Summary
finding focus group materials unreliable and not probative in part because of lack of information regarding methodologies
Summary of this case from Saxon Glass Techs., Inc. v. Apple Inc.Opinion
[Copyrighted Material Omitted]
Abraham C. Reich, Gregory B. Williams, Beth Lillian Domenick, Fox Rothschild O'Brien & Frankel LLP, Philadelphia, PA, Eugene F. Assaf, Gregg F. Locascio, Jeffrey G. Landis, Jonathan Daniel Brightbill, Kenneth S. Clark, Valerie Gutmann, Kirkland & Ellis LLP, Washington, DC, Jordan M. Heinz, Thomas M. Monagan, III, Kirkland & Ellis LLP, Chicago, IL, for Plaintiff.
Alfred W. Putnam, Jr., Drinker Biddle & Reath LLP, Philadelphia, PA, Brian N. Lasky, Clay J. Pierce, Jeffrey D. Rotenberg, Jennifer L. Higgins, Melissa A. Mandrgoc, Steven A. Zalesin, Alexis Gander Deise, David G. Sewell, Karla G. Sanchez, Rosa E. Son, Travis J. Tu, Patterson Belknap Webb & Tyler LLP, New York, NY, Andrea L. D'Ambra, David J. Kessler, Drinker, Biddle & Reath, LLP, Philadelphia, PA, for Defendants.
Memorandum and Order
PRATTER, District Judge.
Merisant Company, Inc. (" Merisant" ) alleges that McNeil Nutritionals, LLC and McNeil-PPC, Inc. (collectively, " McNeil" ) engaged in false and misleading advertising with respect to Splenda No Calorie Sweetener in violation of Section 43(a) of the Lanham Act, 15 U.S.C. § 1125, and the Pennsylvania common law of unfair competition. Jury selection proceeded on April 9, and counsel presented opening arguments on April 10, 2007. The parties engaged in extensive discovery activities and exchanged pretrial memoranda several weeks before trial began, leading to their submission of numerous motions in limine. In addition, Merisant moved to exclude surveys and related testimony from two of the McNeil's expert witnesses, arguing that such evidence is inadmissible under Rule 702 of the Federal Rules of Evidence and pursuant to Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), and its progeny. Merisant filed a Motion to Exclude the Survey Conducted by Susan S. McDonald and all related testimony and opinion (Docket No. 110), and a Motion to Exclude the Survey Conducted by Robert L. Klein and all related testimony and opinion (Docket No. 112). In addition, Merisant filed separate motions in limine seeking to exclude Mr. Klein's and Dr. McDonald' surveys on additional grounds (Docket Nos. 181, 182). Finally, McNeil filed a motion in limine to exclude as evidence at trial certain third-party surveys (Docket No. 167). The Court will address each of these motions in turn.
STANDARDS
A. Standards for Daubert Motions
Federal Rule of Evidence 702, which governs the admissibility of expert testimony, provides:
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 by knowledge, skill, experience, training, or education, 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.
In Daubert, the Supreme Court imposed upon district courts the role of a gatekeeper, in order to " ensure that any and all scientific evidence is not only relevant, but reliable." ID Sec. Sys. Canada, Inc. v. Checkpoint Sys., Inc., 198 F.Supp.2d 598, 601-02 (E.D.Pa.2002) ( quoting Daubert, 509 U.S. at 589, 113 S.Ct. 2786). When " faced with a proffer of expert scientific testimony ... the trial judge must determine at the outset, pursuant to Rule 104(a), whether the expert is proposing to testify to (1) scientific knowledge that (2) will assist the trier of fact to understand and determine a fact in issue." Id. at 602 ( quoting Daubert, 509 U.S. at 592, 113 S.Ct. 2786). This gatekeeping function of the district court extends beyond scientific testimony to " testimony based on ... ‘ technical’ and ‘ other specialized’ knowledge." Id. (quoting Kumho Tire Co. v. Carmichael, 526 U.S. 137, 141, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999)).
Federal Rule of Evidence 702 provides " three distinct substantive restrictions on the admission of expert testimony: qualifications, reliability and fit." Id. (quoting Elcock v. Kmart Corp., 233 F.3d 734, 741 (3d Cir.2000)). The party offering the expert testimony has the burden of establishing that the proffered testimony meets each of the three requirements by a preponderance of the evidence. Id. (citing Padillas v. Stork-Gamco, Inc., 186 F.3d 412, 418 (3d Cir.1999)).
The first requirement, whether the witness is qualified as an expert, has been interpreted liberally to encompass " a broad range of knowledge, skills, and training." Id. (quoting In re Paoli R.R. Yard PCB Litigation, 35 F.3d 717, 741 (3d Cir.1994)).
The second prong requires the expert's testimony to be reliable. Id. When the expert testifies to " scientific knowledge," the expert's opinions " must be based on the ‘ methods and procedures of science’ rather than on ‘ subjective belief or unsupported speculation’ ; the expert must have ‘ good grounds' for his or her belief." Id. (quoting In re Paoli, 35 F.3d at 742). In considering whether there are " good grounds" for the expert's opinions, district courts should look at a series of factors:
(1) whether a method consists of a testable hypothesis; (2) whether the method has been subject to peer review; (3) the known or potential rate of error; (4) the existence and maintenance of standards controlling the technique's operation; (5) whether the method is generally accepted; (6) the relationship of the technique to methods which have been established to be reliable; (7) the qualifications of the expert witness testifying based on the methodology; and (8) the non-judicial uses to which the method has been put.
Id. (quoting In re Paoli, 35 F.3d at 742 n. 8). This list of factors " is non-exclusive and ... each factor need not be applied in every case." Id. (quoting Elcock, 233 F.3d at 746). The Supreme Court has noted that the district court " must have considerable leeway in deciding in a particular case how to go about determining whether particular expert testimony is reliable. That is to say, a trial court should consider the specific factors identified in Daubert where they are reasonable measures of the reliability of expert testimony." Id. (quoting Kumho Tire, 526 U.S. at 152, 119 S.Ct. 1167).
The final prong requires that the expert testimony " fit" by assisting the trier of fact. Id. (citing Oddi v. Ford Motor Co., 234 F.3d 136, 145 (3d Cir.2000)). " Admissibility thus depends in part upon ‘ the proffered connection between the scientific research or test result to be presented and particular disputed factual issues in the case.’ " Id. (quoting In re Paoli, 35 F.3d at 743). The " fit" standard does not require plaintiff to prove " their case twice." Id. (quoting Oddi, 234 F.3d at 145). They need not " demonstrate to the judge by a preponderance of evidence that the assessments of their experts are correct, they only have to demonstrate by a preponderance of evidence that they are reliable." Id. (quoting In re Paoli, 35 F.3d at 744). Thus, the test does not require that the opinion have " the best foundation" or be " demonstrably correct," but only that the " particular opinion is based on valid reasoning and reliable methodology." Id. (quoting Oddi, 234 F.3d at 146).
