Learning Network v. Discovery Communications, 153 F. Supp. 2d 785 (D. Md. 2001)

*786 Mark Andrew Goodin, Halley F. Sexter, Morgan, Lewis and Bockius LLP, J. Kevin Fee, Morgan, Paul D. Weller, Joseph B.G. Fay, Lewis and Bockius LLP, Philadelphia, PA, David Leichtman, Joshua Paul, Hillel I. Parness, Gregory S. Shatan, Stephen W. Feingold, Morgan, Lewis and Bockius LLP, New York City, Gary Mayer Rinck, Pearson PLC, London, UK, Rachel Beth Damelin, Morgan, Lewis & Bockius LLP, Washington, DC, for the Learning Network, Inc., Headland Digital Media, Inc., Pearson, Inc., Pearson, PLC, Phillip Hoffman, plaintiffs.

Charles P. Goodell, Jr., Goodell, DeVries, Leach and Gray, Linda S. Woolf, Goodell, DeVries, Leech and Dann LLP, Baltimore, MD, Christopher M. Curran, Frank Panopoulos, Francis A. Vasquez, Jr., White & Case, LLP, Washington, DC, John Paul Reiner, Robert L. Raskopf, Andrew W. Hammond, Christopher J. Glancy, White and Case LLP, New York City, Coleen M. Meehan, Morgan, Lewis and Bockius LLP, Philadelphia, PA, for Discovery Communications, Inc., defendants.

GARBIS, District Judge.

The Court has before it Channel's Motion to Exclude the Expert Report and Testimony of Henry D. Ostberg, and the materials submitted by the parties relating thereto. The Court has held a hearing on this motion and had the benefit of the arguments of Counsel.

In March of 2001, The Learning Network (hereinafter "Network") commissioned Dr. Henry D. Ostberg (hereinafter "Ostberg"), a market research expert, to design and conduct a survey to determine whether or not there is a likelihood of confusion between Network's "Learning Network" and Discovery Communications, Inc.'s (hereinafter "Channel") "The Learning Channel" or "TLC."

The survey used what is called a "mall intercept" design. Potential respondents were approached by interviewers in shopping malls in ten different cities in the United States. These respondents were asked a series of screening questions, including, but not limited to:

*787 • whether they owned or had access to a computer which has Internet access; and

• whether, they were likely, during the next six months, to use the Internet to:

a) obtain educational information or materials for themselves or a child; or

b) obtain information on how to advance their career or business knowledge.

Respondents who answered "no" to these, and other, screening questions^[1] were terminated.

Respondents who satisfied the screening criteria were taken to a separate room within the shopping mall for the second portion of the interview. The interviewer began by reading the following script:

I would like to show you several pages from a website. These are actual pages from that website, printed and put into a portfolio. Please go through the portfolio I show you, reading whatever pages are of interest to you. Take as much time as you need to become familiar with that website. Please let me know when you have finished looking through the portfolio.

Network's Memorandum of Law In Opposition to Channel's Motion to Exclude The Expert Report and Testimony of Henry D. Ostberg (hereinafter "Opposition"), Ex. 1, App. C at 1 (In-Office Questionnaire).

Each respondent was then shown one of two portfolios containing color pictures of what was said to be a website page. Two hundred thirty-two respondents were shown the "Test" Portfolio containing pages from Network's "Learning Network" site. Two hundred twenty eight respondents were shown the "Control" Portfolio^[2], which was identical to the "Test" Portfolio, but within which the name "Learning Network" had been replaced with "Learning.com."

The cover of the portfolio identified a portfolio number. The first page of the portfolio contained in large type in the center of the page "www.learningnetwork.com^[3]".

The next four portfolio pages (in full color) contained what were represented to be "actual pages from [a] website, printed and put into a portfolio".^[4] As discussed herein, however, the website page, which would be seen by a consumer, would differ from the printed pages in the portfolio in several respects. For example:

• The portfolio pages did not include the title bar at the top of the internet website page, and

• The portfolio pages did not include an "address" designation of "www.learningnetwork.com" which would be displayed to a typical (if not every) consumer gaining Internet access.

