Themis Bar Review, LLC v. Kaplan, Inc.

Filing 92

ORDER denying 51 Motion for Summary Judgment; denying 69 Motion for Summary Judgment. Signed by Judge M. James Lorenz on 3/24/2016. (sjt)

1
2
3
4
5
6
7
8

UNITED STATES DISTRICT COURT

9

SOUTHERN DISTRICT OF CALIFORNIA

10
11
12

THEMIS BAR REVIEW, LLC, a
Delaware limited liability company,

15

v.
KAPLAN, INC., a Delaware corporation,
Defendant and Counterclaim Plaintiff.

16
17

ORDER DENYING CROSS
MOTIONS FOR SUMMARY
JUDGMENT [Docs. 51, 69]

Plaintiff and Counterclaim Defendant,

13
14

Case No.: 3:14-cv-00208-L-BLM

And related counter action

18
Pending before this Court are Themis’ Bar Review (“Themis”) and Kaplan, Inc.’s

19
20

(“Kaplan”) cross motions for summary judgment on the issue of whether Themis has

21

falsely represented its student bar passage rates. The Court decides the matter on the

22

papers submitted and without oral argument. See Civ. L. R. 7.1(d)(1). For the reasons

23

stated below, the Court DENIES both parties’ motions.

24

//

25

//

26

//

27

//

28

//
1
3:14-cv-00208-L-BLM

1

I.

2

BACKGROUND
Themis and Kaplan are both in the business of preparing law students to take the

3

bar examination. For a fee, each company provides instruction and materials tailored to a

4

specific jurisdiction’s bar exam. Themis is a relative newcomer to the business. Themis

5

began advertising its students’ bar passage rates in 2010 in an effort to gain more

6

business. In 2013, Themis circulated an advertisement (the “Themis Ad”) that listed its

7

passage rates for the 2012 bar examination. (See Themis Ad [Doc. 69-6].) The

8

advertisement presents three columns. The left column lists a jurisdiction, and the two

9

columns to the right indicate Themis’ students’ passage rates for that jurisdiction and the

10

overall state passage rate for that jurisdiction. At the top of both the Themis pass rate

11

column and the state pass rate column is a small asterisk. This asterisk corresponds to

12

another asterisk that appears at the bottom of the advertisement, next to fine print

13

language reading “Based on Themis first-time takers who completed 75% or more of

14

their course assignments and on state bar exam first-time takers.” This fine print is

15

notably smaller than the other portions of text printed on the Ad.

16

On January 22, 2014, Kaplan complained to Themis that its pass rate advertising

17

was misleading. (J.S.U.F. [Doc. 86] ¶ 23.) Themis responded indicating disagreement

18

on the issue. (Id. ¶ 24.) Subsequently, Themis filed a complaint with this Court seeking

19

a declaratory judgment that its advertising is accurate and non-misleading. (See

20

Complaint [Doc. 1].) Kaplan later filed a counterclaim alleging false advertising against

21

Themis. (See Counterclaim [Doc. 16].) Each party now seeks summary judgment in

22

their favor. (See Themis MSJ [Doc. 51]; Kaplan MSJ [Doc. 69].)

23
24

II.

LEGAL STANDARD

25

Summary judgment is appropriate under Rule 56(c) where the moving party

26

demonstrates the absence of a genuine issue of material fact and entitlement to judgment

27

as a matter of law. See Fed. R. Civ. P. 56(c); Celotex Corp. v. Catrett, 477 U.S. 317, 322

28

(1986). A fact is material when, under the governing substantive law, it could affect the
2
3:14-cv-00208-L-BLM

1

outcome of the case. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986). A

2

dispute about a material fact is genuine if “the evidence is such that a reasonable jury

3

could return a verdict for the nonmoving party.” Anderson, 477 U.S. at 248.

