Plaintiff Jane Doe No. 2 appeals the district court's March 9, 2017 Order granting Defendant City of Memphis summary judgment, and the district court's May 22, 2018 Order granting Defendant's motion to strike class allegations. For the reasons set forth below, we REVERSE the district court's judgment and REMAND for further proceedings consistent with this opinion.

BACKGROUND

Plaintiffs, Jane Doe No. 1, Jane Doe No. 2, and Jane Doe No. 3, are three women who allege that Defendant failed to submit for testing the sexual assault kits ("SAKs") prepared after their sexual assaults. Each Plaintiff separately reported a sexual assault to the Memphis Police Department ("MPD"), and a bodily fluid sample was taken from each Plaintiff and placed in an SAK. Plaintiffs allege that Defendant possessed over 15,000 such kits that it failed to submit for testing, resulting in spoliation. Plaintiffs sought to certify a class of women whose kits Defendant failed to test.

Discovery revealed that the SAKs of Jane Doe No. 1 and Jane Doe No. 3 were both tested soon after their assaults. However, Jane Doe No. 2's SAK, with which this appeal is concerned, was submitted for testing in 2013, and the results were received in 2014-eleven years after her assault in 2003. This appeal concerns only Jane Doe No. 2 and the purported class, inasmuch as Plaintiffs do not appeal the grant of summary judgment for Defendant on the claims of Jane Doe No. 1 or Jane Doe No. 3.

Plaintiffs filed a complaint on December 20, 2013, alleging various constitutional violations. Pursuant to Defendant's subsequently filed motion to dismiss Plaintiffs' claims, the district court dismissed with prejudice all of Plaintiffs' claims except their claims of sex discrimination in violation of the Equal Protection Clause. Discovery proceeded on Plaintiffs' remaining Equal Protection claim and lasted nearly two years.

During discovery, Defendant produced materials on MPD's policies, procedures, and training relating to sexual assault, investigative files for the three representative Plaintiffs, and affidavits by Major Don Crowe of the Special Victims Unit and Sergeant Margaret Houston of the Sex Crimes Division. Defendant also produced spreadsheets it had prepared containing data from 2001 to 2005 related to MPD's investigations into sexual assaults and other violent crimes. This information constituted 7,200 pages of spreadsheets detailing 15,465 criminal investigations. Defendant claims that producing this discovery cost the City over $1 million. Because the scope of discovery and the events surrounding the discovery process are the crux of this case, more details of the nearly two years of discovery that occurred will be provided below.

On January 25, 2016, Defendant moved for summary judgment on all of Plaintiffs' claims. On February 5, 2016, Defendant filed a Motion to Strike Class Allegations pursuant to Federal Rule of Civil Procedure 23(d)(1)(D). On March 9, 2017, the district court granted Defendant's motions for summary judgment as to Jane Doe No. 1, Jane Doe No. 2, and Jane Doe No. 3.

On March 15, 2017, Plaintiffs filed a Motion to Reconsider Order Granting Summary Judgment as to Jane Doe No. 2, and on May 17, 2018, the district court denied Plaintiffs' motion. On May 22, 2018, the district court granted Defendant's Motion to Strike Class Allegations.

Plaintiffs then appealed to this Court the district court's order granting Defendant summary judgment and its order granting Defendant's motion to strike class allegations. 1

DISCUSSION

Standard of Review

We review de novo a district court's grant of summary judgment. Spadafore v. Gardner , 330 F.3d 849 , 851 (6th Cir. 2003). Summary judgment is proper where there is no genuine issue as to any material fact and the moving party is entitled to judgment as a matter of law. Fed. R. Civ. P. 56(a). In considering such a motion, the court construes all reasonable factual inferences in favor of the nonmoving party. Matsushita Elec. Indus. Co. v. Zenith Radio Corp. , 475 U.S. 574 , 587, 106 S.Ct. 1348 , 89 L.Ed.2d 538 (1986). The central issue is "whether the evidence presents a sufficient disagreement to require submission to a jury or whether it is so one-sided that one party must prevail as a matter of law." Anderson v. Liberty Lobby, Inc. , 477 U.S. 242 , 251-52, 106 S.Ct. 2505 , 91 L.Ed.2d 202 (1986).

"This court reviews a district court's decision on a Rule 56(d) motion for discovery for an 'abuse of discretion.' " In re Bayer Healthcare & Merial Ltd. Flea Control Prods. Mktg. & Sales Practices Litig. , 752 F.3d 1065 , 1074 (6th Cir. 2014) (quoting United States v. Dairy Farmers of Am., Inc. , 426 F.3d 850 , 862 (6th Cir. 2005) ). "An abuse of discretion occurs when the reviewing court is left with the definite and firm conviction that the trial court committed a clear error of judgment." FTC v. E.M.A. Nationwide, Inc. , 767 F.3d 611 , 623 (6th Cir. 2014) (quoting United States v. Hunt , 521 F.3d 636 , 648 (6th Cir. 2008) ). To reverse, this Court must find that the district court's "ruling was arbitrary, unjustifiable, or clearly unreasonable." Id.

Analysis

While Plaintiffs' underlying claim pertains to Equal Protection Clause violations, Plaintiffs acknowledge that their primary argument "is the absence of a meaningful opportunity to conduct discovery" prior to the entry of summary judgment. (Appellants' Br. 20.) Plaintiffs do not argue in their briefs that the district court erred in striking class allegations based on the record before the court. Thus, we emphasize at the outset that all we are asked to consider is whether the district court erred in granting summary judgment and striking Plaintiffs' class allegations before Plaintiffs had a meaningful chance to pursue further discovery.

a. Sex Discrimination Claim

To the extent that the merits of Plaintiffs' underlying claims are relevant to our analysis of whether the district court abused its discretion in denying further discovery, we briefly note the elements of Plaintiffs' claim of sex discrimination in violation of the Equal Protection Clause.