B. Standards for Motions in limine Pursuant to Rules 402 and 403
" Under the Federal Rules of Evidence, subject to certain limitations, all evidence is admissible if it is relevant, i.e., if it tends to make the existence or nonexistence of a disputed material fact more probable than it would be without that evidence." Forrest v. Beloit Corp., 424 F.3d 344, 355 (3d Cir.2005); see Fed.R.Evid. 401, 402. Pursuant to Rule 403, a court may nonetheless exclude relevant evidence if the probative value of the evidence is " substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time, or needless presentation of cumulative evidence." Forrest, 424 F.3d at 355 ( quoting Fed.R.Evid. 403). The Court of Appeals for the Third Circuit has described Rule 403 as an " ‘ umbrella rule’ spanning the whole of the Federal Rules of Evidence," and as such district courts must apply Rule 403 " in tandem with other Federal Rules under which evidence would be admissible." Id. (quoting Coleman v. Home Depot, Inc., 306 F.3d 1333, 1343 (3d Cir.2002)). A district court must articulate the Rule 403 balancing test, i.e., the probative value of potentially relevant evidence must be " carefully balanced, pursuant to Rule 403, against its possible prejudicial effect." Id., 424 F.3d at 356. " ‘ Rule 403 recognizes that a cost/benefit analysis must be employed to determine whether or not to admit evidence; relevance alone does not ensure its admissibility.’ " Id. (quoting Coleman, 306 F.3d at 1343). However, in the Third Circuit there is a " strong presumption" that relevant evidence should be admitted. Id. A party arguing that evidence should be excluded under Rule 403, the party must show that " the probative value of evidence must be ‘ substantially outweighed’ by the problems in admitting it." Id. (quoting Coleman, 306 F.3d at 1343-44). However, " ‘ prejudice does not simply mean damage to the opponent's cause.’ " Goodman v. Pa. Turnpike Comm'n, 293 F.3d 655, 670 (3d Cir.2002) ( quoting 1 McCormick on Evidence § 185 at 645 (John W. Strong, et al. eds., 5th ed.1999)). The court acknowledged that only " unfair prejudice" can tip the scales, noting:
The ... prejudice against which the law guards [is] ... unfair prejudice-... prejudice of the sort which cloud[s] impartial scrutiny and reasoned evaluation of the facts, which inhibit[s] neutral application of principles of law to the facts as found.... Prejudice does not simply mean damage to the opponent's cause. If it did, most relevant evidence would be deemed prejudicial.
Ansell v. Green Acres Contr. Co., 347 F.3d 515, 525 (3d Cir.2003) ( quoting Wagenmann v. Adams, 829 F.2d 196, 217 (1st Cir.1987)).
DISCUSSION
Merisant filed a Motion to Exclude the Survey Conducted by Susan S. McDonald and all Related Testimony and Opinions (Docket No. 110), and Motion to Exclude the Survey Conducted by Robert L. Klein and all related Testimony and Opinions (Docket No. 112). In its motions, Merisant does not challenge Mr. Klein's or Dr. McDonald's respective qualifications, experience or credentials. In general, Merisant argues that Dr. McDonald's survey fails the " reliability" prong and that Mr. Klein's survey fails the " fit" and " reliability" prongs of the Daubert analysis.
The Federal Rules of Evidence " embody a strong and undeniable preference for admitting any evidence which has the potential for assisting the trier of fact." Kannankeril v. Terminix Int'l, 128 F.3d 802, 806 (3d Cir.1997). Rule 702 has a " liberal policy of admissibility." Id. In a Lanham Act case, where the fact-finder must determine whether an advertising claim is deceptive or misleading, courts have recognized that it may be that " the success of a plaintiff's false advertising claim will turn ‘ on the persuasiveness of a consumer survey.’ " Church & Dwight Co. v. S.C. Johnson & Son, 873 F.Supp. 893, 906 (D.N.J.1994) ( quoting Johnson & Johnson-Merck Consumer Pharm. Co. v. Rhone-Poulenc Rorer Pharm., 19 F.3d 125, 129-30 (3d Cir.1994)). The Court of Appeals for the Third Circuit has held that the proponent of a consumer survey has the burden of establishing that it was conducted in accordance with accepted principles of survey research. Pittsburgh Press Club v. United States, 579 F.2d 751, 758 (3d Cir.1978); see also Church & Dwight Co., 873 F.Supp. at 906. A court must consider several factors when determining whether a survey meets applicable standards:
A proper universe must be examined and a representativeA fortiori, sample must be chosen; the persons conducting the survey must be experts; the data must be properly gathered and accurately reported. It is essential that the sample design, the questionnaires and the manner of interviewing meet the standards of objective surveying and statistical techniques. Just as important, the survey must be conducted independently of the attorneys involved in the litigation. The interviewers or sample designers should, of course, be trained, and ideally should be unaware of the purposes of the survey or the litigation. the respondents should be similarly unaware.
Pittsburgh Press Club, 579 F.2d at 758.
Another court has noted that when a survey sample is selected for the purpose of generating data about a population to be offered for its truth " the methods used must conform to generally recognized statistical standards." Reynolds v. Giuliani, 118 F.Supp.2d 352, 366 (S.D.N.Y.2000) (citation omitted). The court considered other factors, including whether " the population was properly chosen and defined; the sample chosen was representative of that population; the data gathered was accurately reported; and the data was analyzed in accordance with accepted statistical principles." Id. In addition, the court stated that when a survey is being offered not for its truth but " in order to gauge opinions, attitudes and beliefs held by a population," which is a common usage for surveys in the false advertising context, such surveys " may be subject to additional scrutiny, especially with respect to how the survey questions were framed and presented." Id. at 366 n. 9.
Courts in the Third Circuit have generally held that a survey's " technical unreliability goes to the weight accorded a survey, not its admissibility." Citizens Fin. Group, Inc. v. Citizens Nat'l Bank, 383 F.3d 110, 121 (3d Cir.2004). Typically, a court will not exclude a survey unless it is so flawed that it would be completely unhelpful or harmful to the trier of fact. See, e.g., AHP Subsidiary Holding Co. v. Stuart Hale Co., 1 F.3d 611, 617 (7th Cir.1993) (noting that " any shortcomings in the survey results go to the proper weight of the survey and should be evaluated by the trier of fact" ). In Citizens Financial, the court of appeals affirmed the district court's exclusion of a survey because the survey's " methodology was fundamentally flawed and because the danger of undue prejudice far outweighed the limited probative value of the survey, especially for a jury." Citizens Financial, 383 F.3d at 121. In noting that courts have " generally" held that unreliability goes to the weight rather than the admissibility of the survey, the court of appeals nevertheless held that excluding the survey was proper because the district court concluded that the " survey did not suffer from mere technical flaws, but from fatal flaws." Id.; see also Spraying Systems Co. v. Delavan, Inc., 975 F.2d 387, 394 (7th Cir.1992) (affirming exclusion of a survey because it was flawed and did not create a genuine issue of material fact; the court found that the questions were biased and, because the case involved consumers' perception of visual trademarks, a telephone survey was inadequate).
In order for a survey to be admitted, its design must " fit" the issues to be decided in this case. In J & J Snack Foods Corp. v. Earthgrains Co., 220 F.Supp.2d 358, 370 (D.N.J.2002), the court stated that:
Above all, the survey's design must fit the issue which is to be decided by the jury, and not some inaccurate restatement of the issue, lest the survey findings inject confusion or inappropriate definitions into evidence, confounding rather than assisting the jury. Only if the expert testimony and related survey are useful, reliable, and have probative value after all their deficiencies are taken into account is the evidence admissible.
Id. In addition, the court noted that it was " essential to consider whether the population and terms were properly defined, whether the design, questionnaires, and interviews met objective standards, whether data was accurately collected and reported, whether data was properly analyzed, whether the questions asked were unrelated to the material issues of the case, whether questions were unfairly leading, and whether questions were confusing." Id. at 369. J & J Snack Foods addressed a dispute arising in the trademark context, where a survey was submitted to show the trademark should be classified as " suggestive" instead of " generic" or " descriptive." However, the survey did not properly define any of the classifications, and, consequently, the court found that the survey had " no bearing on the issue it was submitted for." Id. at 370. The court noted that " the flawed definition permeated the entire survey to make its finding completely untrustworthy and unreliable." Id. at 370-71.
