the FLSA, time spent on job-related training activities are generally compensable. Moreau v. Klevenhagen, 956 F.2d 516, 521 (5th Cir. 1992). Regulations issued by the Wage and Hour Division of the Department of Labor provide the following criteria to determine whether time spent on such activities as attending lectures or training programs is compensable:

Attendance at lectures, meetings, training programs and similar activities need not be counted as working time if the following four criteria are met:

(a) Attendance is outside of the employee's regular working hours;

(b) Attendance is in fact voluntary;

(c) The course, lecture, or meeting is not directly related to the employee's job; and

(d) The employee does not perform any productive work during such attendance.

29 C.F.R. § 785.27. Time spent on training and similar activities must satisfy all four criteria to be exempt from treatment as compensable time.

Dow argues that the Labor Department's regulations improperly expand the limitations on employer liability in the Portal-to-Portal Act by requiring that the training be voluntary. This is not correct. The FLSA and the Portal-to-Portal Act are two separate statutes. The regulations interpret the FLSA only. Even if, for example, the Portal-to-Portal Act did not help Dow because the training was not preliminary or postliminary to employment but was instead indispensable to the plaintiffs' jobs, Dow might still avoid paying overtime if it could show that the four FLSA requirements set forth in the regulation were satisfied.

B. Analysis

The primary arguments raised in the plaintiffs' motion for partial summary judgment are whether, as a matter of law, taking the assessments and preparing for them was "voluntary" and directly related to the employees' jobs.

1. Voluntary

The regulations explain that attendance at events such as classes, lectures, or other education or training sessions is not "voluntary" if it is required by the employer or if "the employee is given to understand or led to believe that his present working conditions or the continuance of his employment would be adversely affected by nonattendance." 29 C.F.R. § 785.28. Dow does not argue that the assessments themselves were voluntary. Dow admits that it required employees to take the assessments. Dow does, however, argue that there are genuine issues of fact material to determining whether the time spent in training and studying for the assessments was voluntary. Dow points out that it required no studying or training. Instead, Dow required a certain level of performance on the assessments — advancing two skill levels per year. An employee who could do so without training at all would comply with Dow policy. Dow cites cases such as Dade County v. Alvarez, 124 F.3d 1380, 1384-85 (11th Cir. 1997), to support its argument that requiring a specific level of performance is different from requiring training that might be necessary or useful to achieve this level. Dow argues that even though it required a specific level of performance on the assessments, which many employees could not achieve without studying, the studying was not required, was "voluntary," and was not compensable.

In Dade County, the court found that off-duty training by police officers seeking to pass a mandatory fitness test was voluntary and not compensable under the FLSA. The officers were not "required to spend a specific amount of time training or to perform certain exercise routines during their off-duty hours." Id. at 1385. There was no evidence that the officers' "employment would be adversely affected if they did not participate in any particular off-duty activities, as long as they could pass the fitness tests." Id. "Given the freedom the officers enjoyed in selecting their off-duty activities," the panel concluded that "the actual off-duty physical activity performed by individual officers was voluntary within the meaning of the regulations." Id. Similarly, in Bull v. United States, 68 Fed. Cl. 212, 256-60 (Fed. Cl. 2005), the court held that off-duty training by customs agents to pass a marksmanship test was voluntary because there was no "directive or requirement to act" and no evidence that employment would have been adversely affected by failing to train. In a similar case, Jackson v. City of San Antonio, 2006 WL 2548545, at *14 (W.D. Tex. Aug 31. 2006), the court reached the same result. It held that off-duty weapons training by police officers seeking to pass a required weapons test was voluntary. Id. The plaintiffs in Jackson did not allege that they would be unable to pass the test without off-duty practice. Id. Similarly, in Price v. Tampa Elec. Co., 806 F.2d 1551 (11th Cir. 1987), the court held that off-duty training offered to industrial product testers was voluntary even though the training was a prerequisite to a salary increase.

As Dow has noted, the plaintiffs in this case consistently testified in their depositions that no one at Dow told them that they were required to train at all, to spend a certain number of hours training, or to engage in a particular type of training. Like the officers in Dade County and Bull, the plaintiffs were required only to pass the assessment tests at a certain level of proficiency. In Dade County and Bull, the fact that passing would have been difficult without training did not make the training involuntary. And in Dade County and Bull, the fact that the employer made certain training options available and paid for them mean that the training was involuntary. See Bull, 68 Fed. Cl. at 257 n. 47 (finding off-duty weapons training voluntary even though employer provided ammunition for that purpose); Jackson, 2006 WL 2548545, at *14 (same); Price, 806 F.2d at 1552 (finding training courses set up by employer voluntary).

The plaintiffs argue that this case is distinguishable from Dade County, Bull, and similar cases. The plaintiffs point to evidence showing that Dow tracked the training hours, receiving spreadsheets from Brazosport College reporting the type and amount of each employee's training. The plaintiffs also argue that in the CBA, they were warned that "[f]ailure of the employee to attempt to reach" the two-skill-level-per-year advancement requirement "could result in corrective disciplinary action," (Docket Entry No. 188, Ex. C at 2). The plaintiffs cite the amended Agreement of February 2005 setting discipline for an employee who failed to advance two skill levels per year. The discipline was based on the amount of training time the employee had. An employee who did not pass two skill levels per year and who had fewer than 40 hours of combined classroom and online time per year would receive discipline that could progress up to and including suspension without pay. An employee who did not pass the required two skill levels per year but had more than forty hours of training time would receive a "letter of expectation" setting a deadline to complete the past year's requirements and the actions that would be taken if this deadline was not met. (Id., Ex.1-B at 3). In short, the plaintiffs argue, Dow set 40 hours of training as a requirement to avoid discipline if the employee failed to meet the performance level. An employee who took 40 hours of training was making a "reasonable effort" to advance two skill levels per year and would not be subject to progressive discipline. (Docket Entry No. 192, Ex. B, Campbell Aff., ¶¶ 17-18).

Dow characterizes the 40 training hours not as a requirement but as an "affirmative defense to discipline if an employee did not meet the requirement of the FSP to advance two skill levels on the assessments per year. (Docket Entry No. 194 at 17). According to Robin Campbell, the February 2005 Agreement came about because approximately 370 employees failed to advance the required two skill levels by December 31, 2004. (Docket Entry No. 192, Ex. B, Campbell Aff. ¶ 17). Campbell testified in his deposition:

So it wasn't Dow's position that we had changed anything, as far as requiring. What we were trying to do is support the Union's concern that we've actually got people who are out there not blowing this off or not paying attention, that we have a signed CBA, but they just cannot get the two levels. Can we look at other ideas on things to work at? And that's how this came to life.

(Id., Ex. A, Campbell Depo. at 78).

The record does not support Dow's position. The record shows that by September 2006, approximately 40 operators had been disciplined under the CBA and the February 2005 Agreement. (Docket Entry No. 188, Ex. A, Singletary Decl. at 2, Ex. A-1 at 61-117). The record does not show how many employees were subject to discipline under Tier II as opposed to Tiers I or III. Dow argues that the February 2005 Agreement did not expose a covered employee to discipline if that employee participated in the program, advanced two skill levels on the assessments, but recorded at least 40 hours of training time. Such an employee would fall under Tier III. Such an employee would, however, receive a letter of expectation stating that disciplinary action could result if the stated expectation was not achieved within the deadline. Many of the letters of expectation required additional training. (Id., Ex. A-1 at 61-117). The February 2005 Agreement did expose an employee to progressive discipline — up to suspension without pay — for failing to advance two skill levels per year if that employee had fewer than 40 hours of training. (Docket Entry No. 192, Campbell Depo., Ex. 5). In Dade County, Bull, and Jackson, by contrast, discipline was not tied to the number of training hours completed; it was wholly based on test results. See Dade County, 124 F.3d at 1384-85; Bull, 68 Fed. Cl. at 256-60; Jackson, 2006 WL 2548545, at *14.