For a plaintiff to succeed on an Equal Protection claim of sex discrimination, she "must show that she is a member of a protected class and that she was intentionally and purposefully discriminated against because of her membership in that protected class." Jones v. Union County , 296 F.3d 417 , 426 (6th Cir. 2002). The latter showing requires a demonstration that it is the policy or custom of Defendant to provide less protection to victims of sexual assault than those of other crimes, and that gender discrimination was the motivation for this disparate treatment. See id. Absent direct evidence of discriminatory motivation, "[a] discriminatory effect which is severe enough can provide sufficient evidence" of such motivation under appropriate circumstances. See United States v. Thorpe , 471 F.3d 652 , 661 (6th Cir. 2006) (quoting United States v. Tuitt , 68 F. Supp. 2d 4 , 10 (D. Mass. 1999) ).

To adequately review the claims made in this appeal relating to discovery, we must scrutinize the details of the discovery process. After a May 2014 scheduling order, Plaintiffs submitted their first set of interrogatories and requests for production of documents, which included, as relevant to Plaintiffs' claims on appeal, the following:

INTERROGATORY NO. 3: For each SAK collected by the Defendant during the years 1980 to the present, identify the following:

a) the name, gender, and race of the individual from whom the SAK was collected;

b) the date of collection of the SAK;

c) the crime being investigated;

d) the physical location(s), person/entity with custody of each SAK, dates that each SAK was in a particular location; and

e) any other identifying information used by the Defendant to distinguish between SAKs.

INTERROGATORY NO. 4: For each SAK identified in the preceding Interrogatory, state and identify the following:

a) whether and when the SAK has been tested for serology;

b) whether and when the SAK underwent forensic DNA analysis;

c) any reason that each SAK was or was not submitted for testing within 96 hours of recollection;

d) any reason that each SAK was submitted for testing on the date(s) it was submitted;

e) any individual or agency that requested and/or approved submission of the SAK for serology/DNA testing;

f) any individual or agency that declined to approve submission of the SAK for serology/DNA testing;

g) any cost associated with the testing of each SAK and what person/entity bore any cost identified;

h) and whether and when the suspect of the crime for which the SAK was collected and charged with a crime for which the SAK was collected.

* * *

INTERROGATORY NO. 5: For each DNA sample collected by the Defendant during the years 1980 to the present for crimes other than sexual assault , identify the following:

a) the name, gender, and race of the individual from whom the DNA sample was collected;

b) the date of collection of the DNA sample;

c) the crime being investigated;

d) the physical location(s), person/entity with custody of each DNA sample, dates that each DNA sample was in a particular location; and

e) any other identifying information used by the Defendant to distinguish between DNA samples.

INTERROGATORY NO. 6: For each DNA sample identified in the preceding Interrogatory, state and identify the following:

i) whether and when the DNA sample has been tested for serology;

j) whether and when the DNA sample underwent forensic DNA analysis;

k) any reason that each DNA sample was or was not submitted for testing within 96 hours of recollection;

l) any reason that each DNA sample was submitted for testing on the date(s) it was submitted;

m) any individual or agency that requested and/or approved submission of the DNA sample for serology/DNA testing;

n) any individual or agency that declined to approve submission of the DNA sample for serology/DNA testing;

o) any cost associated with the testing of each DNA sample and what person/entity bore any cost identified;

p) and whether and when the suspect of the crime for which the DNA sample was collected and charged with a crime for which the DNA sample was collected.

* * *

REQUEST NO. 18: Produce any and all documents and electronically stored information which relate to and/or pertain to your response to Interrogatory No. 3.

* * *

REQUEST NO. 19: Produce any and all documents and electronically stored information which relate to and/or pertain to your response to Interrogatory No. 4.

* * *

REQUEST NO. 20: Produce any and all documents and electronically stored information which relate to and/or pertain to your response to Interrogatory No. 5.

* * *

REQUEST NO. 21: Produce any and all documents and electronically stored information which relate to and/or pertain to your response to Interrogatory No. 6.

(R. 25-2, Plaintiffs' First Set of Discovery Requests, Page ID# 488-90, 500.) Request for Production Nos. 18 through 21 contemplate the production of the investigative files of the victims of sexual assault and other violent crimes, inasmuch as these files were the source material for Defendant's eventual responses to Interrogatory Nos. 3 and 4.

Defendant did not respond to this set of discovery requests by the June 2014 deadline to do so. Instead, Defendant sought to limit the scope of discovery via a motion for a protective order in August 2014, which the court denied in January 2015. Also in August 2014, Plaintiffs sought to compel responses to their discovery requests, which were months overdue.

In September 2014, the district judge allowed Plaintiffs leave to amend their complaint and noted that "Plaintiff's diligence is largely dependent on the depth of discovery, which is currently limited by Defendant." (R. 36, Order, Page ID# 366.)

On January 12, 2015, the court ordered Defendant to respond to Interrogatories 3 and 4 by January 31, 2015, but limited the scope of those interrogatories to SAKs collected between January 1, 1987 and June 1, 2014. The court did not directly address Plaintiffs' requests for production in this order, instead ordering the parties to attempt to resolve their disputes on their own and report back to the court as to the resolution of their dispute.

On January 30, 2015, Defendant moved to extend its deadline for responding to the discovery requests, and the court ordered that "Defendant shall have up to and including February 4, 2015 to properly respond to Plaintiffs' First Set of Interrogatories and Requests for Production of Documents." (R. 60, Order, Page ID# 615.) On February 4, Defendant provided an initial response to Plaintiffs' discovery requests. Responding to Interrogatory Nos. 3 and 4, Defendant referred Plaintiffs to a spreadsheet that it had created containing some information responsive to those requests. Defendant included a disclaimer that while "every attempt has been made to assure the accuracy of the information contained in the spreadsheet," "the only way to verify its accuracy is to review each investigative file." (R. 153-3, Responses to Interrogatories, Page ID# 9098.) The response also made clear that reviewing the investigative files for the spreadsheet was an ongoing process and that the spreadsheet would be updated accordingly. Defendant's response to Plaintiffs' Request for Production Nos. 18 and 19 also referred Plaintiffs to the spreadsheet. Defendant provided several supplemental responses to Plaintiffs' discovery requests from April 2015 2 through August 2015, but never suggested that these supplements made its responses to Plaintiffs' requests complete.