4

The party seeking summary judgment bears the initial burden of establishing the

5

absence of a genuine issue of material fact. Celotex, 477 U.S. at 323. The moving party

6

can satisfy this burden in two ways: (1) by presenting evidence that negates an essential

7

element of the nonmoving party’s case; or (2) by demonstrating that the nonmoving party

8

failed to make a showing sufficient to establish an element essential to that party’s case

9

on which that party will bear the burden of proof at trial. Id. at 322–23. “Disputes over

10

irrelevant or unnecessary facts will not preclude a grant of summary judgment.” T.W.

11

Elec. Serv., Inc. v. Pac. Elec. Contractors Ass’n, 809 F.2d 626, 630 (9th Cir. 1987).

12

“[T]he district court may limit its review to the documents submitted for the

13

purpose of summary judgment and those parts of the record specifically referenced

14

therein.” Carmen v. San Francisco Unified Sch. Dist., 237 F.3d 1026, 1030 (9th Cir.

15

2001). Therefore, the court is not obligated “to scour the record in search of a genuine

16

issue of triable fact.” Keenan v. Allan, 91 F.3d 1275, 1279 (9th Cir. 1996) (citing

17

Richards v. Combined Ins. Co. of Am., 55 F.3d 247, 251 (7th Cir. 1995). If the moving

18

party fails to discharge this initial burden, summary judgment must be denied and the

19

court need not consider the nonmoving party’s evidence. Adickes v. S.H. Kress & Co.,

20

398 U.S. 144, 159–60 (1970).

21

If the moving party meets this initial burden, the nonmoving party cannot defeat

22

summary judgment merely by demonstrating “that there is some metaphysical doubt as to

23

the material facts.” Matsushita Elect. Indus. Co., Ltd. v Zenith Radio Corp., 475 U.S.

24

574, 586 (1986). Rather, the nonmoving party must “go beyond the pleadings” and by

25

“the depositions, answers to interrogatories, and admissions on file,” designate “specific

26

facts showing that there is a genuine issue for trial.” Celotex, 477 U.S. at 324 (quoting

27

Fed. R. Civ P. 56(e)).

28
3
3:14-cv-00208-L-BLM

1

When making this determination, the court must view all inferences drawn from

2

the underlying facts in the light most favorable to the nonmoving party. See Matsushita,

3

475 U.S. at 587. “Credibility determinations, the weighing of evidence, and the drawing

4

of legitimate inferences from the facts are jury functions, not those of a judge, [when] he

5

[or she] is ruling on a motion for summary judgment.” Anderson, 477 U.S. at 255.
“[W]hen parties submit cross-motions for summary judgment, each motion must

6
7

be considered on its merits.” Fair Hous. Council of Riverside Cnty, Inc. v. Riverside

8

Two, 249 F.3d 1132, 1136 (9th Cir. 2001) (internal quotes and citations omitted). Thus,

9

“the court must rule on each party’s motion on an individual and separate basis,

10

determining, for each side, whether a judgment may be entered in accordance with the

11

Rule 56 standard.” Id. (quoting Wright, et al., Federal Practice and Procedure § 2720, at

12

335-36 (3d ed. 1998)). If, however, the cross-motions are before the court at the same

13

time, the court is obliged to consider the evidence proffered by both sets of motions

14

before ruling on either one. Id. at 1134."

15
16
17

III.

DISCUSSION
The Court finds two issues dispositive of these motions. They are (1) did Themis’

18

modification of its pass rate advertisements render Kaplan’s Motion moot? And (2) if

19

not, can the Court decide as a matter of law whether Themis’ pass rate advertisements are

20

or are not misleading? For the following reasons, the Court finds the answer to both

21

questions is “no.”

22

A.

23

Themis argues that the issues presented by the Counterclaim are moot because

Mootness

24

Kaplan is seeking only injunctive relief and Themis has permanently ceased circulating

25

the ad that is the subject of Kaplan’s Motion. (Themis MSJ 13:9–23; Themis Opp’n

26

[Doc. 81] 4:3–7:2.) The Supreme Court has articulated that

27
28

[i]t is well settled that a defendant's voluntary cessation of a challenged
practice does not deprive a federal court of its power to determine the
4
3:14-cv-00208-L-BLM

1

legality of the practice. If it did, the courts would be compelled to leave the
defendant free to return to his old ways. In accordance with this principle,
the standard we have announced for determining whether a case has been
mooted by the defendant's voluntary conduct is stringent: A case might
become moot if subsequent events made it absolutely clear that the allegedly
wrongful behavior could not reasonably be expected to recur.

