Opinion
1:14-CV-333
09-08-2015
THOMAS H. KRAKAUER, Plaintiff, v. DISH NETWORK, L.L.C., Defendant.
MEMORANDUM OPINION AND ORDER
This matter is before the Court on a motion to exclude an expert report, (Doc. 57), filed by the defendant, Dish Network, L.L.C. Because the plaintiff's expert witness is qualified and her report and opinions rest on a reliable foundation, the report and testimony are admissible. The Court will deny the motion.
BACKGROUND
The Telephone Consumer Protection Act ("TCPA") regulates telemarketing and prohibits sellers from making phone solicitations to people who register their phone numbers on a national do-not-call registry ("NDNC list") without consent. 47 U.S.C. § 227(c); 47 C.F.R. § 64.1200(c)(2); see Mims v. Arrow Fin. Servs., LLC, 132 S. Ct. 740, 745-46 (2012). The TCPA also requires sellers and telemarketers to maintain an "internal" do-not-call list ("IDNC list"), that is, "a list of persons who request not to receive telemarketing calls made by or on behalf of that [seller]." 47 C.F.R. § 64.1200(d); see also United States v. Dish Network, L.L.C., 75 F. Supp. 3d 942, 960 (C.D. Ill. 2014), vacated in part on other grounds on reconsideration, 80 F. Supp. 3d 917 (C.D. Ill. 2015). The TCPA prohibits a telemarketer from calling individuals on its IDNC list or on the IDNC list of a seller on whose behalf the telemarketer calls, even if those individuals' phone numbers are not on the NDNC list. See 47 C.F.R. § 64.1200(d)(3), (6).
The TCPA creates a private right of action for injunctive and monetary relief for any "person who has received more than one telephone call within any 12-month period by or on behalf of the same entity in violation of the [TCPA] regulations." 47 U.S.C. § 227(c)(5); see also 47 C.F.R. § 64.1200(d)(3) (liability for IDNC violations). These rules only apply to residential telephone numbers; calls to businesses are not actionable under these subsections. See 47 C.F.R. § 64.1200(c)(2), (d)(3). Calls are also not actionable if a seller has an "established business relationship" ("EBR") with a person, which is created after an individual makes a purchase, inquiry, or application for products or services and lasts for a certain number of months. See 47 U.S.C. § 227(a)(4); 47 C.F.R. § 64.1200(f)(5); see also Snow v. Glob. Credit & Collection Corp., No. 5:13-CV-721-FL, 2014 WL 5781439, at *4 (E.D.N.C. Nov. 6, 2014) (collecting cases).
In 2003, the plaintiff, Thomas Krakauer, registered his residential phone number on the NDNC list. (Doc. 102 at 19.) Dr. Krakauer alleges that Dish Network, a seller of satellite television programming and related services, (Doc. 56-4 at ¶ 5), or its authorized dealer, Satellite Systems Network ("SSN"), called him on this number numerous times between May 2009 and September 2011, including at least two calls in a 12-month period, in violation of the TCPA. (Doc. 32 at ¶¶ 25-30, 54-59.) In these calls, SSN attempted to sell Dr. Krakauer Dish products. (Doc. 32 at ¶ 26.) The calls continued after Dr. Krakauer complained to Dish about SSN's sales tactics and after Dish placed Dr. Krakauer on its IDNC list and instructed SSN to do the same. (Doc. 32 at ¶¶ 27-28; see Doc. 81-51 at 3-13; Doc. 81-54.) During this time, SSN was an authorized dealer for Dish and only marketed for Dish. (Doc. 32 at ¶ 28; see Doc. 48-5 at 6.)
Dr. Krakauer seeks injunctive and monetary relief, (Doc. 32 at 14), and class-wide relief on behalf of two proposed classes: (1) all persons whose telephone numbers were on the NDNC list for at least 30 days, but who received telemarketing calls from SSN to promote Dish between May 1, 2010, and August 1, 2011 (the "NDNC class"); and (2) all persons whose telephone numbers were on the IDNC list of Dish or SSN, but who received telemarketing calls from SSN to promote Dish between May 1, 2010, and August 1, 2011 (the "IDNC class"). (Doc. 47; see also Doc. 32 at ¶¶ 54-59.)
In support of his pending motion for class certification, (Doc. 47), Dr. Krakauer submitted a report by Anya Verkhovskaya. (Doc. 48 at 9-10; see Docs. 48-2 to 48-4.) Ms. Verkhovskaya's report describes analyses she and her team performed on SSN's call records to identify putative class members—individuals who received a call from Dish or SSN in violation of the TCPA during the relevant class period. (See Doc. 48-2 at 2-3.) Dish has filed a motion to exclude Ms. Verkhovskaya's report, (Doc. 57), contending that it fails to meet the standard for admissible expert evidence. (Doc. 58 at 5-6.)
