Arredondo v. Delano Farms Co.

Full title: SABAS ARREDONDO et al., Plaintiffs, v. DELANO FARMS CO., et al.…

Court: UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF CALIFORNIA

Date published: Oct 10, 2014

1:09-cv-01247 MJS (E.D. Cal. Oct. 10, 2014)

Opinion

1:09-cv-01247 MJS

10-10-2014

SABAS ARREDONDO et al., Plaintiffs, v. DELANO FARMS CO., et al., Defendants.

Michael J. Seng UNITED STATES MAGISTRATE JUDGE

ORDER DENYING MOTION FOR PROTECTIVE ORDER (Doc. Nos. 333, 340)

Plaintiffs seek a protective order to prevent Defendants from taking some 196 depositions of absent class members as a part of a "pilot study." Defendants assert the pilot study is a prerequisite to gathering statistical data necessary to respond to Plaintiffs' use of statistical evidence to prove labor violations suffered by the class.

I. FACTUAL AND PROCEDURAL BACKGROUND

The factual and procedural history of this case has been addressed in detail in this Court's earlier orders. The court addresses here only events relevant to the present motion.

Plaintiffs, agricultural workers employed by Defendants, filed this complaint on July 17, 2009 alleging various labor violations, including specifically that Plaintiffs performed uncompensated, off-the-clock work and were not provided compensation for the use of their own tools. Plaintiffs moved for class certification, and on April 19, 2011, the Court granted certification to a class consisting of four subclasses of field workers. (ECF No. 85.)

On November 13, 2012, Defendants moved to decertify the class. (ECF No. 203, 205.) After addressing issues of joint employment, the Court partially denied the motion for decertification on February 21, 2014. (ECF No. 310.) The Court granted the motion for decertification with regard to two of the four subclasses, but allowed the other two subclasses (involving uncompensated pre-shift work and unreimbursed tool purchases) to proceed. (Id.)

In its order on the decertification motion, the Court discussed at length the evidence presented in support of and in opposition to decertification and its pronounced lack of clarity and credibility:

In determining whether there is sufficient commonality, the Court must review and rely upon the declarations provided by each party. The Court is not attempting to determine the underlying merits of the matter, but at times it must address determinative issues to resolve certification issues. With that said, it is noted that there are general credibility concerns with regard to all the anecdotal evidence. Plaintiffs provided declarations supportive of their claims, and Defendants' declarations generally tend to show the contrary. Clearly, the declarations were not taken at random. Because of the state of the evidence, it is also not particularly helpful to combine and statistically analyze the information in the declarations. Because of the differences in number of declarations submitted by the opposing parties, any attempt to calculate the percentage of workers suffering violations based on the number of violations reflected in declarations is not likely to lead to an accurate result. The same problems arise when attempting to show the percentage of violations that occurred in each crew. In many instances only one person in the crew provided a declaration. In the case of some crews, all the declarations were provided by only Plaintiffs or only Defendants.

Arredondo v. Delano Farms Co., 2014 U.S. Dist. LEXIS 22658, 99-100 (E.D. Cal. Feb. 20, 2014).

The parties were then ordered to meet and devise a joint scheduling report. The parties' July 3, 2014, report described their inability to agree on how to conduct discovery relating to the liability phase of trial. (ECF No. 326.) Plaintiffs sought to test a potential survey on three to five class members and then administer the survey through stratified random sampling of approximately 1500 class members. Defendants intended to conduct a pilot study consisting of taking depositions of fourteen foremen and fourteen class members from each of the foremen's crews for a total of 210 depositions. Defendants argued that without conducting a pilot study, it would not be possible to determine if there was a legitimate basis to extrapolate the results of a survey to the remainder of the class members. Because Plaintiffs' class potentially contains a combination of class members who were subject to a policy to perform uncompensated work, class members who performed uncompensated work due to a failure of a policy to prevent them from doing so, and class members who possibly did not perform uncompensated work, Defendants assert that a pilot study should be performed to determine the amount of variability in the class. Defendants assert that the level of variability in the class determines the required sample size of a survey which can provide reliable results, unless variability is too great, in which case survey evidence may not prove to be a reliable tool to assist in establishing liability. (See ECF No. 326 at 20-21 (citing Duran v. U.S. Bank National Assn., 59 Cal. 4th 1, 33 (2014)).)