The DOL Wage and Hour Opinion Letters the plaintiffs cite "constitute a body of experience and informed judgment" but are not binding. Skidmore v. Swift Co., 323 U.S. 134, 140 (1944); Dade County, 124 F.3d. at 1385 (quoting Skidmore). The problem with the Opinion Letters on point is that they conflict with each other. The September 9, 1970, September 27, 1984, and January 15, 2009 letters cited by the plaintiffs all conclude that, under the facts assumed in those letters, training is compensable. The 1970 letter concluded that time spent by telephone operators practicing their typing would not be voluntary "if her attendance [at classes] is required for the continuance of her employment and if such typing practice at home is necessary to" achieve the requisite skills. (Docket Entry No. 188-16). The 1984 letter concluded that time spent by nurses improving their skills for a nursing skills test was not voluntary because failure to perform well on the tests would put a nurse's employment in jeopardy. (Docket Entry No. 188-15). The 2009 letter cited the 1970 letter in finding that time spent doing homework assigned at employer-required training courses was not voluntary. (Docket Entry No. 198-6). On the other hand, the Dade County court cited a June 1, 1994 letter, 1994 WL 1004833, and a September 12, 1985 letter, 1985 WL 1087360, both of which found that off-duty physical training by police officers required to pass fitness tests was voluntary.

Cases such as Dade County and Bull are also distinguishable because they involved employment requirements that were already in place when the plaintiffs began their jobs. As the Dade County court explained: "The sole limitation placed upon the officers by the County was that they maintain the level of physical conditioning required by the job of SRT officer . . . all of whom were active and physically fit when they applied." Dade County, 124 F.3d at 1385-86. The employees could not begin the job without first attaining a certain level of fitness and the employees knew when they began work that they had to maintain that level to remain employed. Id. Similarly, the customs officers in Bull were required to undergo weapons training and attain the minimum marksmanship standards before beginning work. The subsequent tests were in place to ensure that the officers maintained that level of competence and the officers knew of this requirement when they began employment. Bull, 68 Fed. Cl. at 258. The officers in Jackson were also required to satisfy marksmanship standards before beginning work. Jackson, 2006 WL 2548545, at *12.

Here, by contrast, the issue is the application of the new requirements Dow applied to existing employees. To upgrade the skills of the work force, Dow required that all new hires have an associate's degree. The existing employees, hired before any such requirement was in place, had to participate in the FSP until they demonstrated that they had equivalent basic skills. The plaintiffs did not need those skills or an associate's degree to qualify for an operator job when they began work. Unlike the fitness or marksmanship testing for police officers, the FSP did not require covered employees to maintain the skill level required when they began working. The FSP required covered employees to demonstrate by taking assessments that they had developed or improved in six skill areas or face discipline.

The undisputed evidence establishes that for covered employees who failed to advance two skill levels per year, up to 40 hours per year spent in training for which Dow received reports from Brazosport College was not voluntary. There are, however, disputed fact issues as to whether time spent in self-study, which Dow did not count in determining whether an employee had made reasonable efforts to meet the two-skill-level-per-year advancement, was voluntary.

Dow included the online hours reported by the Brazosport College in determining the number of training hours each covered employee had spent. As discussed below, there are disputed fact issues as to the number of hours the covered employees in fact spent in online training as opposed to merely logged in.

2. Related to the Employees' Jobs

Dow argues that there are disputed fact issues material to determining whether the time spent training for assessments was "directly related" to the plaintiffs' jobs. The "directly related" requirement is the third element of the Labor Department regulations. The Labor Department has issued regulations clarifying this prong of the analysis. The regulations provide:

The training is directly related to the employee's job if it is designed to make the employee handle his job more effectively as distinguished from training him for another job, or to a new or additional skill. . . . Where a training course is instituted for the bona fide purpose of preparing for advancement through upgrading the employee to a higher skill, and is not intended to make the employee more efficient in his present job, the training is not considered directly related to the employee's job even though the course incidentally improves his skill in doing his regular work.

29 C.F.R. § 785.29.

Case law has provided additional guidance. In Price, the Eleventh Circuit found that an employee who spent 90% of his time testing rubber goods and 10% working on existing metering equipment could not show that training on upgraded metering equipment was directly related to his job. Price, 806 F.2d at 1552. Citing the Labor Department regulations, the court reasoned that "[a]lthough appellant may have benefitted incidentally from training on the new solid-state meters, the courses were not related to his current job." Id. Similarly, in Dade County, the court found that the police officers' off-duty physical training was not directly related to their jobs. Dade County, 124 F.3d at 1385. The court explained:

The County did not require the officers to acquire or develop a skill unique to their employment as SRT officers. Although physical fitness training allows SRT officers to perform their core employment function of responding to emergency situations, such training also provides the individual officers with benefits that extend beyond their employment position.

Id. These cases provide support for Dow's argument that training is not directly related to an employee's job if it develops skills that are either beyond the requirements of the employee's specific job or so basic and transferrable that the skills would be useful to the employee in any job and beyond. In other words, if the training makes an employee a better person rather than simply better at doing a particular job, the training is not directly related to that job.

Dow also points to a Labor Department Wage and Hour Division Opinion Letter. Such opinion letters "constitute a body of experience and informed judgment" but are not binding. Skidmore, 323 U.S. at 140; Dade County, 124 F.3d. at 1385 (quoting Skidmore). In the letter Dow cites, a restaurant owner provided non-English-speaking employees with basic training in English language skills to "provide them greater opportunities in the workforce and enhance job satisfaction and workplace morale." DOL Wage and Hour Opinion Letter of March 3, 2006, 2006 WL 940661. The employees' current jobs for the restaurant owner did not require English proficiency. The letter concluded that the training was not directly related to the employees' jobs. Instead, the language training was "general in scope" and "designed to help the employee advance in society and in work" rather than to "'handle his job more effectively.'" Id. at *2 (quoting 29 C.F.R. § 785.29).

The evidence in the present record is mixed as to whether the FSP training enabled the operators to do their present jobs more effectively or to upgrade their basic skills so they would be better equipped to perform any jobs, particularly those involving increasingly sophisticated technology. The CBA describes the FSP as intended to "upgrade and maintain skills." (Docket Entry No. 193, Ex. B at 2). But no employee who completed the FSP was promoted to a different job, which the plaintiffs argue shows that the skills were directly related to the existing operator jobs. See Haszard v. American Medical Response Northwest, Inc., 237 F. Supp.2d 1151, 1165 ("the general rule of 29 C.F.R. § 785.27(c) that training time is compensable if it is 'not directly related to the employee's job[]' . . . is meant to encompass all forms of training that do not qualify the employee for advancement or promotion.").

Dow points to Campbell's statement in his declaration that "[t]he FSP was designed to upgrade the skills of covered employees so they could succeed in the next phase under Article XXXIV of the collective bargaining agreement, namely process operator assessments and training." (Docket Entry No. 192, Ex. C, ¶ 7). Campbell explained that the training for the assessments did not relate to day-to-day job functions but instead "was designed to address the position's requirements as they were going to evolve in the future through advancing technology." (Id.).

The record shows that when Dow created the FSP for existing operators, the job qualifications for a newly hired operator had been changed to require new hires to have an associate's degree. Dow added that degree requirement to ensure that operators had the necessary skills for jobs that increasingly required learning sophisticated and complex technical information. (Docket Entry No. 192, Ex. A, Campbell Depo. at 40). The existing operators needed to develop equivalent skills to be able to understand new technology and to perform jobs that would increasingly require training in what Dow referred to as "process technology." Dow set up the FSP out of concern that existing employees "may not have the skills and ability to understand" the process technology training. (Id. at 40-41). The newly hired operators'job description had changed to require an associate's degree; the FSP was put it in place so that existing operators could attain equivalent skill levels. The fact that newly hired operators were required to have an associate's degree and the existing operators were required to achieve certain levels on the assessments as an equivalent weighs in favor of finding that the FSP was directly related to the covered employees' jobs.

Other aspects of the FSP, however, support a different conclusion. The FSP was directed to "foundational skills." The assessments, and the training to prepare for those assessments, were on general, basic skills. Such skills, and the training to achieve and improve them, did not merely apply to specific jobs, but to better performance in the workplace in general and, indeed, in life. In this respect, the training is similar to the English-language training described in the DOL letter and to the fitness training at issue in Dade County. Basic skills in reading, applied mathematics, locating information, teamwork, applied technology, and observation improve the employees' ability to do many jobs and to be productive away from the workplace as well. See Dade County, 124 F.3d at 1385 ("such training also provides the individual officers with benefits that extend beyond their employment position."); DOL Wage and Hour Opinion Letter of March 3, 2006, 2006 WL 940661, at *2 ("The training is designed to help the employee advance in society and in work."); see also 29 C.F.R. § 785.31 (creating a "special" exemption to compensability if an employee voluntarily attends courses as part of a "program of instruction which corresponds to courses offered by independent bona fide institutions of learning" that the employer has established "for the benefit of his employees"); but see Haszard, 237 F. Supp.2d at 1165 ("The unique exception of college-type courses should be restricted to those situations in which the course is clearly offered on a voluntary basis for the benefit of the employee."). Fact issues remain as to whether the training, in whole or in part, was directly related to the plaintiffs' jobs.