I. Mr. Klein's Survey
A. Merisant's Daubert Motion to Exclude the Surveys Conducted by Robert L. Klein and all Related Testimony and Opinions (Docket No. 112)
Robert L. Klein conducted a survey that asked consumers why they purchased Splenda. Mr. Klein's survey was not designed or commissioned in anticipation for the current action but was designed and conducted to address certain class action lawsuits that had been brought against McNeil. Mr. Klein stated that his survey was designed to identify reasons why people purchased Splenda, and to determine whether there was a predominant belief that would support forming a viable class of plaintiffs. See Pl.'s Mot. Exclude Klein 3 n. 1 (Docket No. 112).
McNeil acknowledges that consumers have commenced numerous class action lawsuits against McNeil challenging its advertising for Splenda. Central to the various class complaints was the allegation that consumers who purchased Splenda bought it based on the mistaken belief that Splenda was " natural" or " contained sugar." Def. Mem. Opp'n (Klein) 2. In response, McNeil commissioned Mr. Klein to conduct a survey of consumers who had previously purchased Splenda inquiring as to their reasons for buying the product. Def. Mem. Opp'n (Klein) 2.
Merisant does not contest Mr. Klein's credentials or qualifications. Rather, Merisant argues that the survey Mr. Klein conducted is neither relevant nor reliable with respect to the issues presented in this case. Specifically, Merisant contends that Mr. Klein's survey was not designed to determine, nor should it be used at trial to demonstrate, the issue of consumer's " confusion" or " likelihood of confusion" when purchasing Splenda. Merisant argues that McNeil is seeking to use Mr. Klein's survey for this very reason.
Mr. Klein's survey asked respondents a series of open ended questions similar to the following: " Thinking back to when you first purchased Splenda ... What caused you to first purchase Splenda?" T. Monagan Decl. Ex. 3. The respondents to Mr. Klein's surveys were not exposed to any advertising, marketing materials or product packaging for Splenda. None of Mr. Klein's questions specifically addressed the issues of Splenda's advertising, marketing, packaging, logos, taglines, ingredients or sugar origins, or any potential likeness to " natural" ingredients.
Mr. Klein's questionnaire then instructed the interviewer to record the respondent's answer verbatim and to probe once with the question, " can you explain that further?" T. Monagan Dec. Ex. 3. Next, the respondent was asked " What other reasons, if any, caused you to first purchase Splenda?" Id. Again, the interviewer was instructed to record the answer verbatim and to probe once with the question " what else?" Id. Next, the respondent was asked " After you first purchased Splenda, why did you purchase it again?" Id. The questionnaire then instructed the interviewer to again follow up by asking, " can you explain that further?" Id. Next, the respondent was asked, " What other reasons, if any, caused you to purchase Splenda again?" This question was followed by the follow-up question, " what else?" Id.
McNeil presents several arguments that Mr. Klein's survey should be admitted. First, McNeil contends that the issue of " why people buy Splenda" is relevant to the issues in this case and Mr. Klein's survey informs the jury on that question. Next, McNeil argues that Mr. Klein's survey refutes the findings of one of a survey commissioned by Merisant's expert, Dr. Deborah Jay. Dr. Jay's survey, like Mr. Klein's, is not a " likelihood of confusion" survey. Rather, Dr. Jay's survey was designed to measure why consumers switched to Splenda, or why consumers chose Splenda over Equal. McNeil's final argument is that the Klein survey bears directly on the proper measure of damages, specifically with respect to any potential disgorgement of McNeil's profits. Merisant also contends that Mr. Klein's survey interviewed the wrong universe of people, namely, people who had purchased Splenda in the past, instead of prospective purchasers of Splenda. Mr. Klein's survey was created to determine whether a " class" of Splenda consumers could proceed in a class action lawsuit. Understandably, as to a survey that was designed to test whether a category of consumers constitutes a " class" for purposes of filing a lawsuit against a maker of a certain product, the only permissible universe for such a class must consist of past users of that product. In Lanham Act cases that seek to prove rates of " confusion" among consumers, however, it is typical to survey prospective consumers. See Universal City Studios, Inc. v. Nintendo Co., 746 F.2d 112, 118 (2d Cir.1984) (holding that a survey purporting to measure " confusion" was flawed because the survey utilized an improper universe in that it was conducted among individuals who had already purchased or leased the product rather than those who were contemplating a purchase or lease). McNeil has not fully explained why past users of Splenda would be the appropriate universe in this case.
McNeil maintains that at the damages phase, Merisant carries the burden of proving McNeil's " gains" from sales to consumers as a result of confusion. While plaintiff is responsible for proving the nexus between the false advertisement and any adverse purchasing decision, 15 U.S.C. § 1117(a) states that, " [i]n assessing profits the plaintiff shall be required to prove defendant's sales only; defendant must prove all elements of cost or deduction claimed." 15 U.S.C. § 1117(a). At the damages phase, McNeil would have to prove " costs or deductions" that would reduce Merisant's damages.
B. Merisant' Motion in limine No. 9 to Exclude Testimony of Robert L. Klein (Docket No. 182)
Merisant claimed that its motion in limine is intended to supplement its previous Daubert motion discussed above. In this motion, Merisant seeks to exclude Mr. Klein's survey under Rules 402 and 403 as irrelevant, and argues that any probative value Mr. Klein's survey may have is outweighed by the unfair prejudice Merisant would endure if it was admitted.
C. Analysis
McNeil intends to introduce Mr. Klein's survey to rebut the survey conducted Dr. Jay, which purported to measure why consumers switched from Equal to Splenda. Merisant seeks to exclude Dr. Klein's survey arguing that it is not a " confusion" survey, yet Dr. Jay's survey is not a " confusion" survey either. Merisant seeks to distinguish Dr. Jay's survey arguing that it is a " mistaken belief" survey. The Court does not find Merisant's argument persuasive.
The issue in this case is whether certain of McNeil's advertising for Splenda is false or misleading. However, notably, neither the Klein nor the Jay surveys exposed respondents to any Splenda advertising. Arguably, then, on this element neither of these surveys would be germane to the issue at hand. However, McNeil has not moved to exclude the Jay survey and, as such, a question as to the admissibility of that survey is not before the Court. While McNeil's argument that Mr. Klein's survey is " no less relevant" than the Jay survey is not terribly convincing, the Court will not exclude Mr. Klein's survey on the eve of trial, months after the discovery deadline in this case expired, leaving McNeil no evidence with which to rebut Dr. Jay's survey. The Klein survey is at least as instructive for the jury as the Jay survey.
To the extent either survey is relevant, Mr. Klein's survey is probative enough as rebuttal evidence to permit it to be introduced as evidence. Therefore, both Merisant's Daubert motion and its motion in limine to exclude Mr. Klein's survey will be denied.
II. Dr. McDonald's Survey
A. Merisant's Daubert Motion to Exclude the Survey Conducted by Susan S. McDonald and all Related Testimony and Opinions (Docket No. 110)
Dr. Susan S. McDonald designed and conducted her survey in response to a survey conducted by Dr. Henry Ostberg, Merisant's survey expert. Dr. McDonald's " goal" in designing her survey " was to demonstrate that a properly designed survey, without any of the methodological errors committed by Dr. Ostberg, would prove ... that the vast majority of consumers are not misled by Splenda packaging." S. Zalesin Decl. Ex 1 at 1.