Despite further attempts to meet and confer, the parties could not come to an agreement on how to proceed with discovery. (See ECF No. 329.) On August 19, 2014, the Court ordered the parties to proceed to conduct discovery according to the respective methods that each had described. (ECF No. 330.)

Plaintiffs objected to Defendant's attempts to conduct discovery. After participating in a telephonic discovery dispute conference, Plaintiffs filed the instant motion for a protective order to prevent Defendants from deposing the 196 absent class members as a part of the pilot study. (ECF No. 333.) Defendants filed an opposition to the motion on September 26, 2014. (ECF No. 336.) The parties appeared at the hearing on October 1, 2014, and the matter now stands ready for adjudication.

II. PLAINTIFFS' CONTENTIONS

Plaintiffs contend that discovery from absent class members is ordinarily not permitted, and that there are several reasons why the depositions proposed by Defendants should not proceed. Specifically, Plaintiffs contend that the pilot study is to be used to show variability in the class, and therefore designed to reduce class size or revisit issues of certification. Plaintiffs contend further that Defendants could have obtained the discovery earlier in the litigation. Plaintiffs also argue that Plaintiff's counsel will be overburdened by having to attend the depositions (and attend they must to protect the interests of their client class members). Finally, Plaintiffs assert that Defense counsel, not experts, arbitrarily arrived at the number of depositions to be taken.

In light of the above, Plaintiffs request the Court grant a protective order and deny Defendants' proposed deposition of some 196 absent class members, or alternatively require Defendant to question absent class members by interrogatories or surveys.

III. DEFENDANTS' CONTENTIONS

The joint scheduling report filed by the parties on July 3, 2014, described in detail the reasons Defendants seek a pilot study and why they concluded that as many as 196 absent class members would need to be deposed. Defendants argue that it is essential that the parties first focus on variability in liability among class members and then on determining which groups of class members can establish liability. Defendants' pilot study is offered as an appropriate method for determining a process to ensure consistent, reliable and representative information can be obtained from a sample of class members. Defendants rely upon the testimony of their expert, Joseph Krock (ECF No. 337) and the recent California Supreme Court decision in Duran v. U.S. Bank National Ass'n, 59 Cal. 4th 1, 12 (2014). According to Krock, the greater the variability in the sample, the larger the required sample size needed to obtain, within an acceptable margin of error, a reliable result reflective of the population. (Krock Decl., ¶¶ 7-8); see also Duran, 59 Cal. 4th at 42 ("The more diverse the population, the larger the sample must be in order to reflect the population accurately." Further, "[i]t is impossible to determine an appropriate sample size without first learning about the variability in the population.") (Citations omitted.)

According to Defendants, the challenge in determining whether survey evidence can reliably establish liability of the class as a whole is complicated by the nature of the common questions at issue in this case. This is not like other class-action cases where plaintiffs can establish liability based on a company-wide policy. Here, with regard to the pre-shift subclass,¹ Plaintiffs do not suggest that there was a wrongful companywide policy, but instead allege that wrongful practices resulted from a failure to implement or enforce proper polices. (See Opp'n to Mot. for Deceit, ECF No. 286 at 24.) Plaintiffs therefore have to show what groups of class members actually engaged in pre-shift work. Defendants contend that the stratification built into their pilot study - dividing the larger population of the class into subgroups based on specified factors such as which foreman the class member worked under and dates of employment - will help identify subpopulations with common experiences. (Krock Decl. at ¶ 10.) Krock identified nine subgroups ("strata") taking into account the period and duration of time foremen worked and the time they worked at Delano Farms. (Id. at ¶ 14.) Krock's proposed pilot study is designed to enable the parties to test if subpopulations can be adequately defined and if a precise, statistically reliable, sample size can be determined. (Id. at ¶¶ 15 -16.) Acknowledging the law of diminishing returns, he feels that a pilot study of fourteen (14) foremen crews (deposing each foreman and fourteen crew members per foreman for a total of 210 depositions) reasonably balances the need to discover sufficient information to enable a statistically meaningful survey with the need to avoid undue burden. (Id. at ¶¶ 17, 21.)

