Free Brief in Support of Motion - District Court of Colorado - Colorado

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 1 of 20

IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLORADO Civil Action No. 1:04-cv-1258-LTB-BNB STUDENT MARKETING GROUP, INC., Plaintiff, v. COLLEGE PARTNERSHIP, INC., f/k/a COLLEGE BOUND STUDENT ALLIANCE, INC., Defendant. REPLY MEMORANDUM IN SUPPORT OF PLAINTIFF'S MOTION IN LIMINE TO EXCLUDE THE EXPERT TESTIMONY OF MR. NATHAN ALLEN AND MR. THOMAS HILLER Plaintiff Student Marketing Group, Inc. ("SMG"), by and through its undersigned counsel, files this Reply Memorandum in Support of Plaintiff's Motion In Limine to Exclude the Expert Testimony of Mr. Nathan Allen and Mr. Thomas Hiller. I. INTRODUCTION

In SMG's Motion in Limine and accompanying Memorandum, SMG sets forth various grounds for why Messrs. Allen and Hiller, two of the experts retained by College Partnership, Inc. ("CPI"), should be excluded from offering testimony in this matter. Using testimony from CPI's own witnesses, SMG established that the testimony of Messrs. Allen and Hiller is not based on sufficient facts or data, is not the product of reliable principles and methods, and does not apply principles and methods reliably to the facts of this case. In response, CPI has either conceded the facts which serve as the basis for SMG's Motion or failed to offer sufficient facts or evidence to support the admissibility of the testimony of Messrs. Allen

PI-1411118 v1 0223780-0902

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 2 of 20

and Hiller. Pursuant to the Federal Rules of Evidence and the principles articulated in Daubert and related case law, this Court should exercise its role as a gatekeeper to exclude the testimony of Messrs. Allen and Hiller. During the status conference with the Court on July 22, 2005, the parties and the Court discussed the requirement under Daubert, Dodge and related case law to develop a record sufficient to demonstrate the Court's consideration and application of the appropriate standard governing admissibility of expert testimony. The Court reserved decision regarding whether an evidentiary hearing would be necessary. Since CPI essentially conceded in its Opposition most of the relevant facts and failed to offer evidence to support other contentions, SMG asserts that no further hearing will be required. To demonstrate this and to facilitate the preparation of the detailed findings required by Daubert, Dodge and related case law, SMG has prepared this Reply Brief in the form of proposed findings of fact and conclusions of law. In light of these discussions and the upcoming argument on SMG's Motion in Limine, using the evidence offered by SMG in its Motion and that offered by CPI in its Response in Opposition, this Reply attempts to set forth both the appropriate legal standard and the detailed factual record to serve as the basis for the Court granting SMG's Motion in Limine. II. 1. BACKGROUND

SMG is a marketing company that maintains a database containing names

and addresses of high school students. CPI is an education and career preparation company that provides products and services to college-bound high school students. CPI uses direct-mail advertising to promote its products and services. In doing so, CPI rents addresses from companies such as SMG to facilitate its direct mail program to potential customers.

-2-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 3 of 20

2.

In May of 2002, SMG and CPI entered into a List Rental Agreement (the

"Agreement") for a period of one year. Thereafter, SMG and CPI entered into a List Rental Renewal Agreement ("Renewal Agreement"). Shortly after the expiration of the Renewal Agreement, SMG filed this action against CPI to collect on unpaid invoices. After SMG filed its collection matter, CPI asserted counterclaims against SMG claiming that SMG did not supply the appropriate quantity or quality of student records to CPI. In support of its counterclaims, CPI retained Mr. Nathan Allen and Mr. Thomas Hiller. SMG has filed a Motion in Limine to Exclude the Expert Testimony of Mr. Nathan Allen and Mr. Thomas Hiller. III. A. Nathan Allen 3. CPI retained Mr. Allen individually, to process data files through two CPI'S EXPERTS