To the extent that the plaintiffs seek summary judgment that their training and study time for the assessments was compensable, their motion is granted in part. The record shows that up to forty hours per year, reported to Dow by Brazosport College, was not voluntary; an employee who logged fewer hours and failed to improve two skill levels per year was subject to discipline. There are fact issues that must be resolved to determine whether other types of training were voluntary, because Dow did not count any other types of training in deciding whether to discipline employees for failing to meet the FSP requirements. There are also fact issues that must be resolved to determine whether all or part of the training time was related to the plaintiffs' jobs.

The plaintiffs' motion is for summary judgment is also granted with respect to the plaintiffs' argument that they did not waive their claims by accepting partial payment from Dow in the Labor Department settlement. The record shows that because Dow chose to pay two years instead of three years in back wages, the DOL did not require employees to sign a form waiving their rights. The DOL investigator informed the Union that employees would not waive their rights to pursue claims by accepting settlement payments. (Docket Entry No. 188, Ex. F-1). Dow has conceded that no waiver occurred. (Docket Entry No. 194 at 2 n. 1).

III. Dow's Motion for Partial Summary Judgment

Dow has moved for summary judgment on Maynor's individual claim for overtime pay as well as on his claim that he was discharged in retaliation for engaging in activity protected by the FLSA.

A. Maynor's FLSA Retaliation Claim

Maynor's retaliation claim arises under section 15(a)(3) of the FLSA, which makes it unlawful:

to discharge or in any other manner discriminate against any employee because such employee has filed any complaint or instituted or caused to be instituted any proceeding under or related to this chapter, or has testified or is about to testify in any such proceeding, or has served or is about to serve on an industry committee

29 U.S.C. § 215(a)(3). This claim is analyzed under the McDonnell Douglas burden-shifting approach, borrowed from Title VII. See Hagan v. Echostar Satellite, LLC, 529 F.3d 617, 624 (5th Cir. 2008). Maynor has the initial burden to make out a prima facie case of retaliation. His prima facie case has three parts: (1) participation in a protected activity; (2) an adverse employment action; and (3) a causal link between the two. If Maynor makes that showing, the burden shifts to Dow to articulate a legitimate, nonretaliatory reason for its decision to take the adverse employment action. If Dow can provide such a reason, the burden returns to Maynor to show that the proffered reason is a pretext for retaliation. Id. Maynor's summary judgment burden during this third stage is to raise a fact issue as to whether Dow took the action against Maynor "because of" his protected activity. Id. (citing Kanida v. Gulf Coast Medical Personnel LP, 363 F.3d 568, 576 (5th Cir. 2004)). This is a "but for" causation standard. Kanida, 363 F.3d at 580.

To satisfy the first element of his prima facie case, Maynor alleges that he complained to Robin Campbell, Dow's Labor Relations Manager in Freeport, and others, about Dow's refusal to compensate him and others for time spent training for and taking the assessments. The question is whether Maynor "filed any complaint" under section 215(a)(3). See Hagan, 529 F.3d at 623. The Fifth Circuit has construed that language to allow "an informal, internal complaint to constitute protected activity under Section 215(a)(3)." Id. at 626. But there are limits. "[N]ot all abstract grumblings or vague expressions of discontent are actionable as complaints." Id. (citations and quotations omitted). The informal complaint must "concern some violation of law" and be "framed . . . in terms of the potential illegality" of the action. Id. In Hagan, the panel found that the plaintiff's complaints about a schedule change were not protected activity because he did not contend — or, as he testified, even subjectively believe — that the change was illegal. The schedule change was in fact legal. The plaintiff had raised concerns "about the possibility" of employees receiving less overtime pay, not about any violation of the law. Id.

Aside from a complaint's content, the way it is made can also place it outside the reach of the statutory language. The Seventh Circuit has recently held that only written complaints can amount to protected activity. Kasten v. Saint-Gobain Performance Plastics Corp., 570 F.3d 834, 839 (7th Cir. 2009). The Karsten panel reached that conclusion by analyzing the words "to file." The dictionary meaning and common understanding supported the conclusion that "the phrase 'file any complaint' requires the submission of some writing to an employer, court, or administrative body." Id. at 838-39. The panel also compared the language Congress chose for the FLSA's retaliation provision to the antiretaliation provisions in Title VII and the Age Discrimination in Employment Act. Id. at 840. In Title VII and the ADEA, Congress provided protection to employees who "ha[d] opposed any practice" made unlawful by the statute. See 42 U.S.C. § 2000e-3(a); 29 U.S.C. § 623(d). By contrast, the FLSA antirretaliation provision protects an employee who has "filed any complaint or instituted or caused to be instituted any proceeding under or related to this chapter," "testified or is about to testify in any such proceeding, or has served or is about to serve on an industry committee." The court in Karsten concluded that Congress used the phrase "file any complaint" to limit the antiretaliation provision to employees who had filed written complaints. Karsten, 570 F.3d at 840.

One problem with this line of reasoning is that the FLSA provision was drafted in 1938 and has not since been amended; the relevant provisions of Title VII and the ADEA were enacted in 1964 and 1967, respectively.

The Fifth Circuit has not decided whether complaints must be written to trigger the FLSA antiretaliation provision. The Fifth Circuit has, however, disagreed with the reasoning of a case relied on by Karsten to support the conclusion that filing a document is required. That case, Lambert v. Genesee Hospital, 10 F.3d 46 (2d Cir. 1993), interpreted the FLSA provision as limiting "the cause of action to retaliation for filing formal complaints, instituting a proceeding, or testifying, but does not encompass complaints made to a supervisor." Id. at 55. The Fifth Circuit, in deciding that informal complaints to a supervisor satisfied the statute, rejected the Lambert holding as too narrow. Hagan, 529 F.3d at 626. Whether a reading of section 215(a)(3) requiring a written complaint would also be too narrow is unclear.

The Hagan opinion suggests but does not specify that the complaints at issue in that case were verbal and unwritten. 529 F.3d at 620-21.

Maynor filed three written complaints: two letters to Bill Lewis, the head of Human Resources at Dow's Freeport facility; and an EEOC Charge of Discrimination. Maynor's March 15, 2005 letter to Lewis stated:

On February 3, 2005, I received a written letter for failure to meet annual skills assessment by not signing the college release form. I request the company to remove the letter from my personnel file.

On 2-15-2005 the company and the union agreed to a non compliance agreement which I also disagree with. This grievance should be considered timely as of this date.

(Docket Entry No. 189, Ex. A, Maynor Depo., Ex. 13).

Maynor's second letter to Lewis, sent on August 10, 2005, stated:

On August 10, 2005 I received a written letter, one day suspension without pay, and loss of my 2005 Performance Award, for failure to meet skills initiative requirements. Further, I feel this discipline is also for not signing the college release form.

I request the company to remove this letter from my personnel file, reimburse me one day lost pay, and reinstate my 2005 Performance Award.

I personally feel this is harassment due to the fact that the company is aware that I am working with the EEOC to resolve this problem.

(Id., Ex. A, Maynor Depo., Ex. 16).

Maynor's EEOC Charge of Discrimination is not part of the record. Some information about the charge is provided by Maynor's deposition. Maynor filed his EEOC Charge on June 21, 2005, alleging age discrimination on the basis that the FSP was harder for older employees. (Id., Ex. A, Maynor Depo. at 89-90). There is nothing in the record indicating that the charge addressed Dow's failure to pay for time spent to comply with the FSP requirements. When asked where in the charge he complained about compensation for training hours, Maynor responded, "I don't see it." (Id).

Maynor amended his EEOC Charge twice. Those amendments are irrelevant to the retaliation claim because they were made after termination. Similarly, Maynor's post-termination letter to Robin Campbell cannot form the basis for a retaliation claim.