1 The same arguments are almost certainly applicable to the tool reimbursement subclass.

IV. LEGAL STANDARD

Both parties have set forth the relevant law regarding discovery and, more specifically, the law relating to discovery of absent class members.

We start with the premise that the Federal Rules of Civil Procedure and the decisions of the Ninth Circuit Court of Appeals favor discovery to assist in the underlying goals of litigation. Pursuant to the Federal Rules, "[p]arties may obtain discovery regarding any nonprivileged matter that is relevant to any party's claim or defense" and this "[r]elevant information need not be admissible at trial if the discovery appears reasonably calculated to lead to the discovery of admissible evidence." Fed. R. Civ. P. 26(b)(1). The Ninth Circuit has explained that it favors a broad scope of discovery. "[W]ide access to relevant facts serves the integrity and fairness of the judicial process by promoting the search for truth." Epstein v. MCA, Inc., 54 F.3d 1422, 1423 (9th Cir. 1995); Dysthe v. Basic Research, L.L.C., 273 F.R.D. 625, 628 (C.D. Cal. 2011).

But the right to access information is not absolute and "[t]he court may, for good cause, issue an order to protect a party or person from annoyance, embarrassment, oppression, or undue burden or expense." Fed. R. Civ. P. 26(c)(1). "The party opposing disclosure has the burden of proving 'good cause,' which requires a showing 'that specific prejudice or harm will result' if the protective order is not granted." In re Catholic Archbishop of Portland Oregon, 661 F.3d 417, 424 (9th Cir. 2011) (quoting Foltz v. State Farm Mut. Auto. Ins. Co., 331 F.3d 1122, 1130 (9th Cir. 2003)).

No Supreme Court or Ninth Circuit case law addresses the propriety of conducting discovery on absent class members. See Tierno v. Rite Aid Corp., No. C-05-2520-TEH, 2008 U.S. Dist. LEXIS 112461, 2008 WL 2705089 (N.D. Cal. July 8, 2008) ("The law on discovery directed to absent class members is flexible. Discovery from absent class members is 'neither prohibited nor sanctioned explicitly' by the Federal Rules.") (Citation omitted.) However, courts often apply the standard articulated by the Seventh Circuit in Clark v. Universal Builders, Inc., 501 F.2d 324, 340-41 (7th Cir. 1974). This standard permits such discovery "only where the proponent of the discovery establishes" four criteria:

that (1) the discovery is not designed to take undue advantage of class members or to reduce the size of the class, (2) the discovery is necessary, (3) responding to discovery requests would not require the assistance of counsel, and (4) the discovery seeks information that is not already known by the proponent.

McPhail v. First Command Fin. Planning, Inc., 251 F.R.D. 514, 517 (S.D. Cal. 2008) (citing Clark, 501 F.2d at 340-42).

Other courts have articulated standards that are similar to, or overlap with the factors set forth in, Clark. For example, in McCarthy v. Paine Webber Group, Inc., 164 F.R.D. 309 (D. Conn. 1995), the court required parties seeking discovery to make a "strong showing . . . that the information sought (1) is not sought with the purpose or effect of harassment or altering membership of the class; (2) is directly relevant to common questions and unavailable from the representative parties; and (3) is necessary at trial of issues common to the class." Id. at 313 (citations omitted). It reasoned that "[d]iscovery of absent class members, while not forbidden, is rarely permitted due to the facts that absent class members are not 'parties' to the action, and that to permit extensive discovery would defeat the purpose of class actions which is to prevent massive joinder of small claims." Id. (citations omitted); see also McPhail, 251 F.R.D. at 517 n.3 (citing McCarthy, 164 F.R.D. at 313).