separate computer programs that were designed to (1) "CASS certify" the files (i.e., check addresses against those recognized by the U.S. Postal Service) and (2) check for duplicate names and addresses. Mr. Allen's opinions are limited to the fact that he ran the programs and obtained certain results. Ex. 5, Allen Dep. at 13, 19, 20.1 CPI relies on Mr. Allen's conclusions as the foundation for Mr. Hiller's testimony and to support CPI's claim that SMG did not supply CPI with the appropriate quality of records. 4. Nathan Allen is a computer programmer just two years out of college; he

is not an expert in direct mail. See Ex. 5, Allen Dep. at 21, 29. Although Mr. Allen has some limited experience running the CASS-certification program, id. at 24, 29, Mr. Allen had never

1

Unless otherwise indicated, citations to numeric exhibits are exhibits to SMG's Motion in Limine and accompanying memorandum and citations to alpha exhibits are exhibits to CPI's Response in Opposition.

-3-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 4 of 20

before sorted duplicates from a database using the computer program that he used in this case. Id. at 35-38. 5. In running the CASS certification, Mr. Allen used a program called Satori

to CASS-certify the files. When asked about why he chose Satori to run the CASS certification, Mr. Allen testified that he used Satori because that is what Merrell Remington was already using. Id. at 28-29. Mr. Allen would create a program with the parameters he thought were necessary and when he ran the program information would be sent to Satori who processes the information and gives its results. Id. at 22. 6. Mr. Allen used a program called SoundEx to attempt to identify

duplicates. Id. at 38. Mr. Allen's boss (who has not been designated or retained by CPI) apparently selected the program for sorting duplicates and instructed Mr. Allen as to its use. Id. at 35-37. When asked about his experience with the record-sorting program, Mr. Allen testified that he had never used the particular sound duplication software in a project. Id. at 36-37. Based on Mr. Allen's own testimony, which CPI has not refuted, the extent of Mr. Allen's experience with the SoundEx program is limited to one assignment in a single class taken several years ago and "messing around with it" on his computer at home "to see what it would do." Id. at 37-38. 7. Mr. Allen did no analysis to determine whether the program was

appropriate for identifying duplicate names. Id. at 111. Mr. Allen was unaware of any other organization or entity that used it for such a purpose and was not aware of and did not investigate its error rate. Id. at 111. 8. Mr. Allen knew nothing about the source of the data that he analyzed,

whether it contained data from multiple vendors, who applied the "identifier codes" (a/k/a

-4-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 5 of 20

"reservation numbers") that he used to categorize the data or whether the data had been manipulated by PrimeNet before it had been provided to him. Id. at 30-31, 59. 9. The record is unclear as to the precise source of the data that Mr. Allen did

analyze since Mr. Allen testified that Mr. Russell of CPI provided the files to him, id. at 30-31, but Mr. Russell testified that he never heard of Mr. Allen and did not give him the data that Mr. Allen claims to have analyzed. Ex. 2, Russell Dep. at 34, 54. B. Thomas Hiller 10. CPI retained Mr. Hiller to analyze the results provided by Mr. Allen and to

comment generally on the accuracy of SMG's marketing materials. See generally Ex. 6, Hiller Reports dated 1/12/05 and 2/11/05; Ex. 7, Hiller Dep. at 170. 11. Mr. Hiller is the president of Consultants in Information Management, Inc.

and he has experience in information management and analysis. Ex. A, Hiller Aff., ¶ 2. 12. Mr. Hiller identified multiple errors in Mr. Allen's analyses and requested

that Mr. Allen re-run his analyses on five separate occasions. Ex. 5, Allen Dep. at 70-73; Ex. 6, 2/11/05 Report at 2; Ex. 7, Hiller Dep. at 58-60. Accordingly, Mr. Hiller rendered his opinions and testified in his deposition before Mr. Allen completed his final analyses (although he did review some of Mr. Allen's preliminary analyses). 13. In addition to his review of Mr. Allen's preliminary analysis, Mr. Hiller

also reviewed SMG's marketing brochure and opined that "SMG's 98% accuracy and update frequency statements and deliverability guarantee is misleading and not what list compilers and marketers do as a standard practice." Ex. 6, 1/12/05 Report at 2. Mr. Hiller, however, conceded: · The SMG guarantee in its entirety is not misleading. Ex. 7, Hiller Dep. at 95-96.