MS. VERKHOVSKAYA'S REPORT
Dr. Krakauer hired Ms. Verkhovskaya to analyze telemarketing calls made by SSN on behalf of Dish to identify putative class members in Dr. Krakauer's two proposed classes. (See Doc. 48 at 9-12; Doc. 48-2 at 2-3; Doc. 103 at 3-4.) Dr. Krakauer also proposes to use Ms. Verkhovskaya's report and testimony to establish Dish's liability, as her analysis and report identifies more than 28,000 numbers that received two or more calls from SSN in any 12-month period during the class period, in violation of the TCPA. (See Doc. 48 at 9-12, 17; Doc. 48-2 at 2-3.)
1. Ms. Verkhovskaya's Qualifications
Ms. Verkhovskaya is a Partner and Chief Operating Officer of A.B. Data, a company that "provides a full range of class action and complex litigation support services," most notably, the analysis of call records to identify class members in cases brought under the TCPA. (See Doc. 48-2 at 3-5; Doc. 103 at 6-7.) Ms. Verkhovskaya and A.B. Data have provided such services in a number of TCPA cases in federal court in recent years. (See Doc. 48-2 at 5-7; Doc. 103 at 4-5, 8-9.) Ms. Verkhovskaya has testified as an expert witness in several TCPA cases in the last four years. (Doc. 48-2 at 7-8; Doc. 103 at 8-10.) A.B. Data has served as court-appointed notice and claims administrator in "hundreds" of cases wherein its role was "to use available class member records and data to identify class members in order to send them appropriate notice and, if applicable, claims information." (Doc. 48-2 at 3-4.)
2. Facts and Data Relied Upon
Plaintiff's counsel gave Ms. Verkhovskaya a number of data files listing more than 1.6 million calls placed by SSN between May 11, 2010, and August 1, 2011, to promote Dish. (Doc. 48-2 at 2, 8-9; see Doc. 103 at 14-15.) Dr. Krakauer's attorney received the call records from Five9, Inc., a company that provides telemarketers, including SSN, with software to assist in making telemarketing calls. (See Doc. 48 at 10-11; Doc. 48-2 at 8-9; Doc. 56-8 at 3-5.) These records included information such as the called phone number, the date and length of the call, whether the call was inbound or outbound, and a "disposition" field, which included information such as whether the called person is a Dish customer or the call was not connected. (See Doc. 103 at 11, 20-21; see also Doc. 48-2 at 9-11.)
In addition to SSN's call records, A.B. Data used information from "reputable data vendors" Nexxa Group, Inc., and LexisNexis. (Doc. 48-2 at 4.) Nexxa provided the date of registration on the NDNC list for each phone number. (Doc. 48-2 at 4; Doc. 103 at 24-25.) Lexis provided information on whether a number was associated with a business or residence and the name and address of the telephone subscriber during the class period. (Doc. 48-2 at 5; Doc. 103 at 30, 33-34.) According to Ms. Verkhovskaya, it is "common knowledge" in the industry of class action administration that Lexis "is one of the largest aggregators of public record and proprietary information." (Doc. 48-2 at 5 n.2; see Doc. 103 at 19.) She states that "A.B. Data has longstanding relationships and prior experience" with these data vendors and that she "regularly use[s] these [vendors] in administering class action settlements, a context in which maximum accuracy and reliability is critical." (Doc. 48-2 at 4-5; see Doc. 103 at 19, 41.) She further states that, "[i]n [her] experience, gained over the course of several years of working with these vendors, the vendors provide accurate and reliable information." (Doc. 48-2 at 5; see Doc. 103 at 41.)
3. Principles and Methods Applied
To identify putative class members, A.B. Data first flagged and removed from SSN's records calls with 10 different "disposition codes" that indicated the call was not connected. (See Doc. 48-2 at 9; Doc. 103 at 20-23.) These codes included "Busy," "Fax machine," "No Answer," and "Operator Intercept," and also included calls with a duration of "00:00:00." (Doc. 48-2 at 9.) Ms. Verkhovskaya testified that plaintiff's counsel told her what these codes meant. (Doc. 103 at 21-22.) This step identified 230,121 "connected" calls. (Doc. 48-2 at 10.)
Next, A.B. Data used "programmatic queries . . . , a commonly-used data analysis procedure," to identify connected calls that were second or later calls to the same number in any 12-month period during the class period. (Doc. 48-2 at 10; see Doc. 103 at 15.) The analysis identified 58,151 unique phone numbers that received 164,494 calls. (Doc. 48-2 at 10.) A.B. Data then coordinated with Nexxa to determine which of these numbers were registered on the NDNC list as of April 1, 2010, and identified 23,625 such numbers that received 66,448 calls. (Doc. 48-2 at 10.)