Similarly, in Tierno, the court held that the proponent must demonstrate three factors to justify discovery on absentee class members: (1) the information sought is relevant; (2) the information is not readily obtainable from the representative parties or other sources; and (3) the request is not unduly burdensome and made in good faith. See 2008 U.S. Dist. LEXIS 112461, at *6 (citing Cornn v. UPS, Inc., No. C-03-2001-TEH, 2006 U.S. Dist. LEXIS 69196, 2006 WL 2642540, *2 (N.D. Cal. Sept. 14, 2006).

Accordingly, the Court concludes that as a general rule, discovery from absent class members may be permitted when reasonably necessary, not conducted for an improper purpose, and not unduly burdensome in the context of the case and its issues.

V. ANALYSIS

A. Necessity of the Discovery

Plaintiffs assert that the discovery is not necessary because (1) Defendants had an opportunity to access the absent class members for a significant period prior to class certification, (2) the Court, in addressing issues regarding certification, has already addressed concerns over variability in the practice of class members, and (3) relevant case law prevents such discovery. (P&A in Support of Mot. at 8, ECF No. 333.)

1. Whether Discovery Could be Obtained by Other Means

Plaintiffs contend that the discovery is not necessary because Defendants had an opportunity to investigate and conduct discovery prior to certification. Plaintiffs present examples from other California district courts which have limited discovery of absent class members where alternative means of presenting relevant evidence existed. In McPhail v. First Command Fin. Planning, Inc., 251 F.R.D. 514 (S.D. Cal. 2008), a securities fraud class action, defendants sought interrogatory responses from each of the 178,527 absent class members as to whether they relied on alleged marketing misrepresentations by defendants. Id. at 518. Besides finding the request overly burdensome and an improper attempt to reduce the class size by, for example, seeking to dismiss non-responders, the court held that defendants could rebut the issue of reliance in other ways such as by showing that the marketing script did not contain misrepresentations or that salespersons did not provide uniform statements. Id.

In Tierno, 2008 U.S. Dist. LEXIS 112461, an employment misclassification class action, defendant sought to depose 100 absent class members before the court adopted a discovery plan. In reviewing the decision of the magistrate judge disallowing the depositions, the district court judge affirmed the decision, noting that "some appropriate form of discovery will be allowed" and that the decision was "without prejudice to more limited requests for discovery... made in the context of an agreed-upon discovery plan." Id. at *19.

Neither of these cases address the issues presented before this Court. Tierno does not support Plaintiff's contentions. The court there simply limited discovery until a discovery plan was in place, but espoused a clear intent to allow it. The McPhail court focused on the necessity of the proposed discovery and concluded that the evidence to be discovered could be addressed by other means. Here, Defendants argue persuasively that their discovery is necessary to challenge representative testimony Plaintiffs propose to present on liability.

In this case, both parties seek to determine which class members performed uncompensated pre-shift work. Plaintiffs do not attempt to answer the question with evidence that Defendants had a common policy requiring or allowing such work, but instead by showing that class members in fact performed such work. Given that there are an estimated 25,000 workers who could conceivably fall within the class, but also evidence that only some of them, working under the supervision of only some foremen, worked off the clock during only limited periods of time, sampling is necessary. Plaintiffs intend to conduct a random survey of a portion of the class members and then extrapolate the surveyed workers' responses to the class as a whole.

Because of the evidence of variations among foremen and time periods, Defendants contend that Plaintiffs' proposed survey will produce broad over-generalization across disparate groups. Defendants also argue that insofar as workers claiming compensation for unpaid work have an incentive to respond to Plaintiffs' survey and non-claimants do not, the results of Plaintiffs' survey will be biased. Defendants' concerns regarding reliability of such survey evidence are not taken lightly. Thus Defendants' seek to undertake what they and their expert believe is a more scientific and more reliable approach to the diverse worker groups so that they will be prepared to test the accuracy and reliability of Plaintiffs' survey results. Defendants assert that the discovery they propose is necessary to objectively inquire of various cross-sections of workers to determine issues of variability and whether certain subgroups of class members, for example, those who worked during certain time periods and for certain foremen, had more homogeneous experiences with regard to pre-shift work.