-5-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 6 of 20

· · · · ·

He had no information about actual undeliverable mail sent by CPI. Id. at 81. He has no opinion as to the actual deliverability of mail to the addresses on SMG's lists. Id. at 88. He had no information about SMG's actual list hygiene and list updating practices. Id. at 82-85, 88. He does not know what percentage of the SMG names and addresses were not deliverable. Id. at 112-13. The fact that an address generates a CASS-certification error code during Mr. Allen's analysis does not mean that it is undeliverable, and he has no information on the percentage of such addresses that are not deliverable. Id. at 57, 115.

CPI did not dispute these admissions. 14. Instead of studying SMG's actual list hygiene practices, Mr. Hiller relies

exclusively upon generalized statistical data to support his opinion that SMG's "98% accuracy" statement is misleading. Specifically, he cites data from the U.S. Census Bureau which shows that 3.9% of the population moves every quarter, thereby likely undermining the accuracy of SMG's database. Mr. Hiller did no research to determine if that statistic applies to families with high school age children (the relevant group here). Id. at 104-7. Furthermore, Mr. Hiller concedes that his analysis does not hold true if the party sending the mail (CPI here) chooses to send the mail so it would be forwarded to those who fill out a change of address form. Id. at 9094, 199-200. IV. A. CONCLUSIONS OF LAW

Considerations Governing the Admissibility of Expert Testimony 15. When resolving questions regarding the admissibility of expert testimony,

the Court must consider and adhere to both the Federal Rules of Evidence and the mandates set

-6-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 7 of 20

forth in Daubert and its progeny. Federal Rule of Evidence 702 governs the admissibility of expert testimony and provides that: If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert . . . may testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case. FED. R. EVID. 702. In applying Rule 702, the court is required to perform a "gate-keeping" function to determine whether the opinion rendered is not only relevant, but also reliable. See Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 589-90 (1993); Dodge v. Cotter Corp., 328 F.3d 1212, 1223 (10th Cir. 2003) (performance of the gate-keeping role is not discretionary). 16. In performing its gate-keeping function, the court must create a

sufficiently developed record to demonstrate that it properly considered and applied the appropriate standard governing admissibility of expert testimony in making its reliability determination. See, e.g., Bitler v. A.O. Smith Corp., 400 F.3d 1227, 1232 (10th Cir. 2005). The Court must assess the reasoning and methodology underlying the experts' opinion and determine whether it is both scientifically valid and applicable to the particular set of facts. Dodge, 328 F.3d at 1221. 17. The party offering expert testimony "must show that the method employed

by the expert in reaching the conclusion is scientifically sound and that the opinion is based on facts which satisfy Rule 702's reliability requirements" Id. at 1222 (citing Mitchell v. Gencorp Inc., 165 F.3d 778, 781 (10th Cir. 1999).

-7-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 8 of 20

18.

The Supreme Court in Daubert listed four nonexclusive factors that the

trial court may consider when making determinations regarding the admissibility of expert testimony: (1) whether the opinion at issue is susceptible to testing and has been subjected to testing; (2) whether the opinion has been subjected to peer review; (3) whether there is a known or potential rate of error associated with the methodology used and whether there are standards controlling the technique's operation; and (4) whether the theory has been accepted in the scientific community. 509 U.S. at 593-94. This list is not exclusive and the court has broad discretion to consider a variety of other factors. Kumho Tire, 526 U.S. at 150. 19. Generally, the court should focus on an expert's methodology rather than