The Five9 files begin with SSN's calls placed on May 1, 2010. (See Doc. 48-2 at 10.) Because a phone number must be registered on the NDNC list for 30 days before a telemarketing call may violate the TCPA, see 16 C.F.R. § 310.4(b)(3)(iv); 47 C.F.R. § 64.1200(c)(2)(i)(D), Ms. Verkhovskaya was interested in numbers registered at least 30 days before May 1. (See Doc. 48-2 at 10; Doc. 103 at 24.)
A.B. Data then identified and removed non-actionable calls to business numbers. (See Doc. 48-2 at 10-11); see also 47 C.F.R. § 64.1200(c)(2), (d)(3). First, A.B. Data removed 1,275 calls in SSN's call logs "assigned the disposition 'Business.'" (Doc. 48-2 at 10-11.) Second, A.B. Data coordinated with Lexis and determined that 118 of the remaining numbers were business numbers during the class period; these steps narrowed the list to 22,232 phone numbers. (Doc. 48-2 at 11.)
A.B. Data then identified and removed non-actionable calls to Dish customers. (Doc. 48-2 at 11); see 47 C.F.R. § 64.1200(f)(5), (14)(ii). A.B. Data removed 1,782 calls in SSN's logs "assigned the disposition of 'Dish Customer,'" reducing the class list to 20,450 numbers. (Doc. 48-2 at 11; see Doc. 103 at 32.)
These 20,450 numbers make up Dr. Krakauer's first proposed class, the NDNC class. (See Doc. 47; Doc. 48-2 at 11-12.) A.B. Data determined that SSN placed 57,900 connected calls to these numbers during the class period. (Doc. 48-2 at 11-12.)
Ms. Verkhovskaya attached to her report a list of all 20,450 phone numbers along with, for the vast majority, the "associated name and address" from SSN's logs "and supplemented through additional data research" with Lexis. (Doc. 48-2 at 12, 15-16; see Docs. 48-2 to 48-4.) For numbers in SSN's logs that had incomplete name and address information, A.B. Data used the "standard" process of providing the phone numbers to Lexis and receiving an output file with the names and addresses of telephone subscribers for a given time period. (Doc. 48-2 at 15; see Doc. 103 at 33-34, 37.) A.B. Data appended this information to its list of phone numbers. (Doc. 48-2 at 15-16.) Ms. Verkhovskaya stated that "[i]t is common in [her] profession to rely on [vendors like Lexis] to identify the names and addresses of subscribers of telephone numbers" and that she has used Lexis to perform similar services for courts in many cases. (Doc. 48-2 at 16; see Doc. 103 at 19, 41.)
A.B. Data performed a similar analysis to determine the putative members of the IDNC class. (See Doc. 48-2 at 12-15.) Plaintiff's counsel provided A.B. Data with files containing Dish's IDNC list. (Doc. 48-2 at 12-13.) A.B. Data followed the same methodology above and identified 7,117 unique numbers that received at least one actionable call during the class period. (Doc. 48-2 at 13.) These numbers received 20,549 calls. (Doc. 48-2 at 13.) As to SSN's IDNC, A.B. Data used SSN's call records to identify 2,233 connected calls with "DNC" or "Do Not Call" in the disposition field. (Doc. 48-2 at 14.) Plaintiff's counsel told Ms. Verkhovskaya that these codes meant that the individual was on SSN's IDNC list. (Doc. 103 at 33.) Following the same methodology, A.B. Data identified 714 unique numbers that received at least one actionable call. (Doc. 48-2 at 14-15.) These numbers received 2,058 calls. (Doc. 48-2 at 15.) Adding the 714 numbers from SSN's IDNC list to the 7,117 numbers from Dish's IDNC list results in a total of 7,831 numbers in the IDNC class. (See Doc. 48-2 at 14-15.)
A.B. Data researched and appended the names and addresses for these numbers using the same process as with the NDNC class. (See Doc. 48-2 at 13-16.) Ms. Verkhovskaya attached to her report a list of all 7,831 numbers with names and addresses, where available. (Doc. 48-4 at 36-175.)
ANALYSIS
Dr. Krakauer proffers Ms. Verkhovskaya's report under Federal Rule of Evidence 702. (Doc. 76 at 5-6.) Rule 702 provides that a report from a witness qualified as an expert is admissible if (1) it is "based upon sufficient facts or data," (2) it is "the product of reliable principles and methods," and (3) "the principles and methods [have been applied] reliably to the facts of the case." PBM Prods., LLC v. Mead Johnson & Co., 639 F.3d 111, 123 (4th Cir. 2011); Fed R. Evid. 702. The evidence is admitted if it "rests on a reliable foundation and is relevant." Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 597 (1993).