None of these three written complaints describes an FLSA violation, even in general terms. None of the three written complaints mentions Dow's failure to pay. The case law is clear that a complaint must complain about the failure to pay for time spent on work, whether it is overtime or failure to pay minimum wage. See Hagan, 529 F.3d at 626 (denying relief where the employee "did not frame any of his objections in terms of the potential illegality of the charge." (emphasis original)). But Maynor also made an oral complaint to Robin Campbell in a meeting on December 23, 2005. Campbell's notes from the meeting are part of the record. Campbell observed that "Roy really seemed focus (sic) that his issue was us forcing them to go do this on their own time." (Docket Entry No. 193, Ex. B, Maynor Decl., Attachment B-3). Although Maynor's exact words are not in the record, it appears that he did make a complaint about Dow's failure to pay employees for time participating in the FSP and that his complaint conveyed more than an "abstract grumbling or vague expression[] of discontent." See Hagan, 529 F.3d at 626. Campbell's notes support an inference that Maynor specifically complained that he and other employees were required to train and study for the assessments without pay. Maynor's declaration is consistent. He states: "During that meeting, I made it clear to Mr. Campbell that my main problem was the fact that employees were forced to do it on our own time." (Docket Entry No. 193, Ex. B, Maynor Decl. at 2).

It is not necessary that Maynor's complaint included a specific statement that Dow's policy violated the FLSA. Lambert v. Ackerley, 180 F.3d 997, 1008 (9th Cir. 1999) (en banc) ("it is clear that so long as an employee communicates the substance of his allegations to the employer (e.g., that the employer has failed to pay adequate overtime, or has failed to pay the minimum wage), he is protected by § 215(a)(3)." (emphasis original)). Even though Maynor did not communicate the substance of his FLSA allegations to Dow in a written complaint, he communicated to Campbell verbally that the "focus" of his complaint was Dow's failure to compensate employees for time spent participating in the FSP. Given the absence of Fifth Circuit precedent requiring a written complaint, the record does not allow dismissal on the basis that Maynor's complaints are not protected activity.

The parties have not addressed the two other elements of "protected activity, which are that the plaintiff "stepped outside" his job to make the complaint and had a good faith belief that the complained-of conduct was unlawful. Hagan, 529 F.3d at 625. The record shows that Maynor went outside his normal role when he complained about Dow's refusal to pay for training and assessments under the FSP. The record also shows at least a fact issue as to whether Maynor had a good faith belief in the illegality of Dow's refusal to pay for the time spent in assessments and training. Maynor testified in his deposition that it "isn't right" for Dow to expect employees to train on their own time. (Docket Entry No. 189, Ex. A at 17). He stated in his declaration that he had "always been of the opinion that it was unfair to make the operators go to school on their own time." (Docket Entry No. 193, Ex. B at 2).

The second element of the prima facie case is not disputed. Dow took an adverse employment action against Maynor when it terminated his employment.

The third element requires some "causal link" between Maynor's complaints and his termination. Maynor need not show but-for causation to make a prima facie case. Title VII retaliation cases, which require the same "causal link," are instructive. Under Title VII, a "causal link" for a prima facie is shown if "the employer's decision to terminate was based in part on knowledge of the employee's protected activity." Medina v. Ramsey Steel Co., Inc., 238 F.3d 674, 684 (5th Cir. 2001). "Close timing between an employee's protected activity and an adverse action against [her] may provide the 'causal connection' required to make out a prima facie case of retaliation." Swanson v. Gen. Servs. Admin., 110 F.3d 1180, 1188 (5th Cir. 1997). The Supreme Court has noted that "cases that accept mere temporal proximity . . . as sufficient evidence of causality to establish a prima facie case uniformly hold that the temporal proximity must be 'very close.'" Clark County Sch. Dist. v. Breeden, 532 U.S. 268, 273 (2001). The Fifth Circuit has found temporal proximity of up to four months sufficient to show a causal link. See, e.g., Bell v. Bank of Am., 171 Fed. App'x 442, 444 (5th Cir. 2006) (holding that a seven-month period does not support establish a causal link); Raggs v. Miss. Power Light Co., 278 F.3d 463, 471-72 (5th Cir. 2002) (holding that a five-month period does not support an inference of a causal link); Harvey v. Stringer, 113 F.App'x 629, 631 (5th Cir. 2004) (unpublished) (finding that a ten-month period did not create a causal link); Stroud v. BMC Software, Inc., 2008 WL 2325639 (5th Cir. Jun. 6, 2008) (finding that a three-week lapse between protected activity and adverse employment action was sufficient to show a causal link); Richard v. Cingular Wireless LLC, 233 F.App'x 334, 338 (5th Cir. 2007) (unpublished) (concluding that two-and-one-half months is short enough to support an inference of a causal link); Jones v. Robinson Property Group, L.P., 427 F.3d 987, 995 (5th Cir. 2005) (finding that a period of less than sixty days was sufficiently close to establish a causal link for prima facie case of retaliation); Ware v. CLECO Power LLC, 90 F.App'x 705, 708 (5th Cir. 2004) (unpublished) (finding a fifteen-day period sufficient to support an inference of causation); Evans v. City of Houston, 246 F.3d 344, 355 (5th Cir. 2001) (finding that "a time lapse of up to four months has been found sufficient to satisfy the causal connection for summary judgment purposes" (internal citations omitted)); see also Swanson v. Gen. Servs. Admin., 110 F.3d 1180, 1188 (5th Cir. 1997) ("Close timing between an employee's protected activity and an adverse action against him may provide the 'causal connection' required to make out a prima facie case of retaliation.").

Maynor complained to Campbell on December 23, 2005. He was fired on February 7, 2006, roughly a month and a half later. This temporal proximity is sufficient to support an inference that there was a causal link between Maynor's complaint and his job termination. Maynor has made a prima facie showing on each of the required elements.

Dow has met its burden of articulating a legitimate, nonretaliatory reason for firing Maynor: insubordination by failing to comply with the requirement of signing the FERPA release form. Maynor must identify the basis for a disputed fact issue as to whether, but for his complaint, he would not have been fired.

Maynor argues that Dow's insubordination explanation is a pretext for retaliation because "Maynor's refusal to sign a release was part and parcel of his protected activity done in opposition to a mandatory job related program." Maynor concludes that "Dow's stated reason for Maynor's discharge is directly related to his protected activity." (Docket Entry No. 193 at 11). Dow cites three cases to argue that "[c]ourts have consistently rejected similar arguments." (Docket Entry No. 195 at 4). In Bailey v. Southwest Gas Co., 275 F.3d 1181, 1187 (9th Cir. 2002), the Ninth Circuit upheld summary judgment for the employer on an employee's claim under Nevada law that she had been terminated in retaliation for complaining about having to work overtime. The panel found that Bailey had not met her burden of proving "proximate causation" because "[s]he was not terminated until she repeatedly refused to cooperate with [the employer's] legitimate requests for medical information necessary to evaluate her fitness for duty, and after she had been warned that termination would result if she did not submit a completed physician certification form." Id. In Hazel v. U.S. Postmaster General, 7 F.3d 1, 4 (1st Cir. 1993), the court of appeals reviewed a bench trial in which the district court rejected the plaintiff's Title VII and ADEA retaliation claims. The appellate court applied a clearly erroneous standard of review and stated that "Hazel's admission that he refused to report for work and the evidence that the Postal Service fired him for that very reason provided a 'plausible' basis for the district court's finding that retaliation is not the most likely reason Hazel was fired." Id. In Sullivan v. River Valley School Dist., 197 F.3d 804, 813-14 (6th Cir. 1999), the final case Dow cites, the Sixth Circuit upheld summary judgment dismissing an ADA claim because the plaintiff did not claim that his work suspension for failing to submit to a medical examination was a pretext for discrimination. Instead, he argued that he had been asked to take the examination so that the employer could create a false perception that he was disabled. In other words, the plaintiff refused to comply with a requirement that was imposed to make him appear disabled; the failure to take the test was not claimed as a pretext for the adverse employment action. The appellate court found that such behavior by the employer would be "unprofessional and petty," but not an ADA violation. Id.