the conclusions it generates and if there is any step in the methodology that renders the analysis unreliable, the expert's testimony is inadmissible. Daubert, 509 U.S. at 595. "This is true whether the step completely changes a reliable methodology or merely misapplies that methodology." Mitchell, 165 F.3d at 782. Considering this standard and the role of the court as a gatekeeper, there are serious flaws in the reliability of the information analyzed by CPI's experts and the methodologies employed by them such that the testimony of Messrs. Allen and Hiller should be excluded. B. The Testimony of Allen and Hiller is Not Based on Sufficient Facts or Data 20. Among other things, Rule 702 requires that an expert's testimony be based

upon "sufficient facts or data." FED. R. EVID. 702. The proposed expert's opinion must be based upon facts that satisfy the reliability requirement of Rule 702. See Mitchell v. GenCorp, Inc., 165 F.3d 778, 781 (10th Cir. 1999). Where the data underlying expert opinions and testimony is unreliable and insufficient, expert testimony should be excluded. See Lantec, Inc. v. Novell, Inc., 306 F.3d 1003, 1025-26 (10th Cir. 2002) (holding that the District Court did not err in excluding the testimony of an expert who used unreliable data); The Ninth Ave. Remedial Group v. Allis-

-8-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 9 of 20

Chalmers Corp., 141 F. Supp. 2d 957, 960 (N.D. Ind. 2001) (excluding expert testimony that relied upon "data categorizations, compilations, and conclusions" that were unreliable). 1. CPI's experts did not analyze the data SMG supplied to CPI. 21. CPI challenges the quality of the data provided by SMG, but its experts,

Messrs. Hiller and Allen, never analyzed SMG's actual lists. See Ex. 8, Allen's Report at 1; Ex. 7, Hiller Dep. at 22-23; Allen Dep. at 48. CPI does not dispute this. See Response in Opposition, p. 4. 22. Instead of analyzing SMG's actual lists, Messrs. Hiller Allen relied

exclusively upon data from PrimeNet, which combined SMG's lists with data from various other sources. Ex. 2, Russell Dep. at 68; Ex. 7, Hiller Dep. at 120. In addition to combining SMG's data with other data, PrimeNet also manipulated SMG's data through various processes including, removing duplicates, assigning "common names," i.e., names supplied by more than one vendor to the account of one vendor or another, running various postage certifications, applying a "reservation number standard" (so that CPI could track performance of lists supplied by various vendors) and creating a mail file. Ex. 2, Russell Dep. at 19-21. CPI's witnesses have conceded that this process changed SMG's data. Ex. 7, Hiller Dep. at 100; see also Ex. 2, Russell Dep. at 51-52; Ex. 4, Anderson Dep. at 163. 23. CPI asserts in its Response in Opposition that these processes would not

change the names and addresses. Response in Opposition, p. 7. In support of this assertion, CPI cites to the deposition testimony of Mr. Russell. A review of that testimony, however, reveals that CPI's assertion is incorrect. Mr. Russell testified that he did not know how PrimeNet's manipulation would change SMG's data. See Russell Dep. at 76-77.

-9-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 10 of 20

24.

Although CPI asserts that the PrimeNet data is an appropriate proxy for

SMG's records and that is excused from analyzing the actual data provided by SMG because it does not have that data, Response in Opposition, p. 4, those contentions are insufficient to carry its burden. 25. First, CPI has offered no evidence that the data from PrimeNet is a reliable

source from which its experts can analyze the quality or quantity of the names provided by SMG. To the contrary, CPI's own expert, Mr. Hiller, questioned whether the "reservation number," one of the manipulations that PrimeNet performed on SMG's data, was being applied correctly to the data. Ex. 7, Hiller Dep. at 23-25. If the "reservation number" was applied incorrectly by PrimeNet, then there is no way of knowing which vendor supplied which information to CPI. The only argument offered by CPI in its Response in Opposition to refute this is a conclusory statement in Mr. Hiller's Affidavit. Such statements, however, do not substitute for testimony of the person who actually manipulated and prepared the data. 26. Second, CPI has sued PrimeNet, asserting that PrimeNet's work was