It is undisputed that if the Court excludes Ms. Verkhovskaya's report, there will be insufficient evidence to support Dr. Krakauer's motion for class certification. (See Doc. 48 at 9-12; see generally Doc. 47.) "[W]hen an expert's report . . . is critical to class certification," a court "must perform a full Daubert analysis" and rule on any challenge to an expert's report or to the reliability of information provided by an expert before ruling on a class certification motion. Am. Honda Motor Co. v. Allen, 600 F.3d 813, 815-16 (7th Cir. 2010) (per curiam); see also In re Blood Reagents Antitrust Litig., 783 F.3d 183, 187 (3d Cir. 2015); In re Zurn Pex Plumbing Prods. Liab. Litig., 644 F.3d 604, 627-28 (8th Cir. 2011); Sher v. Raytheon Co., 419 F. App'x 887, 890-91 (11th Cir. 2011); Coleman v. Union Carbide Corp., No. 2:11-0366, 2013 WL 5461855, at *22-23 (S.D.W. Va. Sept. 30, 2013).
The district court "must act as a gatekeeper," E.E.O.C. v. Freeman, 778 F.3d 463, 472 (4th Cir. 2015), and "determine reliability in light of the proposed expert's full range of experience and training as well as the methodology used to arrive at a particular conclusion." Am. Honda Motor Co., 600 F.3d at 816 (internal quotation marks omitted); see also Westberry v. Gislaved Gummi AB, 178 F.3d 257, 261 (4th Cir. 1999). In Daubert, the Supreme Court provided a number of factors for district courts to consider when determining whether to admit expert evidence under Rule 702, including whether the expert's theory or technique is generally accepted in the relevant scientific or expert community; whether the theory has been subjected to peer review and publication; whether the theory can be and has been tested; the known or potential error rate; and the existence of standards and controls. Daubert, 509 U.S. at 593-95; see United States v. Crisp, 324 F.3d 261, 266 (4th Cir. 2003). These factors are not exclusive or dispositive, Daubert, 509 U.S. at 593, and the district court need not consider them "in lockstep fashion." Coleman, 2013 WL 5461855, at *17. Rather, "'the test of reliability is flexible' and 'the law grants a district court the same broad latitude when it decides how to determine reliability as it enjoys in respect to its ultimate reliability determination.'" United States v. Wilson, 484 F.3d 267, 274 (4th Cir. 2007) (quoting Kumho Tire Co. v. Carmichael, 526 U.S. 137, 141-42 (1999)).
Whether expert evidence is reliable is primarily a question of the validity of the expert's methodology, not the quality of the data used or the conclusions produced. Manpower, Inc. v. Ins. Co. of Penn., 732 F.3d 796, 806 (7th Cir. 2013). "The soundness of the factual underpinnings of the expert's analysis and the correctness of the expert's conclusions based on that analysis are factual matters to be determined by the trier of fact, or, where appropriate, on summary judgment." Smith v. Ford Motor Co., 215 F.3d 713, 718 (7th Cir. 2000); see also United States v. Moreland, 437 F.3d 424, 431 (4th Cir. 2006) (noting that a court's focus is "on the principles and methodology employed by the expert, not on the conclusions reached" (internal quotation marks omitted)), overruled on other grounds by United States v. Evans, 526 F.3d 155 (4th Cir. 2008).
The Court also "need not determine that the proffered expert [evidence] is irrefutable or certainly correct. As with all other admissible evidence, expert [evidence] is subject to testing by '[v]igorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof.'" Moreland, 437 F.3d at 431 (internal citation omitted) (quoting Daubert, 509 U.S. at 596).
DISH'S MOTION TO EXCLUDE EXPERT REPORT
Dish contends that Ms. Verkhovskaya's report does not meet the standard for admissible expert evidence. (Doc. 58 at 5.) Dish does not challenge Ms. Verkhovskaya's qualifications as an expert or the qualifications of her team; nor does Dish seriously dispute that the data vendors she and A.B. Data used are commonly used, reliable, and accurate; nor does Dish contend that Ms. Verkhovskaya's method of culling non-actionable calls from the more than 1.6 million calls in SSN's records is unreliable. Rather, Dish contends that plaintiff's counsel improperly defined certain codes or fields in SSN's records, that some of the underlying data Ms. Verkhovskaya used is incorrect, and that some of the conclusions she drew from her analyses are wrong.
Because the Court concludes that any influence by plaintiff's counsel on Ms. Verkhovskaya does not render her report inadmissible and that Dish's other concerns either "implicate[] not the reliability of [Ms. Verkhovskaya's] methodology but the conclusions that it generated," Manpower, 732 F.3d at 807, or the accuracy of the underlying data, see Smith, 215 F.3d at 718, the Court will deny Dish's motion to exclude.