In her May 2006 report, Dr. McDonald states that her sample consisted of 759 respondents. S. Zalesin Decl. Ex 1 at 6. The survey interviews were conducted from March 28 through April 7, 2006. Id. Dr. McDonald employed a " mall intercept" survey, where respondents were approached in 23 malls in various states. Id. Prospective respondents were intercepted by interviewers in the various malls, were asked several screening questions to determine eligibility, and, if they qualified for the survey, they were escorted to an office away from mall traffic where the interview was conducted. Id. at 7. Dr. McDonald employed a three-armed survey, where each respondent was assigned randomly to one of three products: (1) a 3.8 ounce box of granular Splenda; (2) a 3.5 ounce box containing 100 Splenda packets; or (3) a 3.5 ounce box of NutraSweet packets. Id. at 6-7. The respondents were not exposed to the packages themselves, but rather were exposed to a picture of the actual box on a computer screen. Id. at 9.
Dr. McDonald's survey began by showing the respondent one of the three packages noted above, and asking the respondent to view it as if they were considering it for purchase. Next, respondents were asked the following questions:
Q.1a. How would you describe this product, in your own words?
Q.1b. What do you think are the major characteristics of this product?
Probe: Any others?
Def. Mem. Opp'n (McDonald) 2. The respondent was then asked a " filter question" designed to exclude from further questioning anyone who did not feel that the package communicated anything about the ingredients in the product. S. Zalesin Decl. Ex 1 at 10. The filter question in Dr. McDonald's survey was " Q.2: Does the packaging communicate anything to you about the ingredients in this product, or not?" Id. Only respondents who answered " yes" to the filter question where then asked the following question:
Q.3. Please take a look at the statements on this card. Which of these statements do you think most accurately described this product?
• This product is a no-calorie sweetener that contains real, natural sugar
• This product starts with real, natural sugar but is processed into a no-calorie sweetener that does not contain real, natural sugar
• This product is a no-calorie sweetener that does not start with real, natural sugar and does not contain real, natural sugar
• Don't know/can't tell.
S. Zalesin Decl. Ex 1 at 10.
Dr. McDonald states that an attempt was made to validate all of the interviews. Id. at 11. Validation is a method whereby, in this case, Dr. McDonald commissioned an independent survey organization to contact every respondent whose interview had not been confirmed onsite. Dr. McDonald's team originally commissioned " about" 878 interviews, with the goal of ensuring that there would be approximately 750 at the conclusion of the survey. Id. at 12. Of those, 323 were successfully validated. Id. at 12-13. Forty-four (44) interviews were discarded because the respondents denied having been interviewed or appeared not to have qualified. Id. at 13. In addition, 31 interviews from the Kansas City site were removed from the sample because several interviews were not validated and 40 additional interviews (including all thirty (30) from the Philadelphia site) were removed from the sample because the interviewers did not follow instructions. Id. at 13. In total, the results of 115 were rendered void and were not included in the final sample.
In McNeil's opposition brief on this motion, it stated that an additional five (5) interviews were discarded as a result of telephone validation. Def. Opp'n (McDonald) 7 n. 5.
In Dr. McDonald's survey, 25.5 percent of the respondents selected the response that Splenda " contains real, natural sugar." The comparable figure for the NutraSweet " control" arm of the survey was 17.6 percent. Dr. McDonald states that to obtain a true measure of the " confusion" caused by the Splenda packages, one must subtract the percent of " confused" respondents from the NutraSweet arm from the percent of " confused" respondents from the two Splenda arms. Therefore, Dr. McDonald concludes, " the ‘ true’ level of confusion is a modest 7.9 percent." Id. at 23. Merisant argues that the Court should exclude Dr. McDonald's survey and testimony because the survey is " so methodologically flawed and so poorly administered that it cannot possibly be relied upon." Pl. Mem. Supp. (McDonald) 3. Merisant identifies the following alleged problems with Dr. McDonald's survey: (1) Dr. McDonald improperly destroyed approximately 14 percent of the responses to her survey; (2) an insufficiently low percentage of survey responses were validated; (3) Dr. McDonald provided insufficient oversight in conducting the survey and developing the results; and (4) the survey is unreliable because Dr. McDonald adopted " NutraSweet" (which Merisant argues is known as an " ingredient" rather than as a table-top sweetener) as the control stimulus instead of Equal. As to this final point, Merisant notes that the " improper" choice of a NutraSweet as the control resulted in reducing the " confusion" percentage from approximately 26 percent to 8 percent, a difference of approximately 18 percent, to purportedly account for " noise."
McNeil claims that it seeks to offer Dr. McDonald's surveys and testimony to rebut surveys and testimony from Merisant's expert, Dr. Ostberg. McNeil claims Dr. Ostberg planted " the notion that Splenda is ‘ natural’ or ‘ contains sugar’ " " through a series of suggested closed-ended questions." Def. Mem. Opp'n (McDonald) 2. McNeil has not sought to exclude Dr. Ostberg's testimony but devotes a fair portion of is opposition brief attacking Dr. Ostberg's surveys. McNeil argues that Dr. McDonald's surveys " affirmatively disprove[ ]" Merisant's case.
A control is a way of measuring and rooting out whatever error potential, or " noise" exists, that is, something other than what is being measured in the first place. SmithKline Beecham Consumer Healthcare, L.P. v. Johnson & Johnson-Merck Consumer Pharms. Co., No. 01 Civ. 2775(DAB), 2001 WL 588846, at *12, 2001 U.S. Dist. LEXIS 7061, at *39 (S.D.N.Y. June 1, 2001). One type of " noise" that is particularly germane to the false advertising context is preexisting beliefs about what an ad in a particular category is going to communicate. Id.
McNeil counters that Merisant's criticism go to the weight of the evidence and not to its admissibility. McNeil argues that (1) the fact that some interviews were destroyed after Dr. McDonald determined that they were invalid does not invalidate the survey; (2) approximately 40 percent of the responses were validated, which is customary; (3) Dr. McDonald appropriately delegated; and (4) Dr. McDonald's choice of NutraSweet as a control was proper (and revealed a flaw in Dr. Ostberg's survey). As to Dr. McDonald's choice of NutraSweet as a control, McNeil argues that the choice of control goes to the weight of a survey and not to its admissibility, and argues that NutraSweet was appropriate because it is less well known, and because consumers have fewer preconceptions than with Equal, which, according to McNeil, has more " baggage."
The Court finds that, although Dr. McDonald's methods and oversight were far from perfect, as described by the parties in their papers, the survey is not so flawed as to render it inadmissible. Further, Merisant offers virtually no case law in support of its various arguments that Dr. McDonald's survey should be inadmissible. In particular, Merisant offers no case law supporting its argument that discarding data can render a survey inadmissible in court. Merisant bases most of its arguments with respect to Dr. McDonald's survey on the Federal Judicial Center's Reference Guide on Survey Research, which provides that experts should report and produce all data for all completed interviews. The Court does not doubt that this is the most professional approach. However, this shortcoming certainly can be raised at trial.
McNeil claims that discarding the data was an accident, but, nevertheless, the destroyed data pertained to interviews that Dr. McDonald had already invalidated. As explained, this data would not have been included in the survey results in the normal course.
With respect to the validation issue, the Court does not doubt that validation is indeed important. However, McNeil argues that Dr. McDonald validated over 40 percent of her interviews, and that this percentage exceeds the customary norms. Merisant cites no legal precedent for its assertion that a 50 percent validation rate is customary for litigation surveys; nor does it cite a single case where a court excluded a survey because of a subpar validation rate.