inadequate and incompetent. Ex. 3, Grace Dep. at 158. CPI has offered no evidence to reconcile these claims with its assertion that the PrimeNet data is a reliable proxy for the SMG data.2 27. Third, CPI asserts that because it no longer has SMG's files and did not

receive them in discovery, it is appropriate for CPI's experts to use files manipulated by PrimeNet instead. Response in Opposition, p. 4. Notably, however, CPI did not ask SMG for this information in the discovery process. Certainly CPI has not filed a motion to compel or to inspect SMG's database.

2

There are no witnesses from PrimeNet listed on CPI's witness list.

-10-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 11 of 20

28.

Thus, Mr. Hiller and Mr. Allen's testimony must be excluded because it

does not comport with the requirements of Rule 702 that testimony must be based upon "sufficient facts or data" and that the data used is reliable. See The Ninth Avenue Remedial Group, 141 F. Supp. 2d at 960 (in the absence of any evidence of the reliability of the data relied upon by an expert, the expert's testimony is not based upon sufficiently reliable data and must be excluded). 2. The Analysis Encompassed Only a Fraction of the Names SMG Provided to CPI 29. Mr. Hiller and Mr. Allen considered only a fraction of the nearly 9 Million

records SMG provided to CPI during a 12 month period covered by the Renewal Agreement. Messrs. Hiller and Allen studied only a five-month portion of the twelve month period covered by the Renewal Agreement. During that five-month time period, SMG supplied CPI with 3.38 million names. See attachment A of Ex. 1 to SMG's Motion for Summary Judgment which reflects the total number of records SMG supplied to CPI on a monthly basis. The analysis of Mr. Hiller and Mr. Allen, however, attributes only 1.32 million names to SMG. Ex. 7, Hiller Dep. at 69, 74-75. 30. The discrepancy between the number of names supplied by SMG and the

number studied is because PrimeNet apparently assigned certain SMG records, including (1) common records between SMG and ASL and (2) SMG records CPI used for its "creative test mailings," to a "reservation number standard" that does not credit SMG with those records. Ex. 2, Russell Dep. at 32-34, 60-61; Ex. 5, Allen Dep. at 118-22. This reservation number assignment eliminates nearly two-thirds of SMG's records from the already limited data analyzed by CPI's experts.

-11-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 12 of 20

31.

CPI asserts that SMG's lists were of lower quality than other list vendors

(specifically American Student List ("ASL")), but the analyses of Messrs. Allen and Hiller are flawed because neither CPI nor its experts offer any basis for how these records (more than 2 million) compare to the records CPI does attribute to SMG. Specifically, records that SMG would have in common with another list vendor are more likely to be "good" records since both vendors have them. Because the methodology used by Messrs. Allen and Hiller does not attribute these records to SMG but instead credits them to ASL, their analysis undercounts "good" names provided by SMG by an unknown amount. Ex. 5, Allen Dep. at 118-22. 32. Messrs. Allen and Hiller also failed to set forth any reason as to why or

how their conclusions could be extrapolated to all of the data SMG provided to CPI. In CPI's Response in Opposition, CPI asserts that that Mr. Allen and Mr. Hiller chose to analyze mailings for a 23-week period in 2004 "because these mail files were available." Response in Opposition, p. 6. CPI also asserts that Mr. Hiller's opinion is "based on his experience and opinion that the 8 million record sample is sufficient, and SMG provides no reason why 8 million records in an insufficient sample." Notably, CPI offers no evidence for this conclusion, the Hiller Affidavit, which was prepared specifically to rebut SMG's motion, is silent on the issue. 33. Again, CPI has not established the reliability of the facts underlying its

own expert testimony, as it must. Dodge, 328 F.3d at 1221. C. The Principles and Methodology Applied By Allen and Hiller are Unreliable 34. Rule 702 also requires that both the expert's methodology be reliable and

the application of such principles and methodology is reliably applied to the facts of the case. See Mitchell, 165 F.3d at 781. In evaluating the methodology employed by an expert, the court should consider whether the principles and methods have been properly applied to the facts. See