1. The method for determining "connected" calls and calls to numbers on the IDNC lists is flawed.
Ms. Verkhovskaya identified unconnected calls by excluding calls with certain disposition codes, such as "Busy," "Fax machine," or "No Answer," and calls with a call duration of "00:00:00." (Doc. 48-2 at 9.) As to all calls in both classes, Dish contends that because Ms. Verkhovskaya herself did not know what the codes meant and relied on plaintiff's counsel's instructions, (see Doc. 103 at 21-22), that her method for determining "connected" calls is not based on sufficient facts or data. (Doc. 58 at 9-12); see also Fed. R. Evid. 702(b). Similarly, only as to the IDNC class members, Dish contends that because Ms. Verkhovskaya relied on plaintiff's counsel in concluding that calls marked "DNC" or "Do Not Call" meant that the number was on SSN's IDNC list, (Doc. 103 at 33), that this portion of her analysis is not based on sufficient facts or data. (Doc. 58 at 10-12.)
In support, Dish primarily relies upon Southwell v. Mortg. Inv'rs Corp. of Ohio, No. C13-1289 MJP, 2014 WL 3956699 (W.D. Wash. Aug. 12, 2014), appeal docketed, No. 14-36036 (9th Cir. Dec. 5, 2014), another TCPA case that Dish contends has facts "nearly identical" to those here. (Doc. 58 at 11-12.) In Southwell, the expert used "filters" and "parameters" set by plaintiff's counsel and reported the numbers produced. 2014 WL 3956699, at *3. The court noted that the expert had "no idea what the numbers represent and no independent opinion on whether they are accurate representations of what [p]laintiffs purport them to mean." Id. Further, the expert "had no expertise in the TCPA or do-not-call policies or laws or practices," id. (internal quotation marks omitted), and the court noted that the expert did not attempt to remove customer or business numbers or account for the 30-day grace period after an individual registers on the NDNC list. Id.
Here, unlike in Southwell, Ms. Verkhovskaya did not simply perform an analysis set or developed by Dr. Krakauer's attorney. She testified that A.B. Data's "methodology is customized to fit the particulars of each case [depending] on the criteria given." (Doc. 103 at 12.) In her report, she describes the steps that teams under her direction at A.B. Data took to perform the analyses, and she describes the meetings and her instructions before these steps. (See Doc. 48-2 at 16-17.) Plaintiff's counsel told Ms. Verkhovskaya what certain codes meant so she could perform her analysis using her methods. Unlike the cases cited by Dish, Ms. Verkhovskaya's resulting class list is not a regurgitation of what Dr. Krakauer's attorney told her. (See Doc. 58 at 11.) Additionally, Ms. Verkhovskaya has extensive experience in this type of analysis for TCPA cases, which Dish has not disputed, and she took many of the steps to remove non-actionable calls that the purported expert in Southwell did not. (See Doc. 48-2 at 10-15; Doc. 103 at 11, 15.)
Dish does not contend that an expert may never rely on counsel to define certain words or codes in a data set, and it is difficult to understand how any hired expert could perform analysis without some instruction as to what certain words, fields, or codes represent. Notably, as to the unconnected calls, Dish does not contend that plaintiff's counsel or Ms. Verkhovskaya is wrong in perceiving that the disposition codes indicate that a call was not connected. And these codes seem obvious on their face. (See Doc. 48-2 at 9.) Indeed, a Dish expert in a previous case relied on the same types of disposition codes to remove the same calls. (See Doc. 76-1 at 15.) On the whole, it appears reasonable for Ms. Verkhovskaya to rely on plaintiff's counsel's instruction as to what certain disposition codes meant. This influence does not undermine the admissibility of her report, but it may undermine its weight and credibility—matters appropriate for cross-examination. See Young v. Swiney, 23 F. Supp. 3d 596, 622-23 (D. Md. 2014); see also Elm Grove Coal Co. v. Dir., O.W.C.P., 480 F.3d 278, 301 n. 23 (4th Cir. 2007) ("The interplay between testifying experts and the lawyers who retained them should, however, be fair game for cross-examination.").
His expert report also indicates that Dish told him what certain codes in the data represented. (See Doc. 76-1 at 9.)
As to the IDNC issue, SSN's general manager testified that the "DNC" and "Do Not Call" codes "do not indicate a 'do not call' request by a called individual," but rather signal SSN agents to not call the individual because a "particular SSN agent would personally call back that individual (i.e., it was 'their lead')." (Doc. 58-2 at ¶ 4.) Dish makes the same argument about Ms. Verkhovskaya's reliance on plaintiff's counsel to define these codes and asserts that the evidence that counsel was actually wrong about these codes "highlights how [the] implications [of] her factual assumptions impact the reliability of her methodology." (Doc. 58 at 10-11 & n.2.)