2
3
4
5
6

Friends of the Earth, Inc. v. Laidlaw Environmental Services (TOC), Inc., 528 U.S. 167,

7

189 (2000) (internal citations and quotation marks omitted). Kaplan argues that its case

8

is not moot because Themis has failed to provide adequate assurances it will not run

9

advertisements similar to the tested ad in the future. (See Kaplan Reply [Doc. 85] 9:1–

10

28.) The Court agrees.

11
12

The only assurance provided by Themis comes in the form of declaration
testimony from Richard Duffy, the CEO of Themis. Duffy states

13

[a]t no time since September of 2014 has Themis used a version of the ad
with the smaller explanatory language that Kaplan counterclaimed over, and
that Kaplan’s marketing expert tested in his ‘memory test’ survey. Although
I did not and do not agree with Kaplan’s assertion that this version of the ad
was misleading, I voluntarily changed the format of the ad to make the
explanation larger and more prominent, in the hopes of eliminating any
argument over the issue, and because I favor full disclosure to students
choosing between bar exam providers. I do not plan on ever again using that
version of the ad with the smaller explanatory language. Even if Kaplan’s
request for injunction is denied, Themis will continue to use the larger, more
prominent explanation.

14
15
16
17
18
19
20
21

(Duffy Decl. [Doc. 87-2] ¶ 3.)
The problem with Themis’ argument is that Richard Duffy’s statement in no way

22
23

obstructs Themis from recirculating the advertisement at a future time. Rather, it merely

24

announces a present intention (that could later change) not to do so. Because Themis has

25

thus failed to make “it absolutely clear that the allegedly wrongful behavior could not

26

reasonably be expected to recur,” the Court finds this case is not moot.

27

//

28

//
5
3:14-cv-00208-L-BLM

1

B.

2

Under the Lanham Act and California law, a false advertising claim is cognizable

False Advertising

3

only if the advertisement at issue is false or, if true, has a tendency to mislead. Southland

4

Sod Farms v. Stover Seed Co., 108 F.3d 1134, 1139 (9th Cir. 1997) (Lanham Act); Leoni

5

v. State Bar, 39 Cal. 3d 609, 626 (1985) (California law). Themis argues it is entitled to

6

summary judgment because its pass rate advertising is literally true and not misleading.

7

Kaplan, by contrast, argues that it is entitled to summary judgment because Themis’ pass

8

rate advertising, even if literally true, has a tendency to mislead and the other elements of

9

its false advertising actions are uncontested.

10
11

1.

Literal Falsity

Kaplan argues that summary judgment in favor of Themis is inappropriate because

12

(1) the pass rate is literally false when read in isolation from the footnote and (2) even if

13

read in connection with the footnote, Themis’ data collection methods are problematic

14

and thus create serious doubt as to whether the pass percentages are accurate as to first

15

time test takers who complete 75% or more of the course. (Kaplan Opp’n [Doc. 82]

16

11:6–15.)

17

It is true that the published pass rates would be literally false if read in isolation

18

from the footnote. However, to determine whether an advertisement is literally true, the

19

proper focus is on the advertisement as a whole rather than an isolated section. Stover

20

Seed Co., 108 F.3d at 1139 (“When evaluating whether an advertising claim is literally

21

false, the claim must always be analyzed in its full context.”) Thus, the pass rates are

22

literally false only if they are inaccurate with respect to the population defined in the

23

footnote: first time takers who complete 75% or more of the course.

24

Themis contends that the percentages it published are accurate with respect to this

25

population of students. In support, Themis cites to an independent audit performed by,

26

James Schultz (“Schultz”), a Certified Public Accountant, that concludes the data

27

collection process utilized by Themis to calculate pass rates is methodologically sound.