The Ninth Circuit has held that representative testimony may be used to assist in establishing liability in a class action. See Jimenez v. Allstate Ins. Co., 2014 U.S. App. LEXIS 17174 at 14 (9th Cir. Sept. 3, 2014) ("Since Dukes and Comcast were issued, circuit courts including this one have consistently held that statistical sampling and representative testimony are acceptable ways to determine liability..."). However real concerns with regard to reliability of statistical evidence exist. See Duran, 325 P.3d at 933 ("[A] statistical plan for managing individual issues must be conducted with sufficient rigor.").

The Court is unable to say that Defendants realistically could have discovered the relevant information from a representative sample of its 25,000 past and current seasonal workers other than, perhaps, through depositions. It would have been premature to conduct such discovery before the class was certified, and once it was certified Plaintiffs' presumably would have objected as they do now. Regardless, the Court finds that Defendants have made a persuasive case that it is necessary for them to conduct the discovery at this juncture and that such discovery is relevant to the reliability of sampling and the statistical study to be used to establish liability.

2. Whether Variability was Addressed During Certification

Plaintiffs contend that the Court addressed any concerns regarding variability in determining issues of certification. Unfortunately, arguments regarding variability in the amount of damages or whether liability exists have been intermingled. With regard to determining damages (once liability is established) "[t]he amount of damages is invariably an individual question and does not defeat class action treatment." Leyva v. Medline Indus., 716 F.3d 510, 514 (9th Cir. 2013) (citing Blackie v. Barrack, 524 F.2d 891, 905 (9th Cir. 1975)). However, liability is a different question and usually addressed on a class-wide basis. See Jimenez, 2014 U.S. App. LEXIS 17174 at 13 (citing In re Whirlpool Corp. Front-Loading Washer Prods. Liab. Litig., 722 F.3d 838, 853-55 (6th Cir. 2013)).

The Court did not resolve all issues of variability in its order on Defendants' motion for decertification. It did deny decertification of certain subclasses, but also noted in great depth its many concerns about the interpretation and reliability of the evidence presented. In light of those concerns, the Court explained that "Plaintiffs shall be required at trial to present concrete and reliable methods for determining liability and damages." (Decert. Order at 77.) That is still true. The parties are embarked on competing strategies for gathering the necessary evidence and for challenging the methods used. Nothing said by the Court previously during certification or otherwise was intended to, or does, limit discovery into issues of variability in determining liability.

3. Whether Prevailing Case Law Precludes Discovery by Defendants

Plaintiffs rely heavily upon the recent Ninth Circuit decision in Jimenez for the proposition that since liability should be tried as to the entire class, Defendants do not have the right to conduct discovery on absent class members to challenge liability. Instead, they argue that Defendants need only be given an opportunity to present individual issues with regard to damages after liability is established. (See P&A in Supp. Of Mot. at 12-14.)

Jimenez provides little insight into appropriate procedures for presenting or defending against statistical evidence at the liability stage. This is because Jimenez addresses certification in a case where plaintiffs sought to use the results of statistical sampling to justify certification. See Jimenez v. Allstate Ins. Co., 2012 U.S. Dist. LEXIS 65328 at 63-64 (C.D. Cal., Apr. 18, 2012). The district court granted certification and commented as follows regarding the potential use of statistical evidence to prove liability and damages:

As discussed above, the Court finds that the issues of liability are properly subject to class treatment, but has not yet determined how the damages phase of a trial might proceed. Thus, the Court has not been convinced at this time about the propriety of the use of statistical sampling to calculate damages. However, that is an issue that can be addressed in the future, and is not a sufficient basis to find a lack of superiority at the class certification stage. "The 'risk [that individual damage calculations will be unmanageable] is better addressed down the road, if necessary' by altering or amending the class, not by denying certification at the outset." 2 Newberg on Class Actions § 4:26 (4th ed.) (quoting In re Bally Mfg. Sec. Corp. Litig., 141 F.R.D. 262, 268 (N.D. Ill. 1992), order clarified, 144 F.R.D. 78 (N.D. Ill. 1992), aff'd, 2 F.3d 1456 (7th Cir. 1993)). Moreover, when compared to the prospect of separately trying approximately 1200 individual cases as to both liability and damages, the class process envisioned would be superior based on the present facts and circumstances.