First, as discussed, plaintiff's counsel's influence in defining these codes for Ms. Verkhovskaya does not per se render her report inadmissible. Second, this influence and the correctness of the facts underlying Ms. Verkhovskaya's identification of IDNC class members is more properly a question for the trier of fact. Like the disposition codes related to unconnected calls, these codes seem obvious on their face, and it appears reasonable to rely on these codes as identifying calls to individuals on SSN's IDNC list. Every telemarketer who makes residential calls is required by law to maintain an IDNC list, see 47 C.F.R. § 64.1200(d), and should have a system in place to notify its employees of who is on that list. The phrases "DNC" and "Do Not Call" from SSN's call records are, in context, persuasive circumstantial evidence that persons associated with those numbers are on SSN's IDNC list. SSN's explanations to the contrary are inconsistent with its obligations to maintain an IDNC list. If Ms. Verkhovskaya was wrong in her analysis of SSN's IDNC list, this may undermine the weight and credibility of her report, but that does not affect admissibility.
The Court further discusses the issue of SSN's purported IDNC list in the class certification order, which will be filed shortly.
2. Data from Nexxa is unreliable.
Nexxa provided Ms. Verkhovskaya with data showing when individuals registered numbers on the NDNC list. (Doc. 48-2 at 10.) Dish attempted to verify the accuracy of Nexxa's data using Dr. Krakauer's registration information. (See Doc. 58 at 13.) The Nexxa data indicates that Dr. Krakauer registered his number on June 1, 2003, but the NDNC list website shows that he registered on July 3, 2003. (See Doc. 58 at 13; Doc. 56 at 17-18; Docs. 56-14, 56-15.) Dish contends that because "Nexxa's data is inaccurate" for the one person for whom it attempted to verify accuracy, the data and Ms. Verkhovskaya's report should be excluded. (Doc. 58 at 13.)
For one, Dish's challenge to the factual underpinnings of Ms. Verkhovskaya's report is not a proper challenge at this stage. See Smith, 215 F.3d at 718. Second, Dr. Krakauer provided an affidavit from a Nexxa employee explaining that these dates differ because Dr. Krakauer registered his number twice. (Doc. 76-4 at ¶¶ 2-5; Doc. 76 at 14.) The employee testified that Nexxa keeps the date of initial registration while the date on the NDNC list website can be "overwritten" by a later registration. (See Doc. 76-4 at ¶ 5.) Dish's concerns over the reliability of Nexxa's data are unsupported.
3. Ms. Verkhovskaya failed to conduct an independent examination of Lexis or Nexxa data to ensure reliability.
Dish contends that Ms. Verkhovskaya's report should be excluded because she did not test the reliability of data from Lexis or Nexxa. (Doc. 58 at 12-13, 16-18.) Ms. Verkhovskaya testified that she is aware that Lexis does not guarantee the accuracy of its results and that A.B. Data did not and is not able to corroborate Lexis's results. (Doc. 103 at 24, 27.) She also testified that A.B. Data has "not yet tested Nexxa's database." (Doc. 103 at 19.) Dish contends that because she did not attempt to independently verify the reliability of Lexis or Nexxa data, Ms. Verkhovskaya should not be able to rely on Lexis or Nexxa data in her analysis and report. (Doc. 58 at 13, 17.)
"[A]n expert may rely on data that she did not personally collect" and "need not have conducted her own tests." Gussack Realty Co. v. Xerox Corp., 224 F.3d 85, 94-95 (2d Cir. 2000) (per curiam). Rule 703 provides that an expert may rely on facts or data of which the expert has been made aware "[i]f experts in the particular field would reasonably rely on those kinds of facts or data in forming an opinion." Fed. R. Evid. 703; see also Ward v. Dixie Nat'l Life Ins. Co., 595 F.3d 164, 182 (4th Cir. 2010).
First, Dish does not dispute Ms. Verkhovskaya's testimony and statements in her report that Lexis and Nexxa provide "maximum accuracy and reliability," (Doc. 103 at 19; Doc. 48-2 at 4-5), or that it is common practice in her industry to rely on such data vendors. (See Doc. 48-2 at 4-5 & n.2; see generally Doc. 58.) Ms. Verkhovskaya testified that she has never had concerns about the work provided by Lexis or Nexxa. (Doc. 103 at 41.) This may well be enough to allow Ms. Verkhovskaya to rely on the data under Rule 703. See Ward, 595 F.3d at 182. Moreover, as to Nexxa, it would make no practical sense to require A.B. Data to independently verify every phone number against the NDNC list website to determine if Nexxa's data on registration dates is accurate and perform the same data collection Nexxa has already performed, as Dish appears to contend; this would defeat the purpose of relying on third parties for data. Finally, other than the one non-issue addressed supra as to Dr. Krakauer's registration, Dish cannot point to anything else that calls into question the reliability of Nexxa's data.