28
6
3:14-cv-00208-L-BLM

1

(See Schulz Decl. [Doc. 51-5].)1 Kaplan responds by submitting a report from expert Hal

2

Poret. In his report, Poret concludes that Themis’ practice of individually contacting

3

students in states that do not publish pass lists is problematic. Specifically, Poret argues

4

that the study of human psychology suggests relying exclusively on the self-reporting of

5

students may result in a systematic response bias such that students who failed the exam

6

might be ashamed of the fact and therefore lie when asked whether they passed. (Poret

7

Supplemental Report [Doc. 69-19].)
Themis disputes Poret’s conclusion by arguing that law students would not

8
9

misreport their passage because (1) they are honest people and (2) it is in their self-

10

interest to report failure because they would forego the benefit of a free repeat course by

11

lying. (Themis Reply [Doc. 87] 4:17–28.) A reasonable jury could find the self-interest

12

argument persuasive. However, at this stage, the Court must construe all inferences in

13

favor of Kaplan as the non-moving party on this issue. In doing so, the Court finds

14

Poret’s Report sufficient to create a material dispute of fact as to the soundness of

15

Themis’ data collection processes and the pass rates thus derived.

16
17

2.

Tendency to Mislead

The parties take opposing positions on the issue of whether Themis’

18
19

advertisements have a tendency to mislead. If an advertisement is misleading, it can be

20

actionable even if literally true. Stover Seed Co., 108 F.3d at 1140. To determine

21

whether an advertisement is misleading, the Court considers it from the perspective of the

22

advertisement’s actual audience. Id. at 1139. Here, the parties agree that law students

23

are the actual audience for Themis’ pass rate advertising. (Themis MSJ 22:7–24; Kaplan

24
25
26
27
28

1

Kaplan argues that this audit is inadmissible because it was submitted after the discovery cutoff date
for furnishing expert reports. (Kaplan Opp’n 23:17–25:22.) The Court need not reach the issue of the
Audit’s admissibility here because, even if it is admissible, the Court finds that Kaplan has submitted
sufficient evidence in opposition to create a dispute of material fact regarding the validity of Themis’
data collection methods.
7
3:14-cv-00208-L-BLM

1

Opp’n 14:5–20.) Thus, the question presented is whether Themis’ advertising has a

2

tendency to mislead law students.

3

Each party has submitted consumer survey evidence to support its position.

4

Kaplan submitted a study conducted by Hal Poret. (Poret Report [Doc. 69-18].) Poret’s

5

study involved giving an advertisement to 331 current law students, allowing them to

6

look at it for as long as they wanted, and then taking it away from them before asking

7

various questions. Half of the students (the test group) received the actual Themis Ad.

8

The other half (the control group) received a modified version that displayed the footnote

9

text more prominently. After viewing the advertisement, both groups were asked

10

whether the pass rates on the advertisement represented all Themis test takers or only a

11

certain subgroup of test takers. In the test group, 15.7% of the students answered

12

correctly (that the pass rate represented only a subgroup of Themis test takers) and 69.3

13

% answered incorrectly. In the control group, by contrast, 64.8% of students answered

14

correctly and only 17.6% answered incorrectly.

15

Themis submitted a study conducted by David Stewart (Stewart Report [Doc. 51-

16

9].) Stewart’s study essentially followed the same design as Poret’s study save three

17

major modifications. Unlike Poret’s study, Stuart’s study did not utilize a control group

18

and it allowed respondents to keep the advertisement and reference it while answering the

19

questions. Furthermore, the Stewart study did not test the same ad that is the subject of

20

both Kaplan’s Counterclaim and Poret’s study. Rather, the Stewart study tested two

21

different ads that each contained a significantly more prominent footnote than the Themis

22

Ad Kaplan sued upon.2 About 84.5% of the 801 students tested correctly answered that

23

the pass rates referred to a certain subgroup of Themis students. About 12% of the

24

students answered incorrectly that the pass rates referred to the entire population of

25

Themis students.