At a July 9, 2015 status conference, Plaintiffs complained that discovery had not progressed, and the Court admonished Defendant to provide Plaintiffs with responses to their discovery requests.

On August 31, 2015, a status conference was held on Plaintiffs' August 2014 motion to compel. Plaintiffs agreed to withdraw their motion, as some documents had been produced and production was ongoing. By September 2015, Defendant estimated that its responses to Plaintiffs' discovery requests could be completed by May 2016.

In a December 2015 status conference, Defendant acknowledged that its discovery responses remained incomplete and reiterated its estimate that it could complete those responses by May 2016. Yet the following month Defendant moved for summary judgment. It was only after the motion for summary judgment was filed that Defendant produced three of the five years of spreadsheet information that constituted the bulk of its responses to Plaintiffs' discovery requests.

Moreover, it was not until February 2016 that Defendant first produced the complete investigative file of Jane Doe No. 2. It was also in February 2016 that Plaintiffs filed their Rule 56(d) affidavit, as an attachment to their response in opposition to summary judgment. That affidavit complained, as relevant to this appeal, that Defendant had not "allowed the Plaintiffs to review and inspect source documents related to the members of the class." (R. 103-2, Affidavit, Page ID# 1089.) The affidavit argued that summary judgment was improper because Defendant had failed to produce, among other things, "[s]ource investigatory documents in sexual assault cases in which SAKs were created which contain proof of discriminatory and/or stereotypical law enforcement practices." (R. 103-2, Page ID# 1090.)

Turning to the main issue on appeal-the district court's denial of further discovery-when a party moves for summary judgment under Federal Rule of Civil Procedure 56, Rule 56(d) allows the "nonmovant [to] show[ ] by affidavit ... that, for specified reasons, it cannot present facts essential to justify its opposition," and upon such showing, "the court may: (1) defer considering the motion or deny it; (2) allow time to obtain affidavits or declarations to take discovery; or (3) issue any other appropriate order." Fed. R. Civ. P. 56(d). 3

The purpose behind Rule 56(d) is to ensure that plaintiffs receive " 'a full opportunity to conduct discovery' to be able to successfully defeat a motion for summary judgment." Ball v. Union Carbide Corp. , 385 F.3d 713 , 719 (6th Cir. 2004) (quoting Anderson v. Liberty Lobby, Inc. , 477 U.S. 242 , 257, 106 S.Ct. 2505 , 91 L.Ed.2d 202 (1986) ). "A party invoking [the] protections [of Rule 56(d) ] must do so in good faith by affirmatively demonstrating ... how postponement of a ruling on the motion will enable him ... to rebut the movant's showing of the absence of a genuine issue of fact." E.M.A Nationwide, Inc. , 767 F.3d at 623 (quoting Willmar Poultry Co. v. Morton-Norwich Prods., Inc., 520 F.2d 289 , 297 (8th Cir. 1975) ). The affidavit must "indicate to the district court [the party's] need for discovery, what material facts it hopes to uncover, and why it has not previously discovered the information." Ball , 385 F.3d at 720 (quoting Cacevic v. City of Hazel Park , 226 F.3d 483 , 488 (6th Cir. 2000) ).

Moreover, "[t]he party opposing a motion for summary judgment ... possesses no absolute right to additional time for discovery under Rule 56." Emmons v. McLaughlin , 874 F.2d 351 , 356 (6th Cir. 1989). "A district court does not abuse its discretion in denying discovery when the discovery requested would be irrelevant to the underlying issue to be decided." Bayer Healthcare , 752 F.3d at 1074 (quoting Dairy Farmers of Am., Inc. , 426 F.3d at 862 ). Similarly, a district court has "discretion to limit the scope of discovery where the information sought is overly broad or would prove unduly burdensome to produce." Id. (quoting Info-Hold, Inc. v. Sound Merch., Inc., 538 F.3d 448 , 457 (6th Cir. 2008) ).

Although our standard of review is abuse of discretion, this Court has cited approvingly other circuits' view that "[a] ... motion requesting time for additional discovery should be granted almost as a matter of course unless the non-moving party has not diligently pursued discovery of the evidence." E.M.A. Nationwide, Inc. , 767 F.3d at 623 n.7 (second alteration in original) (internal quotation marks omitted) (quoting Convertino v. U.S. Dep't of Justice , 684 F.3d 93 , 99 (D.C. Cir. 2012) ); see also id. ("[D]istrict courts should construe motions that invoke [ Rule 56(d) ] generously, holding parties to the rule's spirit rather than its letter." (quoting Resolution Trust Corp. v. N. Bridge Assocs., 22 F.3d 1198 , 1203 (1st Cir. 1994) ).

This Court has made clear that in reviewing a district court's ruling on a motion for further discovery, there are five factors to consider:

(1) when the appellant learned of the issue that is the subject of the desired discovery; (2) whether the desired discovery would have changed the ruling below; (3) how long the discovery period had lasted; (4) whether the appellant was dilatory in its discovery efforts; and (5) whether the appellee was responsive to discovery requests.

CenTra, Inc. v. Estrin , 538 F.3d 402 , 420 (6th Cir. 2008) (quoting Plott v. Gen. Motors Corp. , 71 F.3d 1190 , 1196-97 (6th Cir. 1995) ).

As an initial matter, it does not appear that the district court considered all five of the Plott factors, as it never acknowledged them. In deciding Defendant's motion for summary judgment, the district court did not specifically rule on whether Plaintiffs had a sufficient opportunity to conduct discovery. Addressing Plaintiffs' motion for reconsideration, the district court rejected Plaintiffs' argument that they had an inadequate opportunity to conduct discovery, but primarily relied on the length of time discovery was open and purported deficiencies in Plaintiffs' counsel's affidavit. The district court also minimized Defendant's noncompliance with discovery requests, reasoning that the noncompliance was primarily a result of administrative burden. In its ruling on Defendant's motion to strike class allegations, the district court again rejected Plaintiffs' arguments for the same reasons, but further explained that no amount of additional discovery would allow Plaintiffs to sufficiently demonstrate commonality.