Merisant's argument that Dr. McDonald provided insufficient supervision also has little merit. Merisant cites to a number of examples from Dr. McDonald's deposition in an effort to show that Dr. McDonald displayed an insufficient level of control over the survey and the validation process, and that she had inadequate knowledge of the survey in general. Merisant cites one case where a court excluded a survey because it found that the survey lacked " sufficient indicia of trustworthiness." Toys " R" Us, Inc. v. Canarsie Kiddie Shop, Inc., 559 F.Supp. 1189, 1205 (E.D.N.Y.1983). In that case, the court also faulted the head of the company that conducted the survey for insufficient oversight. Specifically, the court noted that deficiencies in the survey data included the survey administrator's admission that he was unfamiliar with the industry practice, if any, for validating surveys. In addition, although it was important for the integrity of the survey for the interviewers not to know that the survey was being conducted by Toys " R" Us, at least a couple of the interviewers seemed to know this fact; a significant number of interviews were conducted in a bowling alley; there was evidence that some respondents were present while another respondent was being interviewed, such that they overheard the substance of the interview and were then interviewed immediately thereafter; and there was evidence that there were inaccuracies in one of the computerized tables designed to reflect confusion, and that responses were inaccurately recorded. Id. at 1204. In other words, there were numerous, significant flaws in the survey in Toys " R" Us that, taken for their cumulative effect, rendered the survey untrustworthy, which is not the case here.
Finally, Merisant notes that its expert, Dr. Ostberg, used Equal as the control and found a 32 percent rate of confusion. He adjusted that result downward by 4 percent to account for " noise" from the Equal control, which resulted in a total confusion rate of 28 percent. If one reviews Dr. McDonald's results without subtracting the results from the control group, Dr. McDonald's survey produced a confusion rate of approximately 26 percent, which is not so foreign to the results from Dr. Ostberg's survey. Yet, if these results are adjusted downward by 18 percent due to the control, Dr. McDonald's survey returned only an 8 percent confusion rate. It is not surprising that Merisant is unhappy with the results of Dr. McDonald's survey, but, even considering all of these alleged flaws cumulatively, Merisant has not convinced the Court that Dr. McDonald's survey entirely lacked " sufficient indicia of trustworthiness," Toys " R" Us, 559 F.Supp. at 1205, or that it is so flawed that it would be completely harmful, distracting or confusing to the jury, see AHP Subsidiary Holding Co. v. Stuart Hale Co., 1 F.3d 611, 617 (7th Cir.1993).
Both parties have conceded that a confusion rate of 15 percent is sufficient for Lanham Act purposes.
Merisant's objections to the technical validity of Dr. McDonald's survey go to the weight to be accorded to the survey rather than to its admissibility. As such, Merisant's Daubert motion to exclude Dr. McDonald's survey will be denied.
B. Merisant's Motion in limine No. 8 to Exclude All Survey Data Collected by Dr. Susan McDonald After the Close of Expert Discovery (Docket No. 181)
Merisant filed a second motion seeking to exclude additional research Dr. McDonald performed after the expert discovery deadline had elapsed, after her deposition, and after Merisant filed the Daubert motion discussed above. As Merisant notes, Rule 26(a)(2)(B) of the Federal Rules of Civil Procedure requires disclosure of all opinions to be expressed by the expert and disclosure of all data underlying the expert's research. Merisant argues that six months after the period for expert discovery closed, months after her deposition, and only after Merisant filed a Daubert motion seeking to exclude her survey, Dr. McDonald went back into the field to conduct additional research. Merisant argues that McNeil still has not produced the data underlying the additional surveys. Merisant alleges bad faith on McNeil's part, both in violating the Court's Scheduling Order that established the deadline for expert discovery, deceiving Merisant in requesting additional time to produce expert discovery, and in producing a " preliminary" expert report that it could later tailor to fit McNeil's litigation aims by supplementing it. Merisant argues that the after-deadline-surveys should be excluded under Federal Rule of Civil Procedure 37(c)(1).
Rule 37(c)(1) provides that " [a] party that without substantial justification fails to disclose information required by Rule 26(a) or 26(e)(1), or to amend a prior response to discovery as required by Rule 26(e)(2), is not, unless such failure is harmless, permitted to use as evidence at a trial, at a hearing, or on a motion any witness or information not so disclosed." Fed.R.Civ.P. 37(c)(1). Courts in this Circuit consider four factors in determining whether the exclusion of evidence is an appropriate sanction for failure to comply with discovery duties: " (1) the prejudice or surprise of the party against whom the excluded evidence would have been admitted; (2) the ability of the party to cure that prejudice; (3) the extent to which allowing the evidence would disrupt the orderly and efficient trial of the case or other cases in the court; and (4) bad faith or wilfulness in failing to comply with a court order or discovery obligation." Nicholas v. Pa. State Univ., 227 F.3d 133, 148 (3d Cir.2000); Konstantopoulos v. Westvaco Corp., 112 F.3d 710, 719 (3d Cir.1997) (affirming exclusion of plaintiff's expert witness because producing expert after the expert discovery deadline constituted a " flagrant disregard" of the court's pretrial order, and defendant would be prejudiced). However, " the importance of the excluded testimony" should be considered. Meyers v. Pennypack Woods Home Ownership Ass'n, 559 F.2d 894, 904 (3d Cir.1977). " The exclusion of critical evidence is an ‘ extreme’ sanction, not normally to be imposed absent a showing of wilful deception or ‘ flagrant disregard’ of a court order by the proponent of the evidence." Id. at 905 ( quoting Dudley v. South Jersey Metal, Inc., 555 F.2d 96, 99 (3d Cir.1977)).
In all fairness to McNeil, Merisant can hardly claim that it is surprised or unable to cure any potential prejudice that would result from including Dr. McDonald's additional surveys because Merisant has known that such research was conducted since December 2006 and did not seek to depose Dr. McDonald again prior to bringing the present motion in limine. However, in fairness to Merisant, McNeil has not disclosed the results of these additional surveys to Merisant.
The Court does not find that McNeil engaged in any bad faith effort to surprise Merisant with last-minute evidence, inasmuch as Dr. McDonald seemed to have conducted such additional research in response to one of Merisant's challenges to the technical validity of her survey. In that regard, McNeil maintains that this additional research was not actually necessary-in other words, it stands by the unaltered results of Dr. McDonald's original survey-and would only become necessary if and when Merisant argues that the fact that Dr. McDonald (or her staff) discarded 115 interviews discredits her results.
The Court will credit McNeil's assertion that Dr. McDonald's research is not actually necessary to its case in chief and will grant Merisant's motion to exclude this additional research. However, should Merisant on cross-examination seek to impugn the technical validity of Dr. McDonald's survey on grounds inexorably leading the witness to make reference to the 115 discarded surveys or to the " new" survey " results," thus opening the door for this evidence to be used to rebut those charges, McNeil will be permitted to offer such evidence for that limited purpose, provided that all the customary discovery attendant thereto has been provided to Merisant.
III. McNeil's Motion to Exclude Certain Third-Party Surveys
(Docket No. 167)
McNeil seeks to exclude the following three non-party surveys: (1) an " instant poll" administered over the internet by America OnLine (" AOL" ) that consisted of one check-the-box question that asked " Do you think that Splenda's claim ‘ made from sugar ... tastes like sugar’ is misleading?" ; (2) a survey conducted over the internet by the Center for Science in the Public Interest of " randomly selected" households, which asked respondents whether certain statements, including " is a natural product," described Splenda; and (3) a series of 14 focus groups commissioned by Kraft to test concepts for Kool-Aid and Country Time lemonade, which indicated that consumers were " more comfortable" when Kraft brought the words " sucralose" and " natural" together.
McNeil argues that each of these surveys is hearsay (at times, double hearsay) that is not admissible pursuant to any hearsay exception. McNeil further argues that all three of these third-party " surveys" are unreliable because the methods behind the surveys are not available, they are not trustworthy, they do not conform with accepted principles of survey research, and are generally unreliable due to certain fatal flaws. McNeil argues that the prejudicial effects of introducing the surveys far outweigh any potential probative value, and, accordingly, all three surveys should be inadmissible under Rule 403 of the Federal Rules of Evidence.