26
27
28

2

See Poret Report [Doc. 69-20] 9–11 for a comparison of the three different ads.
8
3:14-cv-00208-L-BLM

Each party criticizes the other’s research design. Themis argues that in high-level

1
2

involvement purchasing decisions such as choosing a bar prep company, a reading test,

3

where the subject can reference the ad while answering questions, is more appropriate

4

than a memory test, where the subject cannot reference the ad while answering the

5

questions. The gist of Themis’ argument is that humans have bad short term memory.

6

Thus, during the short time between looking at the Ad and answering the questions,

7

students are likely to have forgotten the information conveyed in the footnote. However,

8

in real life, a law student choosing a bar prep company would undertake a systematic and

9

comprehensive decision making process that would involve considering the Ad more

10

than just once. Accordingly, Themis argues, a reading test is preferable to a memory test.

11

In support of this argument, Themis offers testimony from its retained expert Stewart

12

(Stewart Decl. [Doc. 51-6]) and cites to a study3 concluding that a “reading test” is the

13

most appropriate survey method for testing consumer perception in high involvement

14

transactions.

15

Kaplan counters that the Poret study did not in fact ask students to “memorize”

16

any data. Rather, if a student noticed the fine print in the footnote, short term memory

17

issues inherent to the human species would not preclude the student from accurately

18

answering, a very short time later, whether the pass rates represented only a subset or all

19

of Themis test takers. (Kaplan Opp’n 17:20–18:16.) Kaplan further argues that the

20

“memory test” design more accurately replicates a real life consumer situation because

21

real life consumers look at an advertisement as long as they need to and form their

22

impressions during this time period. Thus, if a student did not notice the footnote after

23

looking at the advertisement but before being asked about it, that suggests that the

24

advertisement would be misleading in a real life scenario. Bating the student to pay

25
26
27
28

Jerre B. Swann, A ‘Reading’ Test or a ‘Memory’ Test: Which Survey Methodology is Correct. 95
Trademark Rep. 876 (2005)
3

9
3:14-cv-00208-L-BLM

1

attention to the footnote by asking specific questions while reviewing the document,

2

Kaplan argues, simply causes a student to pay more attention to the footnote than they

3

otherwise would have and therefore is an inferior methodology for testing consumer

4

perception.

5

Kaplan cites to a number of cases as well as an academic article that, in the

6

trademark context, suggest the “memory test” methodology employed by Puret is the

7

ideal research design for testing consumer perception of advertising. See Henry D.

8

Ostberg, Response to the Article Entitled, “A ‘Reading’ Test or a ‘Memory’ Test: Which

9

Survey Methodology is Correct?” 95 Trademark Rep. 1446 (2005) (memory test is

10

superior because the reading test “makes it impossible to determine what consumers

11

thought or believed on their own before the questions themselves caused them to seek

12

additional information or clarification”); E. & J. Gallo Winery v. Pasatiempos Gallo,

13

S.A., 905 F. Supp. 1403, 1412 (E.D. Cal. 1994) (approving the use of a memory test

14

design in a trademark dispute).4 After reviewing this literature on consumer surveys and

15

considering the Poret report, the Court finds it highly probative on the issue of the

16

Themis Ad’s tendency to mislead.

17

Themis also argues that, regardless of its validity, the Court cannot consider the

18

Poret Report because the Themis Ad is factually true and facially unambiguous. (See

19

Themis Opp’n 7:19–9:17; Themis MSJ 21:14–23:14.) However, Themis does not cite to

20

any binding authority to support this argument. Further, even assuming it is improper to

21

consider survey evidence in determining whether a factually true and facially

22

unambiguous advertisement is misleading, the fact remains Themis has failed to establish

23

that the pass rates are literally true. Accordingly, the Court finds it can consider the Poret

24

Report. Having considered the Report, the Court finds it sufficient to create a material

25
26
4

27
28

Themis argues that the authority cited by Kaplan is not on point because it involved trademark
infringement claims rather than false or misleading advertising claims. (Themis Opp’n 18:4–20:90.)
However, Themis does not present, and the Court is unaware of, any persuasive reason for finding this
distinction meaningful.
10
3:14-cv-00208-L-BLM

1

dispute of fact regarding whether or not the Themis Ad has a tendency to mislead law

2

students. Accordingly, the Court DENIES Themis’ Motion.