We will consider each factor.

1) Whether Plaintiffs Were Dilatory in Their Discovery Efforts

Although this is the third factor listed in Plott , we have stated elsewhere that our "main inquiry is 'whether the moving party was diligent in pursuing discovery.' " E.M.A. Nationwide, Inc. , 767 F.3d at 623 (quoting Dowling v. Cleveland Clinic Found. , 593 F.3d 472 , 478 (6th Cir. 2010) ). 4 We thus examine this factor first and ask whether Plaintiffs were diligent or dilatory in their discovery efforts, a factor the district court did not specifically analyze.

Plaintiffs filed their first set of discovery requests just a week after the scheduling order was entered. Defendant acknowledges that "Plaintiffs fully participated in the Court's status conferences." (Appellee's Br. 40.) However, Defendant argues that Plaintiffs were not diligent in pursuing discovery inasmuch as they "withdrew their Motion to Compel in September 2015, and despite the Court's invitation [in its September 8, 2015 Order], never filed a more focused and targeted motion to identify the specific, current issues regarding any discovery responses." (Appellee's Br. 39-40 (referencing R. 84, Order, Page ID# 769).)

Plaintiffs respond that the motion that the court invited Plaintiffs to file "was not to compel the City to respond to discovery but rather was to be filed by Plaintiffs if needed - after discovery was produced - to address any deficiencies." (Reply Br. 6.) Plaintiffs argue that their withdrawal of their motion to compel did not affect the court's January 12, 2015 Order requiring Defendant to produce information to Plaintiffs. Indeed, the court's September 8, 2015 Order stated that a new motion to compel could be filed "should the circumstances require it." (R. 84, Order, Page ID# 769.) This order did not remove any obligations from Defendant to produce information, and Defendant in fact continued to produce information after this order. Plaintiffs argue convincingly that because Defendant was producing some documents during the year before Defendant moved for summary judgment, and because the court was overseeing this production through regular status conferences, it was reasonable for Plaintiffs not to file further motions to compel. Further, at status conferences-including one as late as December 2015-Defendant estimated that responding in full to Plaintiffs' discovery requests would take until May 2016. Plaintiffs apparently relied on Defendant's estimate and could not have known that Defendant would instead move for summary judgment in January of that year. Plaintiffs argue that they reasonably believed the discovery process would go on several months longer than it did and that they cannot be faulted for being unable to foresee that their chance to file a second motion to compel would be cut short.

There is no evidence that Plaintiffs delayed discovery or were not diligent in pursuing discovery or choosing not to file a second motion to compel. The record instead indicates that Plaintiffs were diligent in pursuing discovery. Thus, this factor-which represents our "main inquiry" in this type of case-favors Plaintiffs. E.M.A. Nationwide, Inc. , 767 F.3d at 623 .

2) When Plaintiffs Learned of the Issue that Is the Subject of the Desired Discovery

This factor primarily pertains to situations where there was something that prevented a party from learning about a subject of desired discovery until after some discovery had already been sought. See Woods v. McGuire , 954 F.2d 388 , 391 (6th Cir. 1992) (considering an argument that a case decided late in the discovery period made more discovery necessary by changing the law applicable to the plaintiff's claim). Because the desired discovery in this case was sought in Plaintiffs' first set of discovery requests, this factor is not applicable to the facts in this case and is therefore neutral.

3) Whether the Desired Discovery Would Have Changed the Ruling Below

This factor favors Plaintiffs. Plaintiffs seek additional discovery relating to the individual investigative files, as well as statistics, deposition testimony, training documents, internal affairs documents, and internal correspondence. Defendant argues that because Plaintiffs received Jane Doe No. 2's investigative file and have not alleged that the file evidenced discriminatory intent, further discovery would not change the ruling. However, this was a premature conclusion.

To make out their Equal Protection claim, Plaintiffs must show that it was Defendant's policy to provide less protection to victims of sexual assault than those of other violent crimes and that this was motivated by gender discrimination. Jones , 296 F.3d at 426 . Just because gender animus was not present on the face of Jane Doe No. 2's investigative file does not mean that it would not be evident upon a review of other sexual assault victims' investigative files. A plaintiff can prove animus by direct or circumstantial evidence. See Washington v. Davis , 426 U.S. 229 , 242, 96 S.Ct. 2040 , 48 L.Ed.2d 597 (1976) ("Necessarily, an invidious discriminatory purpose may often be inferred from the totality of the relevant facts, including the fact, if it is true, that the law bears more heavily on [a protected class]."). What appears in the spreadsheets might not faithfully reflect, or give a full picture of, the sort of investigation that sexual assault crimes received as compared with other violent crimes. Indeed, Defendant admitted that "the only way to verify [the spreadsheets'] accuracy is to review each investigative file." (R. 153-3, Responses to Interrogatories, Page ID# 9098.)