As to Lexis, contrary to Dish's assertion, A.B. Data has conducted an independent examination to ensure the reliability of Lexis's data generally. Ms. Verkhovskaya testified that A.B. Data has, in the past, "perform[ed] a number of data tests with Lexis[] where [A.B. Data] run[s] known files against [Lexis] databases and measure[s] the output results against known data," and these tests have shown 86-97% accuracy. (Doc. 103 at 19.) As discussed supra concerning Nexxa, there is no requirement that an expert conduct an independent evaluation of data and results each time she uses them. See Ward, 595 F.3d at 182; Fed. R. Evid. 703. And, "[i]n any event, [Dish has] offered nothing to suggest the [Lexis] data are unreliable or inaccurate." Kolbe v. O'Malley, 42 F. Supp. 3d 768, 780 (D. Md. 2014), appeal docketed, No. 14-1945 (4th Cir. Sept. 11, 2014); see also Ward, 595 F.3d at 182.
Finally, Dish also expresses concern with the possible 14% error rate, as it creates the potential for millions of dollars in liability for non-actionable calls. (Doc. 58 at 17-18.) The Court has considered the potential error rate and does not find it unreasonably high for these particular circumstances. See Daubert, 509 U.S. at 594. Like its other arguments, this goes more to the weight of the evidence, not its admissibility.
4. Ms. Verkhovskaya did not reliably apply the Lexis data to her methodology.
Dish states that, contrary to her report and testimony, Ms. Verkhovskaya did not remove all numbers identified as business numbers by Lexis. (Doc. 58 at 18-19; Doc. 56 at 23-25; see also Doc. 48-2 at 11; Doc. 103 at 26.) Dish contends that this shows she failed to reliably apply her methodology to the data. (Doc. 58 at 18-19; Doc. 56 at 25); see also Fed. R. Evid. 702(d).
Dish asserts that "Ms. Verkhovskaya failed to remove hundreds" of business numbers that Lexis data identifies as such and provided "a non-exhaustive list" of 12 numbers. (Doc. 58 at 18-19; Doc. 56 at 23-24; Doc. 56-18.) Dish has submitted no evidence that there are "hundreds" of business numbers on the list compiled by Ms. Verkhovskaya; rather, it has provided only counsel's unsworn statements and an "exhibit" compiled by counsel. (See Doc. 58 at 18-19; Doc. 56 at 23-25; Doc. 56-18); see also Adjabeng v. GlaxoSmithKline, LLC, No. 1:12-CV-568, 2014 WL 459851, at *3 (M.D.N.C. Feb. 5, 2014) (collecting cases holding that counsel's unsworn statements in briefs are not evidence). To the extent Ms. Verkhovskaya missed a few business numbers out of more than 28,000, (see Doc. 48-2 at 10-15), that is insufficient to establish that she has not reliably applied her methodology to the data.
The Court discusses the issue of identifying business numbers in additional detail in the class certification order.
Rule 702 does not require that an expert's evidence be "irrefutable or certainly correct." Moreland, 437 F.3d at 431; see also In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 744 (3d Cir. 1994) ("The grounds for the expert's opinion merely have to be good, they do not have to be perfect."). Indeed, Rule 702 only requires the expert "reliably" apply her methods to the facts, Fed. R. Evid. 702(d); "reliably" cannot mean "perfectly and without error." Contrary to Dish's assertions and based on the admissible evidence, it appears that Ms. Verkhovskaya reliably applied her methodology to the Lexis data.
Further, for purposes of the pending motion, the Court's focus is not on the quality of Ms. Verkhovskaya's conclusions but on the principles and methodology used. See Moreland, 437 F.3d at 431; Smith, 215 F.3d at 718; see also Pugh v. Louisville Ladder, Inc., 361 F. App'x 448, 454 n.4 (4th Cir. 2010) (noting that while the Supreme Court has recognized that, in evaluating expert evidence, "conclusions and methodology are not entirely distinct from one another," a court's focus remains on the expert's principles and methodology (internal quotation marks omitted)). The alleged deficiencies with Ms. Verkhovskaya's results go "more to the weight of the expert [evidence] than to its Daubert admissibility." Pugh, 361 F. App'x at 456; see also Kannankeril v. Terminix Int'l, Inc., 128 F.3d 802, 809 (3d Cir. 1997) (noting that the test is not whether the "expert might have done a better job"); Campbell v. Fawber, 975 F. Supp. 2d 485, 500 (M.D. Pa. 2013) (noting that challenges to the accuracy of an expert's conclusions go to the weight of the evidence rather than its admissibility).