3

The Stewart Study, by contrast, does not suffice to defeat Kaplan’s Motion. The

4

Stewart study is not probative of whether the Themis Ad sued upon is misleading because

5

it did not test the Themis Ad at issue here. The footnotes in the advertisements presented

6

to the subjects of the Stewart Study were significantly larger (both overall and relative to

7

the surrounding text) than the footnote contained in the actual Themis Ad that Poret

8

tested and Kaplan sued upon. Given this disparity in footnote prominence, it is possible

9

that a substantial number of the students who correctly answered the question would not

10

have answered correctly if presented with the much smaller footnote of the actual Themis

11

Ad at issue here.

12

That said, Themis advances several additional arguments in opposition to Kaplan’s

13

Motion. Themis argues that law students constitute a population that is trained to read

14

the “fine print.” Further, given the expense of bar prep and the importance of passing the

15

bar, law students have all the incentive in the world to closely scrutinize the

16

advertisement and read the footnote that is signaled by two asterisks at the top of the

17

Themis Ad. Given this training and incentive, Themis argues its Ad is not misleading

18

because a reasonable law student will actually read the footnote. Furthermore, Ellen

19

Douglas, the Director of Themis, has declared under oath that “[n]o Themis student has

20

ever complained to Themis that Themis’ advertisements are misleading or deceptive.”

21

(Douglas Decl. [Doc. 81-2] ¶ 9.)

22

Viewing all inferences in favor of Themis, the Court finds these arguments

23

sufficient to defeat Kaplan’s Motion. “Credibility determinations [and] the weighing of

24

evidence . . . are jury functions, not those of a judge, [when] he [or she] is ruling on a

25

motion for summary judgment.” Anderson, 477 U.S. at 255. Thus, notwithstanding

26

Kaplan’s survey evidence to the contrary, the Court cannot hold, as a matter of law, that a

27

law student is unlikely to read a footnote signaled by an asterisk when evaluating which

28

expensive bar review course to choose to prepare for one of the most important tests of
11
3:14-cv-00208-L-BLM

1

their lives, especially when there is evidence that no student has ever complained of

2

being misled.
However, Kaplan offers an additional, independent reason as to why Themis’

3
4

advertising is misleading. Kaplan argues that, even assuming students read the footnote,

5

Themis’ advertising is nevertheless misleading with respect to some jurisdictions because

6

the published percentages are not statistically significant. For example, Themis

7

advertised that 100% of its students who were first time takers and completed 75% or

8

more of the course passed the July 2013 Washington D.C. bar exam. (See Washington

9

D.C. Ad [Doc. 69-8].) Themis compared its 100% pass rate to the D.C. district wide

10

average of 71%. Kaplan asserts this is misleading because the 100% pass rate is based a

11

population of four students. The theory is that “extremely small sample sizes cannot

12

realistically indicate whether or not Themis confers upon its students an advantage as

13

compared to others who do not take Themis, especially when those sample sizes are not

14

disclosed.” (Kaplan Opp’n 22:20–23.)

15

The Court agrees that, by comparing Themis pass rates to state or district wide

16

averages, Themis’ advertising might imply that the Themis pass rates are based upon a

17

statistically significant sample size. Thus, when seeing that Themis’ boasts a 100% pass

18

rate in Washington D.C., a student may infer that the 100% figure is based on more than

19

4 students. However, a student might also conclude that the advertisement means

20

nothing more than what it literally says: every Themis student who was a first time test

21

taker and completed 75% or more of the course passed the July 2013 D.C. Bar exam. A

22

reasonable jury could conclude either way. Accordingly, the Court DENIES Kaplan’s

23

Motion. (Kaplan Opp’n 21:23–23:16.)

24

//

25

//

26

//

27

//

28

//
12
3:14-cv-00208-L-BLM

1
2
3

IV.

CONCLUSION AND ORDER
For the foregoing reasons, the Court DENIES both parties’ motions for summary

judgment [Docs. 51, 69].

4
5

Dated: March 24, 2016

6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
13
3:14-cv-00208-L-BLM