As an example of how further discovery might change the outcome below, if Defendant had produced a randomly selected, representative sample of investigative files, it might be evident that much more time was spent investigating other violent crimes than was spent on sexual assault investigations-which might be seen in the level of detail included in the investigative file even though it might not appear in Defendant's spreadsheet purporting to summarize what was in the files. Further evidence of animus might also be found in emails not yet produced and if Plaintiffs were permitted to depose the officers involved in those investigations. Plaintiffs cannot be faulted for not seeking depositions earlier, as it is not unusual to await substantial completion of document production prior to conducting depositions. 5

Finally, Defendant argues that Plaintiffs' Rule 56(d) motion was properly denied because it did not "identify any specific facts relating to alleged discriminatory animus that counsel hopes to uncover." (Appellee's Br. 43.) It is true that a party cannot avoid summary judgment by filing an affidavit which does not "substantiate ... allegations of harassment with any details" and which suffers from a "lack of specificity, without any apparent justification." Emmons v. McLaughlin , 874 F.2d 351 , 357 (6th Cir. 1989). In contrast with some affidavits this Court has found deficient, see, e.g. , Cacevic v. City of Hazel Park , 226 F.3d 483 , 489 (6th Cir. 2000), Plaintiffs' affidavit outlined various types of evidence that would help establish the single dispositive issue: whether Defendant's acknowledged failure to test (or timely test) thousands of SAKs was the result of discrimination on the basis of sex. Further, even assuming the affidavit lacked sufficient specificity, Plaintiffs' Rule 56(d) affidavit contains justification for its lack of details and specificity. The affidavit stated that "although the City has produced spreadsheets purportedly summarizing some [SAKs] and their status, demographic factors, and other factors, the City has not ... allowed the Plaintiffs to review and inspect source documents related to the members of the class." (R. 103-2, Affidavit, Page ID# 1089.)

Without at least a representative sample of investigative files, Plaintiffs cannot be punished for being unable to make more specific allegations about what the files will prove. Because additional investigative files, as well as other discovery Plaintiffs sought but were denied, might well have changed the outcome below, this factor weighs in favor of Plaintiffs.

4) How Long the Discovery Period Had Lasted

The district court emphasized that "the parties conducted expensive and rigorous discovery for a significant period of time-nearly two (2) years." (R. 144, Order Denying Plaintiff's Motion to Reconsider, Page ID# 8887.) However, what constitutes a reasonable length of time for the duration of discovery is so particular to the facts and circumstances of a given case that examining what lengths of time this Court has found sufficient for discovery in the past in not particularly helpful. This Court has stated that a discovery period lasting nearly five months was "a sufficient amount of time ... to conduct some discovery" and therefore the factor did not favor either party. E.M.A. Nationwide, Inc. , 767 F.3d at 625 ; see also Emmons , 874 F.2d at 359 n.8 ("We do not believe that nine months before entry of summary judgment is necessarily too short a period in which to expect discovery to be completed in a [constitutional tort involving alleged police misconduct].").

A discovery period of nearly two years is certainly somewhat lengthy. See Woods , 954 F.2d at 391 (holding there was no abuse of discretion where district court closed discovery almost two years after tort case involving the Westfall Act was filed). However, a claim in which Plaintiffs must prove discriminatory motivation, and might seek to do so using large quantities of data demonstrating a discriminatory effect so severe as to be evidence of discriminatory motivation, see United States v. Thorpe , 471 F.3d 652 , 661 (6th Cir. 2006), will be entitled to a longer discovery period than claims that are easier to prove.

Moreover, for the reasons discussed below, the length of this delay is in large part attributable to Defendant's delays.

This factor is thus largely neutral.

5) Whether the Appellee Was Responsive to Discovery Requests

This factor weighs in favor of Plaintiffs because Defendant was generally unresponsive to Plaintiffs' discovery requests. Defendant missed discovery deadlines, promised to complete its responses to Plaintiffs' initial discovery requests by May 2016, and then instead moved for summary judgment in January of that year. These facts, as well as the fact that Defendant continued to provide Plaintiffs with new information after filing for summary judgment, strongly suggest that Defendant may have purposefully drawn out the discovery process in order to secure summary judgment before having to respond in full to Plaintiffs' interrogatories and requests for production.

Defendant missed the initial June 2014 deadline to respond to Plaintiffs' first set of discovery requests and still had not responded to these requests by August 2014. Defendant was then ordered to respond to certain interrogatories by January 31, 2015, but Defendant moved for an extension the day before its responses were due.

Up until December 2015, Defendant assured Plaintiffs and the court that it would be able to complete its discovery responses by May 2016. However, in January 2016, Defendant moved for summary judgment.

Plaintiffs note that Defendant produced "approximately one hundred ... pages from the investigatory files of Jane Doe No. 1 and Jane Doe No. 2 ... for the first time one (1) week after the City filed its Motions for Summary Judgment." (R. 103-2, Affidavit, Page ID# 1090.) It was also not until after Defendant had moved for summary judgment that it produced three of the five years of spreadsheet information. Thus, a major portion of what Defendant produced in response to Plaintiffs' discovery requests was not turned over until after it had filed for summary judgment.

Although Defendant claims that it spent over $1 million in producing the spreadsheets, it is odd that Defendant would expend such a large amount of money producing spreadsheets, rather than either offering Plaintiffs a representative sample of investigative files that Plaintiffs would bear the cost of reviewing or otherwise negotiating with Plaintiffs about a less costly way to satisfy Plaintiffs' discovery requests. Defendant's expenses were ostensibly for the purpose of answering Interrogatory Nos. 3 through 6; however, the record does not indicate that Defendant ever attempted to respond to Request for Production Nos. 18 through 21, other than with the spreadsheets. It does not appear from the record why Defendant seemed to believe the source material for its responses to Interrogatory Nos. 3 through 6 (namely, the investigative files) would not be subject to Request for Production Nos. 18 through 21. Defendant offered as excuses for its delay in producing the spreadsheets the fact that the case was data-heavy and providing the information involved reviewing both paper and electronic documents, which were stored on different electronic systems because the systems had been updated over time. However, Defendant offered no valid excuse for its failure to produce the documents Plaintiffs requested; indeed, allowing Plaintiffs to review a representative random sample of some of the investigative files could have significantly reduced Defendant's time and effort.

Defendant's delay in producing discovery suggests that it could have been hoping to obtain summary judgment before having to comply in full with Plaintiffs' discovery requests. Defendant's repeated assurances that it would comply by May 2016 before it moved for summary judgment in January 2016 further suggests the possibility of a strategic delay motivated by a desire to deprive Plaintiffs of a full opportunity for discovery.

In any event, Defendant was not responsive to Plaintiffs' discovery requests, and this factor weighs for Plaintiffs.