As a general matter, Merisant argues that each of these surveys is admissible for a legitimate, non-hearsay purpose, for example, as evidence of McNeil's " notice" of consumer confusion.
For the reasons discussed below, the Court finds that all three of these surveys are largely irrelevant, or, at best, marginally relevant to the issues presented in this case, and that any potential probative value of these surveys is substantially outweighed by the danger of unfair prejudice, confusion of the issues, and needless presentation of cumulative evidence. Therefore, McNeil's motion to exclude these three third-party surveys will be granted.
A. The AOL " Instant Survey"
In February 2005, AOL conducted an instant poll over the internet consisting entirely of the question, " Do you think that Splenda's claim ‘ made from sugar ... tastes like sugar’ is misleading?" In response to that question, 42 percent of those surveyed responded " yes," 40 percent responded " no," and 18 percent responded " not sure." See Def. Mot. Exclude Ex. A (Docket No. 167). Merisant argues that the AOL survey is admissible to show notice to McNeil of " consumer confusion," and that it is evidence indication a high level of " public interest" in the outcome of the litigation. Both of these arguments are unavailing. The Court finds that the probative value of this survey is far outweighed by the danger of unfair prejudice to McNeil that would result if this survey were to be admitted. The Court also finds that due to the vagaries of the question presented in the survey, and the complete lack of information about the manner and method in which this survey was conducted, introduction of this survey will unnecessarily confuse the issues in this case.
As an initial matter, viewed one way, the AOL survey asked the respondents to decide the question at issue in this case, by asking them whether McNeil's tagline for Splenda is " misleading." This is not an appropriate survey technique, and the single question posed is plainly biased and improperly suggestive. Both parties properly introduce numerous surveys in this case to endeavor to prove that consumers are confused, or are not confused, as the case may be, by McNeil's advertising for Splenda. Demonstrating that consumers are confused by certain advertising is one element in proving that advertising is misleading in violation of the Lanham Act. However, the AOL survey asked respondents to come to a legal conclusion, i.e., that Splenda's tagline is " misleading." See NFL Props. v. Prostyle, Inc., 16 F.Supp.2d 1012, 1019 (E.D.Wis.1998) (granting a motion in limine and excluding a survey regarding customer confusion because the survey improperly asked respondents for a legal conclusion). If Merisant were to introduce this survey, and explain to the jury that 42 percent of respondents found McNeil's tagline to be " misleading," Merisant essentially would be asking each juror to substitute the survey respondents' judgments for his or her own judgment, which is improper. Just as improper would be a survey question that asked respondents whether McNeil's advertising for Splenda violated the Lanham Act.
However, viewed another way, the AOL survey asks a general, vague and potentially irrelevant question, as it merely asks whether a certain statement is misleading without asking how the statement is allegedly misleading, or why a particular respondent finds such statement misleading. As the Court previously has noted, Merisant's claims in this case are very specific: it alleges that Splenda's " made from sugar, tastes like sugar" tagline is misleading because it conveys to consumers the impression that Splenda contains sugar or is a natural product. However, a question that asks whether the statement is generally " misleading," without inquiring as to why the respondent thinks the statement is " misleading," by no means concludes that any particular individual who responded " yes" necessarily thinks that Splenda's tagline misleading because it implies that Splenda contains sugar or is a natural product. In this regard, it is impossible for the Court to know upon what basis the a certain portion of the respondents to the AOL poll arrived at this conclusion.
The parties have submitted practically no information about the details surrounding this poll. For example, the Court does not know exactly when this poll was conducted, the context in which the poll was conducted, what " universe" of respondents was selected, whether the respondents were aware of this current litigation before they responded, whether the resulting data was properly gathered and accurately reported, in what way the tagline is purportedly " misleading," etc.
Both parties provide an identical page printed from an AOL news website, that displays the results of the survey, which is attached as a sidebar to a New York Times' article entitled, " Splenda's ‘ Sugar’ Claim Unites Odd Couple." See Def. Mot. Exclude Ex. A (Docket No. 167) (noting that the survey received 63,105 total votes). The article is dated February 14, 2005. However, McNeil also included a web print-out of the article dated February 15, 2005. See Def. Mot. Exclude Ex. C. There is no indication whether the AOL instant poll was conducted before, while, or after this article was posted online on AOL's website.
Presumably, the universe " selected" for this poll is comprised of people who happened to have been online while AOL released this poll, and cared enough about the issue to mouse-click on a box entitled " yes," " no," or " not sure." McNeil is on point by describing this sample as the " least representative imaginable." Def. Mot. Exclude 1 n. 2. The Court knows absolutely nothing about the respondents. For example, the Court does not know whether the respondents were over 18, whether they knew that Splenda is a no-calorie sweetener, whether they had ever been exposed to Splenda's packaging or advertising, whether they were aware of this litigation, whether they were employed by one of the parties in this case or another otherwise " interested" party, etc. See Shari Seidman Diamond, Reference Guide on Survey Research 229, 239 n. 41 in Federal Judicial Center Reference Manual on Scientific Evidence (2d ed. 2000) (" Identification of the proper universe is recognized uniformly as a key element in the development of a survey." ).
The Court does not know anything about the methodologies used by the people who conducted this survey. Of particular importance, the Court does not know whether survey respondents could " vote" more than once. Presumably, the same respondent could register as many votes-or clicks of the mouse-as he or she had time to do before the poll " closed." It is impossible to know based on the information available to the Court how many respondents produced the 63,000 plus votes that AOL collected.
It is impossible to know, based on the parties' papers, the exact context in which the AOL instant poll was administered. Both parties provide an identical page printed from an AOL news website that displays the results of the survey, which were attached as a sidebar to a New York Times' article entitled, " Splenda's ‘ Sugar’ Claim Unites Odd Couple." See Def. Mot. Exclude Ex. A (Docket No. 167) (noting that the survey received 63,105 total votes). However, neither party provided documentary evidence or an explanation to indicate in what context or format the poll was conducted. For instance, did people who responded to the AOL poll view the poll in connection with the Times article, was the poll a completely unrelated " pop up" advertisement that a respondent would view independent of the Times article, or was the poll connected to another related or unrelated article? This " unknown" is significant because the Times article references this lawsuit, stating that Merisant is " suing to stop Splenda from making the claim ‘ made from sugar, so it tastes like sugar.’ " Def. Mot. Exclude Ex. C. If people responded to this survey after reviewing the Times article, it is reasonable to infer that some portion of the respondents were improperly influenced by the article. This context would compound the bias inherent in the survey's single question.
In addition, directly above the results of the instant poll is a caption that reads " An unusual partnership of consumer advocates, food conglomerates and sugar growers are all targeting the fast-growing Splenda sweetener's claims." Def. Mot. Exclude Ex. A.
One of the threshold issues in determining whether a survey is reliable evidence in a litigation is whether either the interviewers or, " [a] fortiori, the respondents" are unaware of the litigation. Pittsburgh Press Club, 579 F.2d at 758.
In addition, Merisant's argument that the AOL survey is evidence that McNeil had " notice" of consumers' beliefs about McNeil's advertising for Splenda is unavailing. AOL conducted the instant poll over two months after this lawsuit had been filed. Being sued by a competitor who alleges that the defendant's advertising is misleading because it has caused " consumer confusion" is adequate " notice."