-12-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 13 of 20

id. at 782-83. Under Rule 702, "any step that renders the analysis unreliable . . . renders the expert's testimony inadmissible. This is true whether the step completely changes a reliable methodology or merely misapplies that methodology." See id. at 782 (quoting In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 745 (3rd Cir. 1994)). 35. Testimony which constitutes "ipse dixit" of the expert should be excluded.

See Mitchell, 165 F.3d at 782. An expert cannot establish a fact merely by saying so. See In re Breast Implant Litigation, 11 F. Supp. 2d 1217, 1234 (D. Colo. 1998) (citing Grimes v. Hoffmann-LaRoche, Inc., 907 F.Supp. 33, 38 (D.N.H.1995)). Thus, an expert's opinion connected to the existing data only by the unsupported assertion of the expert is not reliable and should be excluded. See id. 1. 36. The CASS-Certification Analysis is Unreliable To analyze the quality of SMG's records, Mr. Allen performed a "CASS

certification" using a software program designed to match addresses with those recognized by the U.S. Post Office. The records that do not match U.S. Postal addresses are designated as "CASS failures." Ex. 5, Allen Dep. at 41-44. Because postal addresses are constantly changing, the U.S. Post Office updates its database monthly. Ex. 7, Hiller Dep. at 123. 37. The significance of the CASS certification process to conclusions

regarding the quality of SMG's data is itself questionable since CPI's witnesses have testified that an address that fails the CASS-certification process may still be "deliverable." See Ex. 7, Hiller Dep. at 53-58, 79-83; Ex. 1 to SMG's Reply to its Motion for Summary Judgment, Anderson Dep. at 79-83. Nonetheless, there is a significant flaw in CPI's CASS certification analysis.

-13-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 14 of 20

38.

Mr. Allen performed his CASS certification process on data supplied to

CPI for mailings beginning in January 2004. Ex. 5, Allen Dep. at 42-44. To perform the certification, however, Mr. Allen, inappropriately used a program that was updated several times thereafter to be current as of December 2004. Id. at 42-44. This difference in time is significant because the "CASS certification" program updates moves and tracks changes in addresses in the U.S. Post Office. Id. Thus, to get accurate results regarding the number of "CASS failures," the date of the data should be approximately the same as the date of the program. Id. at 42-44; Ex. 7, Hiller Dep. at 124-26. 39. Messrs. Allen and Hiller conceded that the mismatch in time between the

date the records were mailed by CPI and the date of the CASS program would create an unknown error rate and that the results of their "analysis" would be different had the proper version of the program been used. Ex. 5, Allen Dep. at 42-44; Ex. 7, Hiller Dep. at 124-26. 40. While CPI attempts to minimize the significance of this mistake by

asserting that it would be unusual or rare for an address to change, Response in Opposition, p. 6, CPI's explanation is unsupported by any evidence, is inconsistent with the testimony of its own experts and defies common sense. Certainly the U.S. Postal Service would not update the database on a monthly basis if the changes over time were immaterial. 41. CPI contends that because the CASS certification software is "industry

standard," the expert analysis must be admissible. This contention misses the mark since it ignores the flaw in the methodology utilized by CPI's experts, namely applying newer software to older data. Because Mr. Allen used the wrong version of the computer program for the CASS certification, testimony relating to the number of SMG's records that allegedly failed CASS certification is unreliable and must be excluded.

-14-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 15 of 20

2. 42.

The Duplicate Analysis is Unreliable In an effort to establish that SMG's records were unreliable and contained

a significant number of duplicate records, CPI relies principally on the analysis done by Mr. Allen. CPI, however, has not established that the program used by Mr. Allen to identify duplicates is reliable or among those used in the field. CPI also failed to establish that pairing the reservation number standard applied by PrimeNet with the computer program used by Mr. Allen yields reliable results. (a) Mr. Allen's Inexperience with the Computer Program Used to Conduct his Duplicate Analysis

43.