In sum, Plaintiffs were moderately diligent in pursuing discovery, although somewhat blameworthy in relying on Defendant's representations that discovery would be forthcoming. Defendant, on the other hand, was even more blameworthy than Plaintiffs due to its unreasonable delays in producing the discovery material. And it is evident that additional discovery might have changed the outcome below. Together, this suffices to convince this Court that Plaintiffs did not receive " 'a full opportunity to conduct discovery' to be able to successfully defeat a motion for summary judgment." Ball v. Union Carbide Corp. , 385 F.3d 713 , 719 (6th Cir. 2004) (quoting Anderson v. Liberty Lobby, Inc. , 477 U.S. 242 , 257, 106 S.Ct. 2505 , 91 L.Ed.2d 202 (1986) ). It is true that discovery has lasted an extended time and cost Defendant some amount of money; however, as the above analysis illustrates, it is unclear why much of this time and cost was not avoided, and expenditures of time and money alone do not justify terminating discovery where a plaintiff has been diligent and may still discover information that could establish a genuine issue of material fact.

Therefore, this Court finds that the district court abused its discretion in granting Defendant summary judgment and striking Plaintiffs' class allegations before there was a full opportunity to conduct discovery. We remand with instructions that the district court allow discovery to proceed. We note that, rather than requiring a complete production of all of the source files, it may be more appropriate to allow Defendant to produce a sufficient representative sampling of the investigative files or permit Plaintiffs to review a representative sampling of these files. After that, the district court should allow additional discovery as necessary in further proceedings consistent with this opinion.

b. Class Certification Claim

Because of our conclusion that additional discovery might establish a genuine issue of material fact as to sex discrimination against Jane Doe No. 2 by showing that Defendant had a pattern or practice of not taking sexual assault crimes as seriously as other violent crimes, we also hold that additional discovery might also allow Plaintiffs to identify and certify a class.

As discussed above, the investigative files might reveal evidence of gender-motivated discrimination. If so, this evidence could allow Plaintiffs to find further class representatives and could ultimately result in certification of a class.

To certify a class, a party must demonstrate that four prerequisites are met:

(1) the class is so numerous that joinder of all members is impracticable;

(2) there are questions of law or fact common to the class;

(3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and

(4) the representative parties will fairly and adequately protect the interests of the class.

Fed. R. Civ. P. 23(a). Rule 23(b) places an additional requirement on putative class representatives that is inapplicable to this appeal.

The district court struck the class allegations because "discovery thus far does not, and additional discovery would not, allow Plaintiffs to sufficiently demonstrate [the] commonality" requirement of Rule 23(a)(2). (R. 145, Order Granting Defendant's Motion to Strike Class Allegations, Page ID# 8911.) The district court summarized that "given the discretion of different investigators, the discretion in varying supervisor chains, and the multitude of reasons for not testing or timely testing a SAK, Plaintiffs cannot show a common mode of exercising discretion in this case." (R. 145, Page ID# 8918-19.)

In support of its holding, the district court cited Pilgrim v. Universal Health Card, LLC , in which this Court affirmed a district court's order striking class allegations due to the class's inability to meet the predominance requirement of Rule 23(b). 660 F.3d 943 , 946 (6th Cir. 2011). In rejecting the plaintiffs' request for further discovery on the certification issue, this Court noted, "Their claims are governed by different States' laws, a largely legal determination, and no proffered or potential factual development offers any hope of altering that conclusion, one that generally will preclude class certification." Id. at 949 . The instant case is distinguishable from Pilgrim , however, in that there is no choice of law issue, nor any other sort of "prototypical factual issue that will vary from place to place and from region to region" that the Pilgrim Court emphasized. Id. at 948 . The district court noted the discretion individual investigators have in deciding whether to submit a SAK for testing and compared this with the Supreme Court's statement in Wal-Mart Stores, Inc. v. Dukes that absent the plaintiffs establishing a "common mode of exercising discretion that pervades the entire company," the commonality requirement of Rule 23(a) could not be met. 564 U.S. 338 , 356, 131 S.Ct. 2541 , 180 L.Ed.2d 374 (2011). This is true, and the similarities to the instant case are evident; however, the key distinction is that in Wal-Mart Stores, Inc. , it was the plaintiffs who moved for class certification, and the motion was filed after they had an opportunity for meaningful discovery. Id. at 346 , 131 S.Ct. 2541 . By contrast, in this case it was Defendant who moved to strike class allegations, and the motion was filed before Plaintiffs had an opportunity for meaningful discovery-making Wal-Mart Stores, Inc. inapposite. Based on the record before us, Plaintiffs may be unable to satisfy the commonality requirement; however, on remand further discovery could provide Plaintiffs with evidence of commonality and could result in certification of a class.

The district court prematurely concluded that no amount of discovery could possibly demonstrate commonality in this case. The above quote from Wal-Mart Stores, Inc. implies as much, as the Court suggests that a "common mode of exercising discretion that pervades the entire company" would suffice to prove commonality. Indeed, Plaintiffs' theory of the case largely turns on whether there was a pattern or practice of taking sexual assault crimes less seriously than other violent crimes. In reviewing the investigative files and additional evidence, Plaintiffs on remand will be free to pursue their class allegations, as the additional discovery might satisfy the commonality requirement of Rule 23(a)(2).

CONCLUSION

For the reasons set forth above, we REVERSE the district court's grant of Defendant's motion for summary judgment as to Jane Doe No. 2 and motion to strike class allegations, and REMAND for further proceedings consistent with this opinion.

DISSENT

I dissent for three reasons. First, because the majority does not accord due deference to the district court's interpretation of the facts. Second, because the majority incorrectly analyzes the Plott factors of whether Doe was dilatory and whether Doe's desired discovery would have changed the district court's ruling. And third, because the majority's instructions on remand go too far, mandating that the district court permit discovery, rather than instructing the district court to consider whether additional discovery is warranted in light of its holding.