Jimenez, 2012 U.S. Dist. LEXIS 65328 at 63-64. Upon appellate review, the Ninth Circuit affirmed, and found that the district court was correct to allow plaintiffs to present arguments in support of certification based on the use of statistical evidence to prove liability. In accepting that statistical evidence could be used, the Ninth Circuit held that "[t]he district court did not abuse its discretion by entering its class certification order." Jimenez, 2014 U.S. App. LEXIS 17174 at 19. Jimenez focused on the propriety of using survey evidence during certification. Beyond stating that it may be acceptable to use such evidence to prove liability, that court did not specify how or what procedure would be allowed post-certification. Plaintiffs read too much into Jimenez in suggesting it specifies how to proceed with discovery relating to statistical sampling used to prove or disprove liability.

Plaintiffs also argue that the California Supreme Court in Duran imposed limits on Defendants' right to conduct discovery with regard to statistical sampling. This Court does not find in that case any such limits on Defendants' right to present a defense relating to Plaintiffs' methods of statistical sampling. Duran, 59 Cal. 4th at 49. ("If the trial proceeds with a statistical model of proof, a defendant accused of misclassification must be given a chance to impeach that model...").

Plaintiffs assert that Defendants' pilot study is an attempt to obtain information on individual claims, not address class-wide issues relating to liability. Clearly, deposing an absent class member will reveal individual information with regard to that class member. Indeed, discovery performed for the primary purpose of addressing liability and damages could reveal issues regarding certification. See Gen. Tel. Co. of the Southwest v. Falcon, 457 U.S. 147, 160 (1982) ("Even after a certification order is entered, the judge remains free to modify it in the light of subsequent developments in the litigation.")

However, Defendants assert that their objective in deposing representatives from the various identified strata is to determine how to proceed with reliable, representative statistical testing. The Court has seen no persuasive evidence to the contrary. It appears the approach proposed has been designed and sanctioned by a seemingly well-qualified expert in the field. Indeed, conducting a pilot study or some other type of preliminary assessment appears to be a widely accepted method of proceeding. Duran, 59 Cal. 4th at 33 ("In general, when a trial plan incorporates representative testimony and random sampling, a preliminary assessment should be done to determine the level of variability in the class.")

Accordingly, the Court finds that as presented to date, and without foreclosing Plaintiffs' right to present new evidence to the contrary, the discovery requested appears reasonably tailored to obtain information regarding the reliability of statistical sampling. The Court's previous decision to allow Defendants to proceed with its proposed pilot study is well within the limits of the Court's authority over creation of a discovery plan.

B. The Burden Created by the Discovery

Plaintiffs assert that the pilot study should be prohibited because it imposes a large burden on absent class members and Plaintiffs' counsel.² Certainly the taking of some 200 depositions is a major undertaking not permitted without leave of court. See Fed. R. Civ. P. 26(b)(2), 31(a)(2). However, given the extraordinary characteristics of this case, including the large class size, the large amount in controversy, the issues with regard to credibility and reliability of the evidence provided to the Court during the certification phase, and the fact that it will be difficult, if not impossible, to locate and contact many of the absent class members,³ the Court is unwilling to limit a party's seemingly good faith, expertly designed, effort to seek relevant and reliable evidence to enlighten all regarding the need for and quality of representative testimony that may be offered. Absent evidence of abuse or undue burden on deponents, the Court finds that the benefit of the discovery outweighs the burden.

2 Plaintiffs also assert that the pilot study should be prohibited because it requires that which Clark criticized, namely, the assistance of counsel in responding. The Court finds this Clark factor really arose out of concern that the discovery would be overly burdensome. The Court here addresses that broader issue, but not Plaintiff's unfounded suggestion that Clark was condemning all discovery that necessitated Plaintiff counsel's participation.

3 The parties noted during the hearing that they lack reliable contact information for many of the class members, many of whom are migrant farm workers.