Mr. Allen had never before sorted duplicates from a database using the

SoundEx computer program he relies on here. Ex. 5, Allen Dep. at 35.-38. Mr. Allen used this program only because his boss selected it, id. at 35-37, but Mr. Allen was retained by CPI individually and neither his boss nor his employer is designated by CPI as witnesses in this case. Mr. Allen had very little experience with the program and this was the first time he used the particular sound duplication software in a project. Id. at 36-37. 44. More importantly, prior to using the program here, Mr. Allen did not do

any analysis to determine whether the program was appropriate for identifying duplicate names. Id. at 111. Mr. Allen was also unaware of anyone else who used the program for such a purpose and was not aware of its error rate. Id. Likewise, Mr. Allen acknowledged that the SoundEx program could create "false duplicates." See Ex. 5A attached hereto, Allen Dep. at 147-48. Although CPI attempts to salvage the duplicate analysis by making a passing reference to the use of SoundEx by the U.S. Bureau of Census, Response in Opposition, p. 10, such attempt fails because it provides no evidence as to how the program was used, why it was used, when it was used, how often it was used, how long ago it was used, or its rate of error.

-15-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 16 of 20

45.

Mr. Allen also admitted that he knew nothing about the source of the data

that he analyzed, whether it contained data from multiple vendors, who applied the "identifier codes" (a/k/a "reservation numbers") that he used to categorize the data or whether the data had been manipulated by PrimeNet before it had been provided to him. Id. at 30-31, 59. 46. CPI has offered no evidence to refute Mr. Allen's deposition testimony

related to his lack of experience with the SoundEx program or his failure to make an independent assessment as to the appropriateness of the program in this situation. (b) Limitations on the Analysis Based on Reliance on PrimeNet's Application of the Reservation Number

47.

In addition to the fundamental questions regarding the duplicate analysis

performed by CPI's experts, CPI's reliance on PrimeNet's application of the reservation number is likewise unreliable. CPI retained PrimeNet to act as its "mailhouse." PrimeNet's services included combining the information provided by the various list vendors, applying codes that purportedly allowed CPI to attribute the source of the names and addresses to a particular vendor, eliminating duplication arising from names that appeared on more than one vendor's list, running the names and addresses through a computer program designed to ensure the addresses were recognized by the Post Office, and preparing the direct mail packages and sending the mail. See generally, Ex 1. 48. When PrimeNet assigned reservation numbers to allow CPI to track the

source of names, it assigned the letters A through E to records provided by ASL and it assigned F and G to records provided by SMG. Ex. 7, Hiller Dep. at 24; Ex. 2, Russell Dep. at 27-29, 60. 49. On occasion, PrimeNet also would send creative test mailings at the

direction of CPI. See, e.g., Ex. 2, Russell Dep. at 30-32. CPI would send these creative tests to determine whether various characteristics of the mailings (color of envelope, greeting, etc.) -16-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 17 of 20

would have an impact on the response rate. Id. When these creative tests were mailed, CPI would generally instruct SMG to use records supplied by ASL and SMG. Id. at 31. When creative tests were mailed, the PrimeNet was instructed to assign a reservation number that would tell CPI the feature of the mailing being tested rather than the particular vendor that supplied the records. Id. at 31-33. CPI has no way of knowing how many SMG records were used for creative test pieces and whether any of these particular records were duplicates or failed CASS certification. Id. Mr. Allen concedes that his analysis did not take into account that one vendor was assigned multiple alphanumeric designations. Ex. 5, Allen Dep. at 116-21. 50. When Mr. Allen analyzed PrimeNet's data for duplicates, he did not