The issue is whether there are disputed facts material to determining whether Dow fired Maynor because he refused to sign the release to allow Dow to receive the assessment test scores and information on the training he did through the Brazosport College, or whether Dow fired Maynor because he complained about Dow's refusal to pay employees for the off-duty time spent in taking the tests and preparing for them. Maynor argues that Dow's explanation is pretextual because the requirement that employees sign a FERPA release was "directly related" to — "part and parcel of" — the activities that were the basis of Maynor's complaint to Campbell. But the relationship between the FERPA release requirement and Maynor's complaint does not create a factual issue material to determining but-for causation. Nor does the temporal proximity between the Campbell meeting and Maynor's firing. See Strong v. Univ. Healthcare System, LLC, 482 F.3d. 802, 807-08 (5th Cir. 2007) ("temporal proximity alone is insufficient to prove but for causation."); Shirley v. Chrysler First, Inc., 970 F.2d 39 (5th Cir. 1992) (temporal proximity is only "one of the elements in the entire calculation"). The record is clear that Maynor was given ample warning of the consequences of failing to comply with the FSP requirement to sign the release. Long before Maynor was fired, he was warned repeatedly, disciplined, and suspended because he failed to sign the release. The Union encouraged Maynor to sign the release, but he refused to do so. (Docket Entry No. 189, Ex. A, Maynor Depo. at 71-83).

The evidence that does create a fact issue is that Maynor gave Dow a paper copy of the information from Brazosport College, including his training hours and assessment test results, in January 2005, a month before he was fired. Cheryl Weinberger and Robin Campbell received the record and knew what that it was the same type of information they would receive if Maynor had signed the release form. Weinberger told Maynor in an e-mail that Campbell would be meeting with the legal department and that "[a]t that time, they will consider your request to accept the paper-copies of your college information," until which time she would "keep the packet in my office." (Docket Entry No. 193, Ex. F). Dow argues that this does not create a fact issue on but-for causation because "the fact that Maynor may have given a print-out of his purported training time to his supervisor does not change the fact that Maynor refused to sign the College required release form after numerous warnings." (Docket Entry No. 195 at 5). That argument is unpersuasive. Dow required the FERPA release so it could get from Brazosport College each covered employee's training obtained through the college and determine whether the employee had "achieved profile" on the assessments. Weinberger's warning e-mails to Maynor indicated that Dow wanted the release signed in order to get the information from Brazosport College. Her e-mails state: "Without the college information, I will have to assume that you have done nothing to achieve the Skills Initiative requirements." (Docket Entry No. 189, Ex. A, Maynor Depo., Ex. 10). Campbell made the following statement about the FERPA release in his deposition:

What I'm trying to say is that the only way for us to know if you had activity was for the college to release the information. It could only be released from a signing of a FERPA. So, you either don't sign a FERPA, or you didn't do anything in training, if you had done something in training.

(Id., Ex. B, Campbell Depo. at 115). With Weinberger's e-mails, this testimony also confirms that the FERPA release requirement was the means to obtain information from Brazosport College, not an end in itself. There is no evidence as to whether Dow concluded that without the release, it could not look at information compiled by Brazosport College about Maynor's tests and training, even if Maynor himself gave Dow the information. There is no evidence that Dow believed there was any issue about the reliability or accuracy of the transcript information Maynor provided. Given the lack of such evidence and the short time between Maynor's complaint and his firing, there is a disputed fact issue as to whether Dow's explanation that it fired Maynor for failing to sign the release form is a pretext for retaliation for the complaints he made. See Laxton v. Gap Inc., 333 F.3d at 578. Dow's motion for summary judgment on Maynor's retaliation claim is denied.

B. Maynor's Unpaid Hours Claim

Dow argues that, because Maynor did not sign a FERPA release, the company had no actual or constructive notice of his training hours and was not required to pay him for those hours. Dow did not address the paper transcript Maynor provided in the opening brief. (See Docket Entry No. 189 at 11-12). Dow contends in its reply brief that "the fact Maynor provided what he asserted was a paper copy of his transcript to his supervisor does not mean that Dow 'should have known' about his alleged training hours because there is no evidence that his supervisor or anyone else at Dow actually reviewed the report or deemed his copy reliable." (Docket Entry No. 195 at 5). Dow argues that it is "unsubstantiated speculation," an "unsupportable inference," and even a "quantum leap" to conclude that, because Maynor gave Dow a packet from Brazosport College with his training records, Dow should have known what Maynor did to train for the assessments. This argument is not a sufficient basis for granting summary judgment on Maynor's claims. The evidence is undisputed that Weinberger and Campbell knew that the packet included Maynor's training records from Brazosport College. The documents Maynor submitted have the Brazosport College letterhead. There is no evidence that Weinberger or Campbell believed that they could not look at the transcript or that it was unreliable or inaccurate, aside from the implausibly large number of online training hours that also were present on other employees' records obtained directly from the College because the employees had signed releases.

On the present record, Dow's argument that because it did not examine the packet containing Maynor's training records, it had no reason to know that Maynor had taken training offered through Brazosport College or how many hours he had spent, does not establish that, as a matter of law, Dow does not owe Maynor compensation for those hours. "An employer who is armed with knowledge that an employee is working overtime cannot stand idly by and allow an employee to perform overtime work without proper compensation, even if the employee does not make a claim for the overtime compensation." Newton v. City of Henderson, 47 F.3d 746, 748 (5th Cir. 1995) (alterations and quotations removed) (quoting Forrester v. Roth's I.G.A. Foodliner, Inc., 646 F.3d 413, 414 (9th Cir. 1981)). In Newton, the court held that the employer had no liability for failing to pay overtime because under the circumstances, the employer had no reason to know that the employee was working overtime hours. 47 F.3d at 748-50. The plaintiff employee was a police officer. Although he worked for a city, he was assigned to a federal Drug Enforcement Agency task force. While working on the task force, the plaintiff accumulated overtime hours. The city had not authorized many of these hours. Indeed, the city explicitly told its police officers not to work unauthorized overtime hours. City policy required any overtime hours to be reported within 72 hours, which the plaintiff did not do. The plaintiff gave reports of his hours to the DEA for purposes unrelated to payroll but did not give the reports to the city until he resigned from the police department. Id. at 747-48. In his FLSA lawsuit, the plaintiff offered two reasons why the city should have known about his overtime hours. First, he argued that the police chief was on the task force's board of directors and could have obtained the reports of the hours he worked from the task force. Second, he pointed to deposition testimony by the city's police chief that undercover task force officers often have to work overtime. The district court entered judgment for the plaintiff after a bench trial. The Fifth Circuit reversed, holding that it was not enough to show that the employer "had the ability to investigate" the number of hours an employee was working, especially given evidence that the employee was told that he could not work unauthorized overtime. The appellate court noted the police chief's testimony that although undercover officers might have to work extra hours, they were expected to take "flex time" to compensate instead of pay at overtime rates. Id. at 748-750.

In this case, Maynor gave Dow a printed copy of the records from the Brazosport College showing the number of hours he spent in training for the assessments Dow required. Dow did not have to conduct an investigation in the absence of any reason to believe that Maynor was spending time training and taking assessments. Dow had to look at the records Maynor provided. No other investigation was required. Unlike the employer in Newton, which told the employee not to work overtime without specific authorization, Dow expected that most covered employees, including Maynor, would have to spend off-duty hours training. Maynor, unlike the officer in Newton who had been told not to work unauthorized overtime, had been told that if he did not take at least forty hours of training and failed to progress two skill levels, he could be subjected to discipline, up to suspension without pay. Although not all employees required training to advance two skill levels per year, most of the covered employees had to spend some time in some form of training to do so. After February 2005, Dow's policy subjected them to discipline if they spent less than forty hours and failed to advance.

Maynor has raised a genuine issue of material fact on his pay claim. Dow's motion for summary judgment on that claim is denied.

IV. The Decertification Motion

A. The Legal Standards

The FLSA requires employers to compensate employees who work more than forty hours in a workweek at overtime rates. The statue states in relevant part:

Except as otherwise provided in this section, no employer shall employ any of his employees who in any workweek is engaged in commerce or in the production of goods for commerce, or is employed in an enterprise engaged in commerce or in the production of goods for commerce, for a workweek longer than forty hours unless such employee receives compensation for his employment in excess of the hours above specified at a rate not less than one and one-half times the regular rate at which he is employed.

29 U.S.C. § 207(a)(1). "'An employer who is armed with [knowledge that an employee is working overtime] cannot stand idly by and allow an employee to perform overtime work without proper compensation, even if the employee does not make a claim for the overtime compensation.'" Harvill v. Westward Comm'cns, L.L.C., 433 F.3d 428, 441 (5th Cir. 2005) (quoting Newton v. City of Henderson, 47 F.3d at 748).