When respondents indicated that they had finished looking at the portfolio, it was removed. Respondents were asked the following series of questions:

*788 1a. Is this website known by any name or designation other than the one you saw on the website?

1b. (If "Yes") By what other name or designation is this website known?

2a. As far as you know, is the company, organization or people who ownor put outthis website engaged in any activities, businesses or services other than this website?

2b. (If "Yes") In what other activities, businesses or services are they engaged?

(If Cable or TV Mentioned) What specifically do you have in mind? Please tell me specifically.

3a. Do you think that this website received or needed to receivepermission or approval from anyone for any reason?

3b. (If "Yes") Who gaveor needed to givepermission or approval to this website?

Why do you think permission or approval was required from (NAME MENTIONED)? Any other reason?

4a. Have you even seen this particular website before today?

4b. (If "Yes") What was the name or designation of that website?

5a. Does your household currently subscribe to cable television or satellite television which requires a monthly fee?

5b. (If "Yes") Which do you subscribe to?

The survey results, according to Ostberg, established that:

1. None of the respondents seemed to confuse the Learning Network with The Learning Channel (TLC), as reflected by the fact that the Learning Network was said not to be known by any other name and not to be engaged in either television or cable broadcasting.

2. A substantial number of respondents thought that the Learning Network had received, or needed to receive, "permission or approval" from someone else to operate; however, none thought that such permission or approval was required from The Learning Channel (TLC).

3. Six percent (6%) of respondents said that they had seen the Learning Network website before; but none said that the website which they had seen previously was called "The Learning Channel" (or "TLC").

4. No likelihood of confusion can be attributed to the word "Network" in the "Learning Network" name, since beliefs about the website were substantially the same whether the website was called "Learning Network" or simply "Learning.com."

Ostberg Report at 3.

Channel seeks to have the Court exclude the Ostberg survey from evidence at the jury trial of this case. The Court will herein discuss some principal defects in the survey. The omission of certain of Channel's arguments from the instant discussion does not indicate rejection of those points, but rather, is reflective of the limited time available for writing this Memorandum and Order.^[5]

It is well established that a district court must act as the "gatekeeper" for *789 expert witness testimony in a jury trial. See Kumho Tire Co. Ltd. v. Carmichael, 526 U.S. 137, 152, 119 S. Ct. 1167, 143 L. Ed. 2d 238 (1999); Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 589, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993); Oglesby v. General Motors Corp., 190 F.3d 244, 249 (4th Cir. 1999). The instant motion relates to survey evidence which is intrinsically different from the types of evidence most frequently encountered in a Daubert or Kumho context.

As stated by Professor McCarthy:

One must keep in mind that there is no such thing as a "perfect" survey. The nature of the beast is that it is a sample, albeit a scientifically constructed one .... Like any scientific method related to statistics in the social sciences, every survey, no matter how carefully constructed and conducted, has some potential flaws somewhere. The proper approach is to view such evidence with some understanding of the difficulty of devising and running a survey and to use any technical defects only to lessen evidentiary weight, not to reject the results out-of-hand.

5 McCarthy on Trademarks § 32:178, at 32-294 (citations omitted).

Indeed, recent events have demonstrated that the most "accepted" of surveys can be considered of doubtful validity if the manner in which questions are presented and answers are given is subjected to detailed scrutiny. See Bush v. Gore, 531 U.S. 98, 121 S. Ct. 525, 148 L. Ed. 2d 388 (2000).

Nevertheless, it is, and has been even before Daubert, necessary for a district court to pass upon the threshold question of the validity of a survey before submitting the results to a jury for their consideration. As noted in Simon Property Group, L.P. v. mySimon, Inc.:

Expert testimony concerning survey results may not be reliable if the survey format does not accurately gauge consumer confusion between the marks at issue.

104 F. Supp. 2d 1033, 1040 (S.D.Ind.2000).

The question here presented is whether the asserted defects in the Ostberg survey are so great as to warrant exclusion rather than grist for the mill of cross-examination, rebuttal evidence, and jury evaluation.

Channel raises the question of whether the survey format is defective because the stimulus is presented through a medium different from that used by the consumer in the marketplace. Specifically, Channel contends than a printed picture of a webpage has a different effect from the webpage as it appears on a computer monitor screen.