Unlike other cases where Defendants sought discovery from significantly larger numbers of (and at times, all) absent class members, here Defendants have limited their request to roughly 200 - less than one percent of the 25,000 total class members. Accordingly, more than 99% of the absent class members will suffer no imposition at all from the discovery. The two to four hours⁴ of time the 196 deponents will be asked to devote to the deposition is a relatively small time investment relative to a case of this size and one which should help provide clarity on the issue of liability to class members.

4 Defendants originally estimated that early depositions would take no more than four hours and that the time devoted to each likely would decrease as the attorneys grew familiar with the process and learned to deal with practical obstacles. After the first few days of attempted depositions, Defense counsel represented that those that did go forward were taking only about two hours each.
--------

Considering all of the foregoing, the Court concludes that the imposition which will result from Defendants' pilot study depositions, though not insignificant, is outweighed by the potential benefits to the process.

To the extent that personally appearing at a deposition is likely more time consuming and burdensome than replying to a written survey or interrogatories, the Court believes that given the complexities of this case and the population being questioned, depositions will ultimately prove to be a more efficient and reliable method of obtaining detailed information regarding absent class members' experiences.

Finally, the Court appreciates the time commitment such depositions will impose on counsel. However, the commitment is mutual, not inconsistent with the tens of millions of dollars Plaintiffs' counsel seeks to recover on behalf of the class, and given the rather large stable of counsel aligned with Plaintiffs, certainly doable.

C. Proper Purpose

As noted above, the Court has no reason to believe that Defendants' pilot study, and the depositions to be taken pursuant to it, are designed for any reason other than because they were rationally and legitimately believed by Defendants and their experts in good faith to be necessary to address the complex issues presented in this case. If at any time Plaintiffs discover evidence to the contrary, they may present it to the Court and ask for reconsideration.

The primary concern, at least as reflected in cases cited by Plaintiffs as denying discovery (see, e.g., McPhail, 251 F.R.D. at 518), seems to be that Defendants will seek to use non-responses to move to impose issue sanctions on, and indeed dismiss, non-responding class members. However, at the hearing, Defendants assured the Court that they will not seek to remove absent class members from the action for failure to appear at depositions. Further, the Court reserved Plaintiffs' right to seek relief if there was strong affirmative evidence of bad faith solicitation of non-responses. Based on the information presented to the Court, the discovery does not appear to be sought for an improper purpose.

VI. PLAINTIFFS' OBJECTIONS

On September 30, 2014, Plaintiffs filed objections to the declaration of Defendants' expert, Joseph A. Krock. (ECF No. 340.) Plaintiffs object that several statements in Krock's declaration do not meet standards of scientific reliability as required under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 582 (1993).

Plaintiffs' attempt to exclude expert testimony at this juncture is premature. Krock, in his declaration, makes statements that explain his reasoning behind proposing the pilot study. He does not, however, express an ultimate opinion on or relating to the statistical relevance of the data to be collected, the conclusions to be drawn therefrom, or otherwise. To the extent Plaintiffs' objections intend to challenge Krock's proposed pilot study technique or the principals upon which it is based, the Court finds the Daubert criteria met. Certainly, the concept and practice of stratified sampling is grounded in the "methods and procedures of science." Daubert at 590; see In re Countrywide Fin. Corp. Mortgage-Backed Secs. Litig. v. Countrywide Fin. Corp., 984 F. Supp. 2d 1021, 1036 (C.D. Cal. 2013) ("The Court's gate-keeping task is not so strict as to allow only the best possible scientific evidence, but to allow only reliable scientific evidence. Stratification is a widely accepted statistical technique."). The Court finds Krock's methodology for conducting a pilot study to determine the potential stratification of the class sufficiently reliable under Daubert.

This case is in the nascent stages of liability discovery. Plaintiffs retain the right, and will have the opportunity, to question or attack the reliability of Krock's methods and any opinions and conclusions he bases thereon. Their objections are denied without prejudice at this time.

VII. ORDER

Accordingly, it is hereby ORDERED that Plaintiff's Motion for Protective Order is DENIED. Further, Plaintiff's objections to the testimony of Joseph Krock are dismissed without prejudice. IT IS SO ORDERED.

Dated: October 10, 2014

/s/ Michael J. Seng

UNITED STATES MAGISTRATE JUDGE