determine whether duplicates existed between the categories defined by the alpha characters in the reservation number. Accordingly, his analysis fails to determine whether there were duplicate records provided by ASL but assigned different letters in the reservation number standard; for example, if ASL provided duplicate names that were assigned to categories A and B, Mr. Allen would not count them as duplicates. Because ASL had five different letters assigned to it and SMG had only two letters assigned to it, this could significantly undercount the number of duplicates provided by ASL. Mr. Allen conceded that, in fact, he did not capture all of the duplicates provided by each vendor and would, for example, undercount duplicates provided by ASL "by who knows how much." Id. at 116-21. 51. While CPI asserts that SMG should not complain about an error that

undercounts duplicates, Response in Opposition, pp. 6-7, that assertion ignores the fact that CPI is comparing SMG's percentage of duplicates to ASL's and that comparison is fundamentally flawed. 52. Furthermore, the reservation number standard, the letter assigned to each

record depended, in part, upon whether an ASL record was duplicated by an SMG record, in -17-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 18 of 20

which case it was assigned a letter reflecting that it was a "common" name as opposed to a "unique" name. ASL was given "credit" for all the names it had in common with SMG. Ex. 2, Russell Dep. at 27-29. Mr. Allen's analysis does not take this into account. Id.; Ex. 8, Allen's Report. This too would affect, in an unknown way, the relative number of duplicate names assigned to SMG versus ASL. CPI does dispute that this application of the reservation number standard would have an effect on the percentage of SMG's claimed duplicates. 3. 53. Mr. Hiller's Opinions on the Marketing Materials is Unreliable Mr. Hiller applied no discernable methodology in rendering his opinion

that SMG's marketing materials are misleading. He admittedly did no research regarding the underlying facts or the actual records supplied by SMG. Ex. 7, Hiller Dep. at 81, 82-85, 88, 11213, 115. His opinions are concededly based upon fragments of the materials as he testified that the guarantee is not misleading when viewed in its entirety. Id. at 95-96. Accordingly, Mr. Hiller's testimony about SMG's marketing materials is pure ipse dixit. Such testimony is inadmissible as ipse dixit. See Mitchell, 165 F.3d at 782 (excluding expert testimony where the opinion evidence was connected to the data only by the ipse dixit of the expert). V. CONCLUSION

For each of the foregoing reasons and for those set forth in SMG's Motion in Limine and accompanying Memorandum, SMG respectfully requests that this Court grant SMG's Motion in Limine to Exclude the Expert Testimony of Mr. Nathan Allen and Mr. Thomas Hiller. Dated: July 28, 2005 Respectfully submitted: s/R. Daniel Scheid ______________ LEWIS SCHEID LLC R. Daniel Scheid -18-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 19 of 20

River Point Building 2300 Fifteenth Street, Suite 320 Denver, CO 80202 Telephone: (303) 534-5040 Facsimile: (303) 534-5039 KIRKPATRICK & LOCKHART NICHOLSON GRAHAM LLP Patrick J. McElhinny, Esquire Dianna S. Karg, Esquire 535 Smithfield Street Henry W. Oliver Building Pittsburgh, PA 15222 Telephone: (412) 355-6500 Facsimile: (412) 355-6501 Counsel for Plaintiff, Student Marketing Group, Inc.

-19-

Case 1:04-cv-01258-LTB-BNB

Document 105

Filed 07/28/2005

Page 20 of 20

CERTIFICATE OF SERVICE I hereby certify that on this 28th day of July, 2005, I electronically filed the foregoing REPLY MEMORANDUM IN SUPPORT OF PLAINTIFF'S MOTION IN LIMINE TO EXCLUDE THE EXPERT TESTIMONY OF MR. NATHAN ALLEN AND MR. THOMAS HILLER with the Clerk of the Court using the CM/ECF system which will send notification of such filing to the following e-mail addresses: Rosemary Orsini, Esquire Brian Matise, Esquire BURG, SIMPSON, ELDREDGE, HERSH, JARDINE, P.C. [email protected] [email protected]

s/Claudia Cooper