Merisant's other argument-that 63,000 responses by " consumers" is evidence that consumers have a significant interest in the truthfulness of McNeil's advertising for Splenda-also fails. Merisant argues that the " public interest" is relevant in responding to McNeil's laches defense. See Conopco, Inc. v. Campbell Soup Co., 95 F.3d 187, 193 (2d Cir.1996) (" Laches may be properly applied so long as its application is equitable in light of the public's interest in being free from confusion and deception." ). The Court acknowledges that the " public interest" is a relevant consideration in the laches content. However, the Court is not aware of any cases where the number of responses to an online survey has been used to by a court to infer the public's " interest" in a particular subject matter, in connection with the laches defense or otherwise. A raw (and, as explained above, see supra note 12, unreliable) number of respondents in no way explains how or to what extent the public interest is affected.
For the numerous reasons noted above, Merisant's motion will be denied with respect to the AOL instant poll.
B. The 2004 CSPI Survey
According to McNeil, the Center for Science in the Public Interest (" CSPI" ) " is an organization that purports to advocate for food nutrition and safety." Def. Mot. Exclude 2 n. 3. In April 2004, CSPI commissioned a survey that included questions about Splenda. The survey was conducted over the internet, and polled 1009 " randomly selected" households. Id. at 2. This survey was not created by or at the request of the parties in this case or any agent thereof. The only indication as to why CSPI created this survey is CSPI's statement in a subsequent press release that it commissioned the survey to " understand how consumers perceived Splenda's slogan." Def. Mot. Exclude Ex. F.
According to McNeil, CSPI commissioned the survey " to support [its] ongoing public relations campaign against artificial sweeteners." Def. Mot. Exclude 2.
The survey asked whether " based upon anything that you have seen or heard, or just your impression, which, if any, of the following describes Splenda?" Def. Mot. Exclude Ex. E at EMCN-MER 00821142. Among the available options a respondent was able to choose from were the statements " it is an artificial sweetener," " it is made from sugar," " it is a natural product," " its only ingredient is sugar," and " it is made from sugar and chlorine." Id. at EMCN-MER 00821144. Of the 1009 respondents surveyed, 60.9 percent chose " it is an artificial sweetener," 37.8 percent chose " it is made from sugar," 7.7 percent chose " its only ingredient is sugar," and 29.9 percent chose " it is a natural product." Id. In February 2005, CSPI publicly announced the results of the survey. It is not clear that McNeil was aware that the survey was conducted or aware of the results prior to February 2005. In its press release announcing the results of the survey, CSPI admonished McNeil for its " misleading labeling and advertising." Def. Mot. Exclude Ex. F.
No report describing the survey administrator's methodologies, the survey's stated purpose, or the results has been submitted for the Court's consideration of this motion. The only evidence presented for the Court's consideration as to the admissibility of this survey is a presentation of the results of the survey that describes the numerical and percentile breakdowns of the answers to each question, and the press release noted above. Merisant has not indicated that it intends to call any representatives for CSPI or the survey administrators who conducted its survey to testify in this case, nor has it indicated it that intends to have its survey experts testify as to any aspect of this survey.
Merisant argues that the CSPI survey would be offered for proper non-hearsay purposes, because it would be introduced to show respondents' beliefs about the composition of Splenda, and McNeil's " notice" of consumer confusion. Merisant further argues that the survey was conducted by a reputable research company that McNeil frequently uses, and that survey administrators employed a proven and reliable methodology. It also claims that the results of the CSPI survey are entirely consistent with the proportion of respondents who said that Splenda is a " natural product" in the survey conducted by Dr. Henry Ostberg, Merisant's survey expert. Pl. Mem. Opp'n 2 n. 2 (Docket No. 205).
The Court does not embrace Merisant's arguments with respect to the CSPI survey either. First, consumer " beliefs about the composition of Splenda" are irrelevant to this lawsuit unless the questions asked establish a nexus to Splenda's advertising and marketing. This survey was conducted over the internet, and-to the Court's knowledge, based on the papers received from the parties-respondents were not shown any packaging for Splenda or any of McNeil's marketing or advertising materials for Splenda. Of particular importance to this law suit, the CSPI survey did not reference the allegedly misleading statements in this case, such as McNeil's " made from sugar, tastes like sugar" tagline. Rather, it asked each respondent to rely upon " anything that [he/she] may have seen or heard," or his or her " impression" in forming an opinion about what statements describe Splenda. Reliance on respondents' " impressions" and not on the advertising itself is not appropriate. See Doris E. Long, Unfair Competition and the Lanham Act 326 (1993) (" Generally, in order to be given probative weight, the survey must be designed so that the responses clearly are based on the content of the advertisement and not simply on the respondents' prior personal experience." ). This nexus is too attenuated to be relevant to this action.
Further, many of the same technical flaws that the Court highlighted above with respect to the AOL survey are present in the CSPI survey as well. For example, no specific universe of respondents was chosen for this survey. Instead, the survey administrators surveyed " randomly selected households." Def. Mot. Exclude Ex. E at EMCN-MER 821141. No proper " filter" question was asked to determine whether respondents were past or prospective consumers of Splenda, and to weed out inappropriate respondents. In other words, respondents were not chosen because they were either past or prospective consumers of Splenda. In contrast, all of the " randomly selected" participants were questioned, even if they had never heard of Splenda.
The first question asked of each respondent was " how familiar are you with Splenda, a brand of sugar substitute?" Id. at EMCN-MER 821142. The results of the survey indicate that all 1009 respondents were asked this question. In response, 243 respondents chose the provided answer " I use it now," 184 chose " I have tried it, but do not use it," 451 chose " I have heard of it, but never tried it," and 132 chose " I have never heard of it." Id. at EMCN-MER 821143. In sum, 57.8 percent of the respondents had either heard of Splenda but never tried it or had never heard of it, only 24 percent could be described as current users of Splenda, and 18.2 percent as past users. Respondents were not asked whether they had ever viewed Splenda's packaging, or seen any of its advertising. It is not clear why the Court should admit evidence, or permit the jury to assign weight to any evidence, concerning the " impression" of an individual who has " never heard of" the product at issue in this case, let alone one who has possibly never seen the marketing or advertising claims at issue in this case.
For the reasons stated above, the Court finds the CSPI survey to be irrelevant and unreliable as evidence relating to Merisant's specific claims in this case. As a result, the Court finds that any marginal relevance or any probative value that this survey may potentially have, is outweighed by the unfair prejudice McNeil would suffer if this survey were to be introduced.
In addition, the Court does not find Merisant's argument that the CSPI survey constitutes " notice" of consumer confusion to be persuasive because, like the AOL survey, McNeil did not receive the results of the CSPI survey until February 14, 2005-over two months after Merisant filed the present law suit. As noted above, the timing of this survey precludes any relevance it may have for " notice" purposes.
Further, as Merisant itself acknowledges, the results of CSPI survey are " entirely consistent" with the results of Dr. Ostberg's survey. For this reason, the Court also finds that this survey should be excluded as " needless presentation of cumulative evidence" pursuant to Rule 403 of the Federal Rules of Evidence.
C. The Kraft Survey
In 1999, Kraft Foods, Inc. (" Kraft" ) conducted a series of 14 focus groups to test different concepts for its Kool-Aid and Country Time lemonade products. McNeil acknowledges that it received portions of the results of the study in the ordinary course of business in August 2000. What McNeil actually received from Kraft was a " final presentation," dated November 12, 1999, which amounts to a collection of slides (from a PowerPoint or similar presentation) that purports to describe the results of focus group sessions conducted by a third-party consulting company commissioned by Kraft. See Def. Mot. Exclude Ex. G.