Section 216(b) of the FLSA creates a cause of action for employees against employers violating the overtime compensation requirements:

An action . . . may be maintained . . . by any one or more employees for and in behalf of himself or themselves and other employees similarly situated. No employee shall be a party plaintiff to any such action unless he gives his consent in writing to become such a party and such consent is filed in the court in which such action is brought.

29 U.S.C. § 216(b). Section 216(b) establishes an "opt-in" scheme under which plaintiffs must affirmatively notify the court of their intention to become parties to the suit. See Mooney v. Aramco Servs. Co., 54 F.3d 1207, 1212 (5th Cir. 1995).

Certification of a collective action suit is generally analyzed under a two step process. The first step is the conditional certification, or "notice stage," in which the district court decides whether to issue notice to potential class members. See Mooney, 54 F.3d at 1213-14 (citing Lusardi v. Xerox Corp., 118 F.R.D. 351 (D. N.J. 1987)). The court's decision is often based only on the pleadings and affidavits that have been submitted. Id. "Because the court has minimal evidence, this determination is made using a fairly lenient standard, and typically results in 'conditional certification' of a representative class" that provides potential class members with notice and the opportunity to opt-in. Id. at 1214 n. 8. At this first stage, there must be a showing of "some identifiable facts or legal nexus [that] bind the claims so that hearing the cases together promotes judicial efficiency." Barron v. Henry County Sch. Sys., 242 F. Supp. 2d 1096, 1103 (M.D. Ala. 2003) (citing Sheffield v. Orius Corp., 211 F.R.D. 411, 416 (D. Or. 2002)); see also Basco, 2004 WL 1497709, at *5 (quoting Heagney v. European Am. Bank, 122 F.R.D. 125, 127 (E.D.N.Y. 1988) (stating that certification is appropriate when some factual nexus binds the named plaintiffs and potential class members as victims of a particular alleged policy or practice)).

If a court conditionally certifies a class, the action proceeds as a collective action during discovery. See Mooney, 54 F.3d at 1214. The second stage typically occurs when discovery is largely complete and the defendant moves to "decertify" the conditionally certified class. See id.; Lusardi, 118 F.R.D. at 359. At that point, the court makes a factual determination as to whether there are similarly situated employees. Id. If the district court finds that the claimants are similarly situated, the collective action may proceed. See Mooney, 54 F.3d at 1214; Basco, 2004 WL 1497709, at *3. "A court may deny plaintiffs' right to proceed collectively if the action arises from circumstances purely personal to the plaintiff, and not from any generally applicable rule, policy, or practice." England v. New Century Fin. Corp., 370 F. Supp. 2d 504, 507 (M.D. La. 2005); see also Barron, 242 F. Supp. 2d at 1104 ("[T]he mere fact that violations occurred cannot be enough to establish similarity, as that would not ultimately be sufficient to establish a pattern and practice without a showing that the violations were more than sporadic occurrences."). If the court decertifies the class, the opt-in plaintiffs are dismissed without prejudice and the original plaintiff proceeds with individual claims. England v. New Century Fin. Corp., 370 F. Supp. 2d 504, 508 (M.D. La. 2005).

This case has reached the second stage and Dow has moved to decertify the class. At this stage, the plaintiffs have the burden to prove that they are similarly situated. Proctor v. Allsups Convenience Stores, Inc., 250 F.R.D. 278, 280 (N.D. Tex. 2008) (citations omitted). With discovery materials available to aid the court's analysis, the "similarly situated" inquiry is more searching than it was at the conditional certification stage. Id. But "[w]hile the plaintiffs must show that they are 'similarly situated,' this does not mean they must establish that they are 'identically situated.'" Id. The analysis turns on: "(1) the disparate factual and employment settings of the individual plaintiffs; (2) the various defenses available to the defendant which appear to be individual to each plaintiff; and (3) fairness and procedural concerns." Id. (quoting Basco, 2004 WL 1497709, at *4). "The three factors are not mutually exclusive and there is considerable overlap among them." Johnson v. Big Lots Stores, Inc., No. 04-3201, slip op. at 17 (E.D. La. Jun. 20, 2008).

Dow argues that there is no common policy or practice affecting all class members because there was no requirement for training for the assessments. Dow also argues that there are individualized defenses that make collective action treatment improper. Both arguments are analyzed below.

B. Applying the "Similarly Situated" Factors

1. Factual Setting

The presence of a common policy, plan, or practice affecting all class members, although not necessarily required, is helpful to showing a similar factual setting. Falcon v. Starbucks Corp., 580 F.Supp.2d 528, 534-35 (S.D. Tex. 2008) (collecting cases). It is undisputed that all of the plaintiffs were subject to the FSP and had to take the assessments. Dow argues, however, that the FSP is not a "common policy" because it did not require training for the assessments in general or any type or amount of training in particular. The plaintiffs agree that they trained for the assessments in different ways — classroom study, computer labs, tutoring, self-study, and online study — and for different amounts of time. The question is whether these variations preclude finding a common policy, plan, or practice affecting the covered employees.

In granting motions for decertification, many cases emphasize that class members were supervised by a variety of managers and worked in a variety of locations, stores, or offices. See Proctor, 250 F.R.D. at 281 ("The Plaintiffs' claims vary from store to store and manager to manager"); Johnson v. TGF Precision Haircutters, Inc., 2005 WL 1994286, at *2 (S.D. Tex. Aug. 17, 2005) ("The evidence here is that there was a substantial variance of experiences by different Plaintiffs under different managers and at different shops around Texas."); Basco, 2004 WL 1497709, at *8 (noting breadth of possible experiences at Wal-Mart stores across different regions in Louisiana and even under different managers within the same store). Such differences in the plaintiffs' employment settings were important because they affected the amount and type of off-the-clock time required. In the present case, the variations in amount and type of training did not depend on where the individual plaintiff worked in the plant or who his supervisor was. The FS P was part of the CBA and applied to every plaintiff, regardless of manager or department. The evidence as to what Dow said about the type and amount of training for the assessments is common to each plaintiff. Every plaintiff was subject to the FSP. Although Dow asserts that it did not require any plaintiff to train in a particular way, for a particular number of hours, or even to train at all, Dow did require each plaintiff to take assessments and to advance two skill levels per year. Dow knew that most of the employees spent hours preparing for the assessments. After February 2005, Dow policy was to discipline employees who failed to advance as required if they had not put in at least 40 hours of training reported by Brazosport College. That the plaintiffs trained in different ways and for different amounts of time does not eliminate the "meaningful identifiable facts or legal nexus that bind the claims." Falcon, 580 F.Supp.2d at 535-36 (quotations and alterations removed) ("That [plaintiffs] did not perform exactly the same duties off-the-clock does not undermine the conclusion that the putative class is similarly situated."). This is especially true when, as here, subclassing provides a mechanism for addressing differences among groups within the class and bifurcation of liability and damages provides a mechanism for addressing some individual issues raised by claims for compensable self-study or online study hours. The first factor does not prevent the plaintiffs from proceeding collectively.

2. Individualized Defenses

Dow's primary argument for individualized defenses is that it did not know, and could not know, what kind of training and how much training many of the plaintiffs used. Dow points to the evidence that some of the plaintiffs are seeking compensation for hours spent in self-study, which was not reported on the spreadsheets provided by the Brazosport College. Dow also relies on evidence that many of the hours that Brazosport College reported for online study were unreliable because of the "flaw in the system" that showed all the hours an employee was logged on as hours spent training, even if the employee was away from the computer much or all of the time. (Docket Entry No. 192 at 16-19). Neither of these arguments leads to decertification.

It is undisputed that many of the plaintiffs' training hours were reported to Dow on spreadsheets provided by Brazosport College and are not challenged as unreliable or as requiring individualized inquiry. Dow received spreadsheets from Brazosport College showing how many hours each employee spent in such training as attending classes, working in computer labs, or taking tutoring. Dow does not assert it lacked information about these hours or that they were unreliable. As to employees reporting such training, the legal issues as to Dow's liability are common; there are no individualized defenses; and the calculation of any damages appears to be straightforward. See Anderson v. Mt. Clemens Pottery Co., 328 U.S. 680, 687, 66 S. Ct. 1187 (1946), superseded on other grounds by 29 U.S.C. § 254(a) ("When the employer has kept proper and accurate records the employee may easily discharge his burden by securing the production of those records"); Moss v. Crawford Co., 201 F.R.D. 398 (W.D. Pa. 2000) (refusing to decertify because the employer kept records of the number of hours the employees billed clients and compensation was based on such billings).