The issue appears to be a significant one. The matter may be of particular concern in the instant case because the consumer likely would have received exposure to Channel's product through a television screen.

It would not have been overly burdensome for the surveyor to have provided the "portfolio" on a computer monitor. For example, the pictures in the portfolio could easily have been placed on a CD-Rom disc and viewed on a computer monitor.

The Court concludes, however, that in the present context, with neither judicial precedent nor scientific writings shedding light on the matter, it should not hold that the display of the stimulus on a paper page rather than on a computer monitor, is a determinative, or even significant, factor. Accordingly, the Court is not relying upon this factor.

The Court finds that the survey is defectively designed in that it does not engender reasonable confidence that all, a majority, or even a substantial number, of the subjects^[6] were answering the questions posed with the understanding that the stimulus was the website of, or presented by, "Learning Network." If subjects did not have this understanding, their responses to the "substantive" questions would be meaningless.

In the real world or, better put, the "virtual" world of the internet, a consumer typically^[7] gets to the Learning Network website by a route that includes the affirmative use of the Learning Network name or, at least, an indication that he is going to a website published by a "learning" entity. The consumer can get to the website by typing in the Uniform Resource Locator (hereinafter "URL") "www.learningnetwork.com", by using a search engine, or through a link to the site which would typically, if not necessarily, have a reference to the Learning Network or, at least, "learning." The only other way to get to the website would be if the consumer had "bookmarked" the website which means that the consumer had gotten there previously by way of the URL, a search engine, or a link. The Court does not find the printing of the purported URL on the first page of the portfolio (prior to the website page pictures) remotely equivalent to the route by which the consumer gets to the site.^[8] Further, once the subject reached the portfolio pages purportedly depicting the website; he was denied significant location information that would be available to a real world consumer. Subjects were provided with a picture of the website which omitted the "title bar" at the very top of each website page. The "title bar" for the Learning Network home page includes the statement "Welcome to the Learning Networkyour educational resource.^[9]" Other pages within the website also carry equivalent identification notice in the title bar.

Moreover, the portfolio website pictures omitted the "tool bars" that would likely appear on a real world consumer's monitor stimulus, including the location tool bar which would tell the consumer that he was connected to http://www.learningnetwork.com.

In sum, the subject looking at the portfolio website arrived at the site without the real world affirmative action that would reinforce the identity of the location and viewed a purported picture of a computer monitor screen without the "title bar" and location information that reinforces the "Learning Network" or at least "learning" route used to get there.

The subject then looked at a picture of a website that displayed the Learning Network logo in the upper left corner, a Gateway Computer advertisement at the top in the center as well as advertisements and links for Gateway, AOL and other vendors (including the Learning Network Store) down the right side. It is probably true that if asked to name the owner of the *791 website, with time to study the picture, one could find sufficient clues to figure out the right answer. However, the subjects looked at the pictures without knowing what would be asked. Then, without being able to look back, they were supposed to have known that it was the Learning Network website and, based on this knowledge, answer substantive questions.

The Court finds that the survey was severely defective by virtue of its failure to ask any question that would test the critical question of what website it is that the subject thought they had seen.

A significant indication that the survey generated confusion is provided by subjects' responses to Questions 4a and 4b. Of the 232 subjects in the test condition, 14 answered that they had seen "this particular website before today." Of these 14, only two identified the website^[10] as the Learning Network. These 14 subjects, having recognized (or thought they recognized) the website, would have been the most likely to know what the website was. Only two of these got the name right, andit must be notedthe two who got it right would have seen the actual website on the internet thus having had the identification stimuli of obtaining access, the "title bar," and the location tool bar.

The Court concludes that the subjects have not been shown to have known that they had seen the Learning Network website as distinct from real world consumers who would have had this knowledge. Accordingly, the survey evidence does not establish that the subjects were answering questions with critical knowledge equivalent to that possessed by the relevant universe of consumers of the Learning Network website.