I

We may find that the district court abused its discretion if its "ruling was arbitrary, unjustifiable, or clearly unreasonable." F.T.C. v. E.M.A. Nationwide, Inc. , 767 F.3d 611 , 624 (6th Cir. 2014). This discretion must at least mean that we will not second guess the district court's judgment and reweigh the Plott factors on appeal. We should constrain our review to determining whether the district court's conclusions are supported by the record and correct as a matter of law. See Curtiss-Wright Corp. v. Gen. Elec. Co. , 446 U.S. 1 , 10, 100 S.Ct. 1460 , 64 L.Ed.2d 1 (1980) (explaining that in reviewing a district court's decision to certify a question for appeal for an abuse of discretion "the proper role of the court of appeals is not to reweigh the equities or reassess the facts but to make sure that the conclusions derived from those weighings and assessments are juridically sound and supported by the record."). Instead, the majority effectively applies de novo review to the district court's decision, conducting its own analysis, reweighing the Plott factors, and dictating how discovery should proceed.

While the district court did not formally consider each Plott factor, its analysis is sufficiently robust to warrant deference. As the majority points out, the district court neither acknowledged the Plott factors nor reviewed each in a formal manner. A review of the opinion, however, shows that the district court considered three of the five factors: whether the desired discovery would change its ruling, how long the discovery period lasted, and whether the City of Memphis (the City) was responsive to discovery requests. See Plott v. Gen. Motors Corp., Packard Elec. Div. , 71 F.3d 1190 , 1196-97 (6th Cir. 1995). Presumably, it found the combination of those three factors dispositive. The district court did not address whether Doe was dilatory and when Doe learned of the issue, the latter of which the majority concludes is not applicable. The district court's failure to address the dilatory Plott factor could warrant reversal if none of the Plott factors the district court analyzed are dispositive. As I discuss below, the appropriate instruction on remand would be for the district court to reweigh the Plott factors taking the dilatory factor into account, not mandating discovery.

The majority errs in making a credibility determination to support its conclusion that the responsiveness Plott factor-whether the City was responsive to discovery requests-weighs in Doe's favor. Id. at 1196 . The district court noted that "to the extent Defendant was non-compliant with discovery requests, the record show[s] that such difficulties were largely due to the administrative burden of producing thousands of documents spanning multiple years." R. 144, at PID 8889. Instead of deferring to the district court, the majority effectively makes a credibility determination that the City's excuses for its delays in producing some requested discovery were not valid. The majority does not explain why it believes the district court's conclusions were unwarranted. Instead the majority speculates that the City's actions suggest some strategic motive to deprive Doe of discovery. Although it is possible to interpret the City's actions as strategic, the record also supports the district court's conclusion that the City's noncompliance was due to a significant administrative burden. Where the facts support multiple interpretations, we should not impose our own over the district court's.

The majority similarly errs in imposing its own interpretation of the facts to weigh the length-of-discovery Plott factor without explaining why the district court lacked support in the record for its determination. The district court explained that Doe had adequate time for discovery because, "[p]rior to the Order granting Defendant's Motion for Summary Judgment, the parties conducted expensive and rigorous discovery for a significant period of time-nearly two (2) years." Id. at 8887. The majority acknowledges that what constitutes a reasonable length of time for discovery is highly particularized to the facts and circumstances of the case. Instead of deferring to the district court, however, the majority decides that proving discriminatory motivation requires large quantities of data and therefore that the length-of-discovery Plott factor is neutral. Determining whether a certain case is one in which the scope of the claim requires more than two years of discovery is precisely the kind of particularized fact determination that should be left to the district court.

II

In addition to erring in reweighing the Plott factors instead of giving the district court its due deference, the majority errs in its analysis of two of the Plott factors: (1) whether Doe was dilatory in seeking discovery and, (2) whether Doe's requested discovery would have changed the district court's ruling.

1. Whether Doe Was Dilatory in Seeking Discovery is Not This Court's "Main Inquiry"

I agree that Doe was not dilatory in seeking discovery. I disagree, however, with the significance the majority gives that fact. We have held that when reviewing the decision to stay discovery, "[t]his Court's main inquiry is 'whether the moving party was diligent in pursuing discovery.' " E.M.A. Nationwide , 767 F.3d at 623 (citation omitted). Diligence is determined through an analysis of all the Plott factors, of which a party's dilatoriness is just one factor. See Dowling v. Cleveland Clinic Found ., 593 F.3d 472 , 478 (6th Cir. 2010) ("The overarching inquiry in these [ Plott ] factors is whether the moving party was diligent in pursuing discovery."). The majority, however, equates finding that Doe was not dilatory with finding that Doe was diligent. The majority's error is understandable. In common parlance, finding someone to not be dilatory might be the same as saying that they are diligent. Our caselaw, however, makes clear that diligence in pursuing discovery is a term of art. Simply because Doe was not dilatory does not mean, so far as the Plott factors are concerned, that she was diligent. The majority is incorrect to attribute the significance it does to the fact that Doe was not dilatory.

2. Doe's Desired Discovery Would Not Have Changed the District Court's Decision

I also disagree with the majority's conclusion that Doe's desired discovery would have changed the district court's decision. Doe fails to offer any reason why she believes that most of the additional discovery identified in her Rule 56(d) affidavit exists. Much of the material Doe requests rests only on conclusory allegations that documents containing discriminatory practices exist and have been withheld from production. For example, Doe alleges that the City has emails detailing discriminatory practices that Doe has not yet received. Doe does not, however, contend that the City has failed to turn over emails generally, or provide any explanation for why she believes emails containing discriminatory practices exist. Of the discovery Doe identifies in her Rule 56(d) affidavit that is not based on conclusory allegations, that material has already been substantially produced. The majority focuses on the City's failure to turn over "investigative files," the provision of spreadsheets in response to interrogatory requests with the caveat that they were subject to verification by the source material, emails that were allegedly not produced, and the lack of depositions, so I address those with specificity.