Dow argues that as to the plaintiffs who are claiming compensation for self-study hours, an individualized inquiry into what each plaintiff told Dow about these hours will be required. To recover under the FLSA, the plaintiffs must prove that Dow knew or should have known that they worked overtime hours. See Newton v. City of Henderson, 47 F.3d at 748. There is a common question as to whether Dow knew or should have known that self-study was one of the training options employees subject to the FSP were using. That general question can be answered by examining the evidence as to the FSP and the information available to Dow independent of specific communications by each plaintiff about his training hours. The inquiry will turn on evidence about whether Dow knew that Brazosport College made manuals and other written materials available to employees to study themselves, that many employees were using these manuals outside any class or lab that Brazosport College reported on the spreadsheets submitted to Dow, and that employees were using other kinds of self-study as well, are questions common to each employee who is claiming compensation for self-study hours. The inquiry into Dow's information about self-study training hours raises issues that are not present in the claims for compensation for attending classes, computer labs, or tutoring. Those differences are best addressed by treating the plaintiffs claiming compensation for self-study training hours as a subclass. But these differences do not defeat collective action treatment either generally or as to this subclass.

The absence of a mechanism for employees to report self-study hours does not defeat class treatment. Anderson v. Mt. Clemens Pottery Co., 328 U.S. 680, 687-88, 66 S. Ct. 1187 (1946), superseded on other grounds by 29 U.S.C. § 254(a) ("The employer cannot be heard to complain that the damages lack the exactness and precision of measurement that would be possible had he kept records in accordance with the requirements of . . . the [FLSA]."). Many courts have found that class treatment is proper even when the number of overtime hours must be proven by the plaintiffs' testimony without the benefit of written records. See, e.g., Moreau v. Klevenhagen, 956 F.2d 516, 522-23 (5th Cir. 1992) (remanding to allow deputy sheriffs to testify in damages proceeding as to the number of hours they spent in training for firearms qualifications); Falcon, 580 F.Supp.2d at 536-37 (denying decertification of a class of Starbucks assistant managers who were given more than 40 hours worth of duties but discouraged from reporting overtime and who worked off the clock to compensate); Hill v. Muscogee County School Dist., 2005 WL 3526669, at *4 (M.D. Ga. 2005) (denying motion to decertify class of teacher's assistants who were given more duties than could be completed in an eight-hour day but told not to record more than eight hours on their time sheets); Thiebes v. Wal-Mart Stores, Inc., 2004 WL 1688544, at *1-2 (D. Or. July 26, 2004) (submitting 108 verdict forms to the damages-only jury, asking the jury to determine whether each Wal-Mart employee plaintiff worked unpaid hours off-the-clock — and how many — based on the plaintiffs' own testimony about their hours worked).

The need for individualized inquiries to determine the number of hours spent in self-study does not defeat certification. Similarly, to the extent the issue is the amount of hours each individual plaintiff spent in online training, this issue affects damages, not liability. In such cases, the Fifth Circuit has endorsed bifurcation in FLSA collective actions. See Moreau v. Klevenhagen, 956 F.2d 516, 521-23 (5th Cir. 1992). In Moreau, a group of deputy sheriffs sued seeking compensation for time spent in weapons training. The district court bifurcated the case into a liability phase and a damages phase, with separate discovery to be conducted on each issue. After the liability phase, the court granted a defense motion for summary judgment. The Fifth Circuit reversed, finding clear evidence of a written policy under which deputies would not be paid overtime for firearms training. "Whether any deputies were actually deprived of overtime compensation because of the County's firearm qualification policy should have been addressed only at the damages stage of the proceedings." Id. at 523. The panel remanded to the district court to determine whether any deputies lost out on pay, and if so, to what extent. Id. The Fifth Circuit has voiced no objection to bifurcating at the trial stage either. See Barfield v. Madison County, 212 F.3d 269, 271 (5th Cir. 2000) (describing the district court's decision to bifurcate FLSA collective action into separate trials for liability and damages); Mireles v. Frio Foods, Inc., 899 F.2d 1407, 1410 (5th Cir. 1990) (same).

Other courts have found bifurcation useful in FLSA actions as well. See, e.g. Falcon, 580 F.Supp.2d at 541 ("the Court is willing to consider bifurcating the liability and damages stages of the trial if it will be in the interest of judicial economy."); Thiebes v. Wal-Mart Stores, Inc., 2004 WL 1688544, at *1 (D. Or. July 26, 2004) (describing the court's earlier decision "to bifurcate the case for separate trials on liability and damages); see also FED R. CIV. P. 42(b) ("For conveinece or to avoid prejudice, or to expedite and economize, the court may order a separate trial of one or more separate issues, claims, crossclaims, counterclaims, or third-party claims."); MANUAL FOR COMPLEX LITIGATION (4th) § 21.141 ("Bifurcation and severance under Rule 42 are available as tools that might make a case more manageable by separating out discrete issues for a phased or sequenced decision by the judge or at trial."); WILLIAM B. RUBENSTEIN, ALBA CONTE HERBERT B. NEWBERG, NEWBERG ON CLASS ACTIONS § 24:1, n. 14 (4th ed.) ("Bifurcation is possible in FLSA class suits."); 7B CHARLES A. WRIGHT, ARTHUR R. MILLER MARY KAY KANE, FEDERAL PRACTICE AND PROCEDURE § 1807 at 503-04 (3d ed. 2005) ("In the few reported cases that have proceeded to trial, courts have bifurcated the proceedings, allowing the collective action to proceed first on a representative or group basis for the liability stage of the litigation. Certification is then reviewed at the remedial stage, in order to determine whether the litigation should continue as a group action, whether subclasses should be established, or whether the action should be dismantled into individual trials.").

Dow also argues that for employees claiming compensation for hours spent in online training, there are individualized issues that defeat certification. Dow does not dispute that it knew that employees were spending training time online because these hours were reported by Brazosport College. Dow points to the flaw in the system in arguing that the number of hours reported is unreliable and will require individualized determination. Dow argues that the unreliability of the online hours presents individualized issues that defeat certification. The specific issues presented by those claiming compensation for online hours can adequately be addressed by subclassing and bifurcating liability from damages. How many hours each employee spent in actual online training as opposed to merely being logged on is appropriately determined as part of the damages calculation for the subclass of those claiming compensation for online hours. The need for individual plaintiffs to establish the amount of uncompensated time does not defeat certification. See, e.g., Falcon, 580 F. Supp.2d at 536-37 (Starbucks assistant managers); Hill, 2005 WL 3526669, at *4 (teaching assistants); Thiebes, 2004 WL 1688544, at *1-2 (Wal-Mart employees).

Dow also argues that even as to a subclass of plaintiffs claiming compensation for self-study hours and of plaintiffs claiming compensation for online hours, whether those hours are compensable raises individualized inquiries. Dow points out that not only did the plaintiffs use different kinds of self-study, ranging from reading the manuals provided by Brazosport College to reading the newspaper, but some plaintiffs claimed self-study time for hours during which they were "on the clock" at Dow. Roy Maynor testified that he read the printed materials provided by Brazosport College while driving his car. As a small part of his training hours, Anthony Moody testified that he read the newspaper to test his retention. Dow argues that whether hours spent training while on the clock at Dow, doing other activities such as driving, or doing things that the employee would have done anyway, are individualized inquiries that should defeat certification.

The evidence shows that most of the employees spent their self-study hours reading the manuals provided by the Brazosport College (and did not read while driving) or in online study through the College's web portal. Whether time spent in self-study or online study while also being paid "on the clock" during a regular shift can be determined for all plaintiffs claiming compensation for such time. The question of whether time spent in training while on the clock at Dow can be decided on a common basis and does not defeat certification.