5. The Lexis data does not represent what Ms. Verkhovskaya contends it does.
In a related argument, Dish contends that, contrary to Ms. Verkhovskaya's report and testimony, the Lexis data does not cover the proposed class period and Lexis does not identify the telephone number subscriber. (Doc. 58 at 14-16.)
As to the time period, Dish's argument that Ms. Verkhovskaya has misunderstood the Lexis data appears to be little more than a dispute between witnesses. (See Doc. 58 at 14-15; Doc. 58-5 at ¶¶ 45-49.) It does not render her report inadmissible, but goes to weight. See, e.g., Westberry, 178 F.3d at 265-66 (collecting cases); Spesco, Inc. v. Gen. Elec. Co., 719 F.2d 233, 237-38 (7th Cir. 1983). Also, Ms. Verkhovskaya testified that she asked Lexis to provide user information during the class period, that she has worked with Lexis "several hundred times" in the past 15 years, and that she has never had concerns about Lexis's results. (Doc. 103 at 27-28, 41.) In light of this explanation and for purposes of this motion, Ms. Verkhovskaya's principles and methodology remain reliable. See Westberry, 178 F.3d at 265-66.
Further, Ms. Verkhovskaya testified that she first used name information in the SSN call logs to identify individuals and then used Lexis data to supplement the call logs where they were incomplete. (See Doc. 103 at 34; Doc. 48-2 at 15.) Dish's expert contends that roughly 40% of the numbers in the Lexis dataset that Ms. Verkhovskaya used do not have name information for the 2010-2011 period. (See Doc. 58-5 at 18, 48.) Elsewhere, Dish's expert's report appears to indicate that the SSN call logs had at least partial name information for more than 95% of the numbers and complete name information for almost 75%. (See Doc. 58-5 at 14, 41.) Neither Dish nor its expert has made clear how many numbers from the SSN call logs that make up the putative class lists lacked complete name information and whether and for how many of these numbers Lexis also lacked name information during the class period. Without more, Dish's passing argument that "Ms. Verkhovskaya used inapplicable data for half of the putative class," (Doc. 58 at 15), is not persuasive.
As to the telephone number subscribers, Ms. Verkhovskaya testified that she submitted the list of numbers to Lexis "and asked them to reverse append historic information of who was the subscriber of that phone number and what the address of that individual was during the [class period]." (Doc. 103 at 34; see Doc. 48-2 at 15.) A Lexis employee testified that the Lexis "data does not purport to identify the subscriber to the telephone number." (Doc. 58-3 at ¶ 7; Doc. 58 at 15.) For this reason, Dish contends that Ms. Verkhovskaya's methodology is flawed because her belief that Lexis identifies the subscriber is based "solely on her unfounded and inaccurate" opinion. (Doc. 58 at 15.)
Ms. Verkhovskaya did not testify that Lexis identifies subscriber information; she only testified that she asked Lexis to provide subscriber information. (See Doc. 103 at 34, 37-38.) Even if the data Lexis provided does not actually include subscriber information, that does not make Ms. Verkhovskaya's methods unreliable. She stated in her report that it is common in her industry to use third parties like Lexis to identify telephone number subscribers, (Doc. 48-2 at 5, 16), and Dish has not disputed this. Therefore, her belief that the Lexis data identifies subscribers is not based solely on her unfounded opinion; as discussed supra, it is based on working with Lexis on many cases over many years, having no concerns about the accuracy of Lexis's results, and asking Lexis to provide subscriber information, all to which she testified under oath. (See Doc. 103 at 34, 37, 41.) The fact that Lexis does not guarantee that the name provided is the "subscriber" is not fatal.
Further, for the reasons stated in the Court's class certification order, standing to sue under the section of the TCPA at issue is not limited to subscribers, so even if Ms. Verkhovskaya is wrong and Lexis does not identify the subscriber, that is not material to the case. --------
CONCLUSION
Dish has not challenged Ms. Verkhovskaya's qualifications, the reliability of the data or resources on which she relied, or her methods. Dish's challenges concern plaintiff's counsel's instructions as to what certain codes in the data meant, the accuracy of the underlying data, and the soundness of her conclusions, none of which affect admissibility, but go to weight. Dish also points out some potential errors or inaccuracies in Ms. Verkhovskaya's analysis or assumptions. Even taken together, these potential errors are not enough to show that Ms. Verkhovskaya has not reliably applied her methods to the data and facts of the case.
It is ORDERED that the motion to exclude expert report, (Doc. 57), is DENIED.
This the 8th day of September, 2015.
/s/_________
UNITED STATES DISTRICT JUDGE