Doe never requested the City's "investigative files," so the majority's speculation as to what Doe might be able to prove with the investigative files is unwarranted. All Doe asked for in the requests for production and Rule 56(d) affidavit was the source material used to respond to certain interrogatory requests. Those interrogatory requests, and thus the requested documents, included specific information about each DNA sample that had been collected for sexual assault and violent crimes other than sexual assault. For instance, Doe sought the name, gender, and race of the individual; the collection date; and the crime investigated. For sexual assault crimes, Doe also sought any reason why the sexual assault kit was or was not submitted for testing within 96 hours of collection. The City provided all of that information, albeit for a limited range of years, in spreadsheets as part of its response to Doe's interrogatory requests.

Certainly, some of the source documents used to answer the interrogatory requests would be contained in investigative files. But none of the interrogatory requests, the requests for production, or Doe's Rule 56(d) affidavit is sufficiently broad in scope to require the City to turn over information or documents in investigative files that were not the source materials used to respond to the interrogatory requests. Doe never even defines the term "investigative file." If Doe had actually requested the investigative files it would be reasonable for her to define the term so that the City would know what Doe's request encompassed. The closest Doe's Rule 56(d) affidavit gets to requesting the investigative files is in subparagraph (g) of paragraph 10. There, Doe lists as documents that have yet to be produced "investigatory documents in sexual assault cases ... which contain proof of discriminatory and/or stereotypical law enforcement practices." R. 103-2, at PID 1090. As an initial matter, Doe's request is conclusory, merely assuming that investigative documents containing proof of discrimination exist in investigative files. Further, Doe's request does not ask for all investigative documents for violent crimes. Nor does Doe's request encompass the entire investigative file. Instead, Doe's request is limited to only documents in investigative files containing proof of discriminatory and or stereotypical law enforcement practices.

Even if Doe received the requested investigative file documents, Doe would still not receive a statistically meaningful sample of investigative files from which she could draw inferences. By receiving only what was requested-the specific pages or parts of pages with the source material relevant to her interrogatory requests-Doe would have nothing more in substance than she did before the Rule 56(d) motion. Doe could not, for example, determine whether it was evident that more time was spent investigating other violent crimes than sexual assault crimes because she could not analyze the number of documents in each file and would not receive source material detailing the time and resources used to investigate the various crimes. In sum, there is no basis for the majority to conclude that the investigative files could change the district court's ruling.

Neither would production of the source materials used for the City's interrogatory spreadsheet responses change the district court's ruling. The majority makes much of the caveat accompanying the City's interrogatory response spreadsheets that the information was subject to verification with the source material. Using this statement to conclude that the source material could change the district court's ruling, however, goes too far. First, the district court did not indicate that the caveat accompanying the spreadsheets impacted its decision. Therefore, even were the information verified with the source material, the verification itself would not change the district court's ruling. Second, Doe does not contend that the spreadsheets themselves failed to answer the interrogatory requests or were unreliable. Instead, Doe is unsatisfied with the spreadsheets because they did not cover the entire period from 1980 to the date of the request. Doe contends that there has been on-going discrimination since 1980. So, while adding additional years of data might impact the extent of her claim, the availability of only some years of data should not affect whether she can make out a claim in the first place. Finally, even if the spreadsheets could not be relied on to the same degree without production of the source material, there is no indication that the spreadsheets contain systemic errors. The lack of confidence affects only the statistical significance to assign statistical inferences that can be drawn from the data, not whether inferences can be drawn at all. There is no reason to think that marginally increased statistical confidence would provide evidence that was different in kind, rather than merely different in degree. And since the district court made its ruling on the absence of any evidence, not the lack of a sufficient degree of evidence from the spreadsheet data, the source material would not change its ruling.

Although the majority indicates that the requested depositions and alleged lack of emails could change the district court's decision, the majority only supports its declaration by vaguely referencing the possibility that the depositions and emails could show animus. The majority does not explain its basis for that conclusion and does not address the fact that Doe's requests concerning emails and depositions are based on conclusory allegations. Doe's Rule 56(d) affidavit seeks only emails and internal correspondence to the extent that they contain discriminatory and or stereotypical law enforcement practices. Doe's requests, therefore, make the conclusory allegation that emails containing discriminatory practices exist and were withheld by the City. Doe does not offer any basis for her allegation that emails containing discriminatory practices exist. She does not, for example, allege any bad faith on the part of the City in suppressing responsive material. Since the City has already made substantial productions, including of internal documents, there is doubly a lack of reason to believe that emails containing evidence of discrimination exist and were withheld. Similarly, for the depositions, although it is true that depositions might reveal animus, Doe provides no reason why she believes that depositions would provide evidence of animus. It is always true that a deposition might provide smoking gun evidence. But without any indication that such smoking gun evidence exists and could be found via deposition, mere speculation that depositions could change the district court's ruling is just that-speculation. As all Doe offers for why emails and depositions could change the district court's ruling are conclusory allegations and speculation, Doe's assertions to the contrary in the Rue 56(d) affidavit should not be credited. See Emmons v. McLaughlin , 874 F.2d 351 , 358 (6th Cir. 1989).

In sum, Doe failed to identify anything in the Rule 56(d) affidavit that could change the district court's decision. Therefore, I disagree with the majority that this Plott factor weighs in Doe's favor.

III

Even if remand were justified, our instructions on remand should account for the deference we give the district court. Unless our review of the Plott factors means that the district court could reach only one conclusion, we should not foreclose the district court from reconsidering the Plott factors on remand in light of our decision. Here, the majority does not indicate that, taken as a whole, the Plott factors could mean only that the district court should grant additional discovery. Nor has the majority identified any Plott factor as dispositive, meaning the district court would have to permit additional discovery regardless of how it weighed the other factors. Thus, even taking the majority's opinion to announce what the record could support, rather than its own interpretations of the facts, the proper instruction on remand would be for the district court to determine whether additional discovery is warranted in light of the majority's opinion. It is incorrect to simply mandate that the district court permit additional discovery.