Nor does the variation in the type of activities used for self-study defeat certification. Dow does not dispute that reading manuals provided by the College off the clock at Dow is appropriate self-study. The amount of such time, given the absence of reporting through the College, is a question of damages, not liability. Whether time spent on activities such as reading the newspaper (that is, an activity that the employee already did for reasons unrelated to training) is compensable is a liability question that can be resolved for all employees claiming such hours within the subclass of those plaintiffs claiming compensation for self-study hours. The issue of whether reading training manuals while driving is compensable appears to be raised only as to one employee — Maynor — and can appropriately be resolved as part of the damages inquiry into the number of self-study hours that were in fact spent on training. In short, the common issues predominate and the "individualized" liability issues apply to groups of plaintiffs and can be resolved on that basis.

3. Fairness and Procedural Concerns

There are some liability issues raised only by those claiming compensation for self-study hours, best addressed by a subclass of such plaintiffs. There are damages calculation issues raised by those claiming compensation for self-study and for online study that are best addressed by bifurcating liability from damages. On balance, it is more efficient to try the common issues together. "It would certainly not be in the interest of judicial economy to decertify a class of similarly situated plaintiffs simply because, after discovery, it becomes apparent that the alleged policy was not as uniform as plaintiffs believed at step one." Falcon, 580 F. Supp.2d at 536.

The initial step at the damages trial — assuming that liability is found — will be applying the restrictions on compensable hours defined during the liability trial to each plaintiff's claim. Once the noncompensable hours are identified, there are two distinct computational issues that will arise during the damages phase. First, there are no records of self-study time, as the only records are those maintained by the College, which only tracked online training hours and hours spent at the College. Second, the online hours in those records are inflated because the plaintiffs were able to log on and leave the clock running while not actually using the online portal for training. The deposition testimony and documentary evidence of inflated hours already in the record make it clear that the online study time was not recorded accurately. This does not make class treatment improper. See, e.g., Falcon, 580 F.Supp.2d at 536-37 (denying decertification for Starbucks assistant managers whose only written records inaccurately stated that they had worked only 40 per week); Hill, 2005 WL 3526669, at *4 (denying decertification for teaching assistants for whom there were no accurate written records of hours worked); Thiebes, 2004 WL 1688544, at *1-2 (upholding jury verdict for Wal-Mart employees for whom there were no written records of off-the-clock hours in spite of Wal-Mart's detailed timekeeping reflecting the employees' regular working hours).

Although the record shows that employee training hours were not submitted to Dow until after the Addressing Non-Compliances Agreement was signed in February 2005, records of those hours appear to be available because Dow was able to determine that employees had taken an average of 18.5 training hours to advance one skill level. (Docket Entry No. 192, Ex. B, Campbell Aff., ¶ 16).

With respect to the online hours, under Dow's contract with Brazosport College, the hours covered employees spent in online training were tracked electronically on the ACT servers and reported to Dow. Although Dow did not know of the flaw in the system that made the reports inaccurate, the inaccuracy is not a basis to avoid FLSA liability. See Anderson v. Mt. Clemens Pottery Co., 328 U.S. 680, 687-88, 66 S. Ct. 1187,90 L.Ed. 1515 (1946), superseded on other grounds by 29 U.S.C. § 254(a) ("The employer cannot be heard to complain that the damages lack the exactness and precision of measurement that would be possible had he kept records in accordance with the requirements of . . . the Act."); Donovan v. Grantham, 690 F.2d 453, 457 n. 3 (5th Cir. 1982) ("[The employer] cannot now be heard to complain that the calculations are not completely accurate because of his own failure to keep adequate records."). Over 60 years ago, the Supreme Court explained what should happen when an employer failed to keep precise records of time compensable under the FLSA. The Court wrote:

The solution, however, is not to penalize the employee by denying him any recovery on the ground that he is unable to prove the precise extent of uncompensated work. Such a result would place a premium on an employer's failure to keep proper records in conformity with his statutory duty; it would allow the employer to keep the benefits of an employee's labors without paying due compensation as contemplated by the Fair Labor Standards Act. In such a situation we hold that an employee has carried out his burden if he proves that he has in fact performed work for which he was improperly compensated and if he produces sufficient evidence to show the amount and extent of that work as a matter of just and reasonable inference. The burden then shifts to the employer to come forward with evidence of the precise amount of work performed or with evidence to negative the reasonableness of the inference to be drawn from the employee's evidence. If the employer fails to produce such evidence, the court may then award damages to the employee, even though the result be only approximate.

Anderson, 328 U.S. at 687-88. This framework has been used consistently since. See Von Friewalde v. Boeing Aerospace Operations, Inc., 2009 WL 2391400, at *4, *10 (5th Cir. Aug. 4, 2009) (applying the Anderson rule); Harvill, 433 F.3d at 441 (same); Beliz v. W.H. McLeod Sons Packing Co., 765 F.2d 1317, 1330-31 ("the workers may satisfy their burden with admittedly inexact or approximate evidence."). The absence of precise evidence of the full extent of each plaintiff's damages is not cause for decertification.

The parties seem to agree that Dow would be entitled to an offset against any damages awarded for payments made to plaintiffs as part of the DOL settlement. Because the record indicates that these payments were documented, this is a mechanical task that can be completed at the end of the damages phase. It does not present manageability concerns.

Proceeding collectively does not impair Dow's due process rights. Dow will have an opportunity to present its arguments and defenses at the liability phase, which will include the right to cross-examine the representative witnesses. Dow will also be able to contest each individual plaintiff's claim at the damages stage.

At the motion hearing held on September 18, 2009, Dow advanced the argument that proceeding collectively would interfere with its Seventh Amendment right to a jury trial. In support of that argument, Dow cited Cimino v. Raymark Industries, 151 F.3d 297 (5th Cir. 1998). Cimino was an asbestos class action certified under Rule 23 in which the Fifth Circuit held that the trifurcated trial plan has violated the defendant's Seventh Amendment rights by failing to provide an individual inquiry into causation; into whether and to what extent each plaintiff was exposed to the defendant's asbestos products. Id. at 315. The court found that the class action device did not change the ordinary proof requirements established by the substantive law. In reaching that conclusion, it noted that it was bound by the Rules of Decision Act and Erie, which mandated applying the Texas law principle that each plaintiff must make a separate showing of causation and damages. Id. at 313. Because the FLSA is a matter of federal law and causation is not an issue, Cimino is distinguishable on its facts. For both reasons, there is no need for an individual showing of causation. Although Dow is entitled to contest each individual's damages, bifurcation will allow it to do so.

In sum, the three-factor analysis reveals that the plaintiffs are similarly situated, such that their action should proceed on a collective basis. Dow's motion to decertify the class is denied.

V. Conclusion

The plaintiffs' motion for summary judgment is granted as to the absence of a waiver by accepting the DOL partial payment and granted in part and denied in part as to the compensability of the time spent in conjunction with the Foundational Skills Program. Dow's motions for summary judgment as to Maynor's claims and for decertification are denied. A status conference is set for December 15, 2009 at 5:00 p.m. in Courtroom 11B.

Maynor v. Dow Chem. Co.

Summary

Opinion

MEMORANDUM AND ORDER

I. Background

A. The Site Foundational Skills Program

B. The February 2005 Agreement

C. Discipline For Failing to Comply with the Foundational Skills Program; Maynor's Job Termination

D. The DOL Investigation and the November 2006 Contract Modification

E. This Suit and the Conditional Collective Action Certification

II. The Plaintiffs' Motion for Summary Judgment

B. Analysis

III. Dow's Motion for Partial Summary Judgment

B. Maynor's Unpaid Hours Claim

IV. The Decertification Motion

B. Applying the "Similarly Situated" Factors

V. Conclusion

Maynor v. Dow Chem. Co.

Maynor v. Dow Chem. Co.

Case Details

Citations

Citing Cases

Maynor v. Dow Chem. Co.

Summary

Opinion

MEMORANDUM AND ORDER

I. Background

A. The Site Foundational Skills Program

B. The February 2005 Agreement

C. Discipline For Failing to Comply with the Foundational Skills Program; Maynor's Job Termination

D. The DOL Investigation and the November 2006 Contract Modification

E. This Suit and the Conditional Collective Action Certification

II. The Plaintiffs' Motion for Summary Judgment

B. Analysis

III. Dow's Motion for Partial Summary Judgment

B. Maynor's Unpaid Hours Claim

IV. The Decertification Motion

B. Applying the "Similarly Situated" Factors

V. Conclusion

Maynor v. Dow Chem. Co.

Maynor v. Dow Chem. Co.

Case Details

CitationsCopy Citation

Citing Cases

Citations