The purpose of the survey was to ask subjects a question (or questions) to determine whether there was confusion between the source/ownership of the website they had seen and The Learning Channel. In the survey, the "confusion" question was phrased in the following manner: "Do you think that this website receivedor needed to receivepermission or approval from anyone for any reason?" The results were as follows:

   Yes           57
   No           113
   Don't Know    62

The fifty-seven subjects who gave affirmative answers made a wide variety of statements describing the permission needed or given. None stated that The Learning Channel would have to give permission. The Court finds that the responses to this question are without evidentiary value.

As noted above, there has not been a showing that the subjects knew that the potential permission seeker was the Learning Network.^[11] Moreover, the appearance of the website, once stripped of the "title bar" and URL, could lead to the conclusion that it was a "bazaar" type website presenting goods and services from various providers. Several subjects referred to advertisers (Gateway, "advertisers", AOL) as those from whom the publishers of the website would need permission.

*792 The wide open nature of the question as to this website's^[12] seeking approval from anyone for any reason, requires one to speculate as to what the subjects may have been considering. A more focused question could, for example, have sought to know whether the website consumer needed to get permission to use the name of the website. Of course, absent confidence that the subject knew the name of the website at issue, even the proposed question would not yield useful information.

Finally, although perhaps not of the utmost significance in the current case, the question could lead subjects who were the most confused as to source to give a negative answer. In other words, a subject who thought, by virtue of the similarity of the names, that the Learning Network website was actually put out by the owners of The Learning Channel would not think that the website owner (being TLC) would need to seek permission from TLC.

In Question 2a, the subject is asked whether those "who ownor put outthis website engaged in any activities, business or service other than this website." The responses were as follows:

  Yes                25
  No                 84
  Didn't Know       123

The 25 subjects giving affirmative responses mentioned a variety of other "activities" in which the website publisher was engaged,^[13] but none mentioned television programming.

Network seizes upon this response as an indication that no subject thought that The Learning Channel owned or published the website. However, the Court finds the question to have been phrased so vaguely as to yield no meaningful information. As noted above, it is not possible to determine what website the subjects thought the stimulus was. The 123 "don't know" answers would include those who did not think they knew the website identification as well as those who thought they knew this information, but did not know what the website producers did apart from production of the website. The Court does not find it sound to conclude that the absence of references to television programming establishes that the subjects knew they had seen the Learning Network rather than a Learning Channel website.

The Court finds it unsound to seek to discern, by inference, from responses to a vague question, information that the survey could have, and should have, obtained directly and unambiguously.

For the foregoing reasons:

1. Channel's Motion to Exclude the Expert Report and Testimony of Henry D. Ostberg is GRANTED.

2. The Expert Report and Testimony of Henry D. Ostberg shall not be admitted in evidence.

[1] For clarification, a respondent was terminated if he gave a "no" answer to both subsections of the second question listed above. However, if the respondent gave a "no" answer to one subsection and a "yes" answer to another subsection, he was not terminated.

[2] The "control" portfolio was used to determine whether the use of the word "learning" contributed to any confusion.

[3] "www.learning.com" was used for the "control" portfolio.

[4] The Exhibit hereto is not reproduced in color and, therefore, does not adequately demonstrate the extent to which the portfolio page replicated, in general appearance, the image on a computer monitor screen.

[5] The Court has had full argument on the motions and ample time to review the materials and precedents and to consider the motions. The matter has been fully, and carefully, considered. In the event the instant motions were to be the subject of an appeal, the Court may provide a supplemental memorandum.

[6] The term "subjects" is used to refer to those taking the survey.

[7] If not inevitably.

[8] This criticism might have been somewhat overcome had the subjects been placed before a computer and told to type in the URL to get to the website at issue.

[9] The "title bar" also includes information that may vary from consumer to consumer for example, the browser. This additional information does not detract from the message conveyed by "Welcome to the Learning Network."

[10] The subjects who had seen the particular website before were asked to provide the "name or designation" of the website they had seen before. Inasmuch as the website seen before was the same website they had seen "today," the subjects were being asked to identify the website they had just seen in the portfolio.

[11] Indeed, two subjects thought that the website owner would need permission or approval from Learning Network.

[12] There may be a problem with the question's asking if "this website" needed permission. At least as a literal matter, it would be the people who published the website, not the website that would need permission.

[13] The highest number of like responses was 8. Those respondents indicated that the owners of the website were also engaged in the production of educational software.