According to the portions of the study that McNeil received, an unascertainable number of individuals in the focus groups viewed sucralose " as another kind of artificial sweetener," and said that the name " Splenda" sounded " more artificial." Def. Mot. Exclude 4. In addition, some participants were frustrated by the " inability to describe sucralose as ‘ natural’ " and were " more comfortable" when Kraft brought " the words ‘ sucralose’ and ‘ natural’ together." Id. The only indications that the Kraft focus groups discussed " made from sugar" positioning at all, are two slides out of the entire presentation. One slide, entitled " Sucralose Language Learnings," states that sucralose is " made from natural sugar," Def. Mot. Exclude Ex. G at EMCN-MER 281590, and another states: " Certain descriptions of Sucralose, e.g., ‘ made from natural sugar’, seem to give consumers permission to believe that it is a natural product and will taste like sugar," id. at EMCN-MER 281592. However, as far as the Court can tell-the presentation one of these two slides is dark and blurry-the word " Splenda" is not mentioned on either of the slides that mention the phrase " made from sugar."
McNeil argues that the Kraft survey should be excluded as double hearsay, because it is a summary of the results of focus group sessions, and because the focus group process in this case, or the recording of such process in the form of a " final presentation," is flawed. Merisant, in turn, argues that the survey is admissible for non-hearsay purposes such as " notice" of consumer confusion.
The Court notes that the results of a focus group are inherently different than the results of a survey. Whereas a survey is designed to collect data from an appropriate numerical sample of individual respondents, focus groups are designed to gather data from " groups" of people assembled together for that very purpose. As the Court of Appeals for the Fourth Circuit stated, " the very nature of a focus group seems, to some extent, to limit its ability to identify the message an advertisement conveys to an individual consumer." Scotts Co. v. United Indus. Corp., 315 F.3d 264, 277 (4th Cir.2002) (emphasis added). The court noted that: Because the participants in a focus group freely voice their opinions, the opinion of a participant can be shaped by those of the others. Thus, a participant who may have derived no false message from an advertisement viewed outside the context of the focus group might well change his opinion about the message conveyed by the advertisement after considering the views expressed by the other participants.
In Scotts, the court of appeals did not opine whether evidence of focus groups would ever be admissible, because it found that the focus group study at issue was so flawed as to render it inadmissible in that case. 315 F.3d at 277.
The Court finds that the Kraft focus group materials introduced here are unreliable and are not probative on the issue of whether McNeil's claims about Splenda at issue in this case are false or misleading.
Like the AOL survey and the CSPI surveys discussed above, the Court knows very little about the methodologies, purpose or results of the Kraft focus groups. Indeed, no actual data from the focus group sessions conducted by Kraft has been submitted to the Court. As such, the Court is left to ascertain Kraft's purpose in commissioning these focus groups, and every other detail of the focus group process, from the 10 slides submitted to the Court. In particular, the Court does not know whether or what questions were asked of the respondents (i.e., whether the questions were leading or not leading, open-ended or closed-ended, suggestive or non-biased, etc.), and what the respondents' actual answers to such questions may have been, either individually or in the aggregate. The Court does not know whether the administrators of the focus groups ignored some responses in favor of others in assembling their " final presentation" for Kraft.
The Court, and, indeed, the jury, would never be able to know any of this information because there is no indication that Merisant intends to call any of the administrators of these focus groups as witnesses at trial. Merisant has not cited any cases where a court has admitted the results of a focus group, over the objection of the opposing party, where the proponent of such evidence cannot produce any supporting evidence attesting to the reliability of the focus groups at issue.
In this case, the Court only knows what information the survey administrators chose to include in their " final presentation" to Kraft. The information that is available, however, does not inspire confidence in the reliability of this survey or the relevance with respect to the issues presented in this case.
For example, unlike the AOL or CSPI surveys where the individuals who participated in the surveys were either self-selecting or " randomly selected" by the survey administrators, the participants in the Kraft focus groups were chosen based on specific criteria: the level of their use of either Kool-Aid or Country Time products. Participants were designated as " moderate" or " heavy" Kool-Aid users, and " moderate" or " heavy" Country Time users. Def. Mot. Exclude Ex. G at EMCN-MER 281584-281586. In its motion papers Merisant did not seek to explain why or how this particular universe of individuals is in any way relevant to issue to be decided in this case. Because Splenda was not on the market in the United States when Kraft commissioned the study or when the results were disclosed to McNeil, it is reasonable to infer that none of the individuals who participated in the focus groups had ever viewed any of McNeil's marketing or advertising materials for Splenda, including those that are at issue in this case.
Credible evidence indicating that McNeil was aware that consumers were confused about Splenda's attributes, based on McNeil's advertising and marketing of Splenda, would be relevant and would generally be admissible. However, because the Kraft presentation does not address the issues to be decided in this case, it is irrelevant. For the reasons stated above, the Court finds that it is also unreliable, and will be excluded.
CONCLUSION
For the reasons stated above, Merisant's " Daubert " motion and additional motion in limine to exclude the survey and related testimony of Mr. Robert L. Klein will be denied, Merisant's " Daubert " motion to exclude the survey and related testimony of Dr. Susan S. McDonald will be denied, but Merisant's motion in limine to exclude the survey research conducted by Dr. McDonald after the deadline will be granted. In addition, McNeil's motion to exclude certain third-party consumer surveys will be granted. An appropriate order follows.
ORDER
AND NOW,
this 12th day of April, 2007, upon consideration of the various motions in limine and Daubert motions submitted by the parties, and the responses thereto, and any reply brief submitted, for the reasons stated in the accompanying Memorandum, IT IS ORDERED that:
1. Merisant's Motion to Exclude the Survey Conducted by Robert L. Klein and all related testimony and opinions (Docket No. 112) is DENIED.
2. Merisant's Motion in limine No. 9 to Exclude Testimony of Robert L. Klein (Docket No. 182) is DENIED.
3. Merisant's Motion to Exclude the Survey Conducted by Susan S. McDonald and all related testimony and opinions (Docket No. 110) is DENIED.
4. Merisant's Motion in limine No. 8 to Exclude All Survey Data Collected by Dr. Susan McDonald After the Close of Expert Discovery (Docket No. 181) is GRANTED and such evidence is INADMISSIBLE during McNeil's counsel's direct examination of Dr. McDonald; however, should Merisant on cross-examination seek to impugn the technical validity of Dr. McDonald's survey on grounds inexorably leading the witness to make reference to the 115 discarded surveys or to the " new" survey " results," thus opening the door for this evidence to be used to rebut those charges, McNeil will be permitted to offer such evidence for that limited purpose, provided that documentary discovery relating thereto has been provided to Merisant prior to the witness's testimony.
5. With respect to McNeil's Motion to Exclude Evidence Concerning Third-Party Surveys (Docket No. 167),
a. The Court finds that the AOL survey, the CSPI survey and the Kraft focus group presentation are all irrelevant to Merisant's claims and McNeil's defenses in this proceeding;
b. To the extent that any such evidence is marginally relevant, the Court finds that its introduction at trial would cause danger of unfair prejudice to McNeil, confusion of the issues, and needless presentation of cumulative evidence, as more fully explained in the accompanying Memorandum; therefore,
c. McNeil's Motion to Exclude Evidence Concerning Third-Party Surveys (Docket No. 167), is GRANTED and the AOL survey, the CSPI survey and the Kraft focus group presentation will be considered excludable at trial.
Notably, Merisant does not contest the form or the wording of the questions included in Dr. McDonald's survey, with one exception. Rather, Merisant argues that Dr. McDonald's survey is " so methodologically flawed and so poorly administered that it cannot possibly be relied upon." Pl. Mem. Supp. (McDonald) 3. Merisant's one content-based objection to the survey itself is that Dr. McDonald adopted an improper control stimulus-she switched the control stimulus from Equal to NutraSweet-which, Merisant notes, caused the results of Dr. McDonald's survey to be adjusted downward by 18 percent to account for " noise."