The Consumer, The Character and The Consumed: The Three Cs Model of Product Placement Effects

Cristel Antonia Russell and Barbara B. Stern

1 The Consumer, The Character and The Consumed: The Three Cs Model of Product Placement Effects

... product placements act as implied celebrity endorsements. Whatever an actor is seen using or driving is seen as a statement of preference and choice (Marshall, 2001, www.normmarshall.com).

The purpose of this paper is to investigate the influence of product placements in television serial comedies on consumer attitudes. We draw from Heider's balance theory (1946,

1958) to explain the attitudinal effects of placed products on consumers in terms of a triadic relationship between consumer viewers of a sitcom, characters in it, and products that the characters use. The theory, originally developed in social psychology, was based on the assumption that humans have a preference for harmonious situations in which the relations in a triad composed of a person, another person, and an object is balanced when "all three of the relations are positive or when two of the relations are negative and one is positive" (Heider 1958, p. 202). Heider (1958) and later Osgood and Tannenbaum (1955) identified two types of relations (1958): liking refers to a person’s degree of positive or negative evaluation of something or someone and unit refers to the level of association between a person and another person or object, based, for example, on similarity, causality or ownership. A basic assumption of this theory is that liking relations and unit relations are mutually interdependent and tend toward a balanced state. In other words, if person X has a strong and positive relationship with person Y, then person X will tend to align his/her attitude toward an object/issue with that of person Y and this attitude alignment will be especially strong if the object/issue is central to person Y (Davis and Rosbult 2001).

Balance theory was later applied to advertisements in terms of a balanced relationship

2 among consumers, celebrity endorsers and products to explain the transfer of attitudes from the celebrity to the product endorsed. Our transposition of the triad from celebrity advertisements to product placements rests on reconceiving the celebrity endorser as a program character who uses a placed product such that the entities are conceived of as consumers, characters in programs, and embedded products that are consumed. In accordance with the reframed theory, we present the "3Cs Model of Product Placements" to explain that a consumer's attitude toward a product placed in a program will depend on the strength and valence of two relations: one between the consumer-character and the other between the character-consumed. The model contributes to greater understanding of placement effects by extending prior research in which two factors were found to affect viewers' attitudes -- the way products are integrated in the program (d’Astous and

Seguin 1999; Russell 1998, 2002), and the relationships viewers have with the program characters (Russell, Norman, and Heckler 2004) -- to include product-character relationships within a program genre, hitherto unexplored research areas.

Genre theory in literary criticism is used to analyze product-character relationships specific to a media vehicle, for each one -- soaps, reality shows, mysteries, and so forth -- contains elements common to one genre but not to others. Sitcoms are comedic works in which the dramatic elements of character, plot, setting, dialogue and so forth interact to form a unique pattern (NOTE – note sure it is in our favor to state early on that there would be genre differences). Analysis of the pattern requires a turn to genre theory -- the branch of literary criticism dedicated to the classification of works -- as the most useful source of information about sitcom structure and purpose (Feuer 1992, Fowler 1982). The point of classification is to impose "order into the wide range of texts and meanings that circulate in our culture for the convenience of both producers and audiences" (Fiske 1987, p. 109) presumed to possess genre

3 knowledge by dint of exposure to similar texts.

Second, to explain the relationships between viewer and character, we draw from parasocial interaction theory, instrumental in explaining the strength of the relationships between viewers and characters (Horton and Whol 1956). As viewers establish closer relationships with television characters, they tend to identify with them (Feilitzen and Linn 1975; Levy 1962), come to know and approve of their consumption practices, and even model their own consumption on that of the models (Russell et al. 2004). Figure 1 depicts the 3Cs Model of product placement attitudes with contributions of genre theory to the understanding of product- character relations and the contributions of parasocial interaction theory to the consumer- character relations.

INSERT FIGURE 1 ABOUT HERE

The Placement Industry and Sitcoms

The reason for studying placements is that clearer understanding can contribute not only to persuasion research but also to the placement industry's marketing strategy. Placements themselves are based on a marketing practice in which a firm pays for the right to feature its branded products in mass media films and television programs. The practice of integrating brands and entertainment has become a $1 billion dollar industry (Lasswell 2004) and is so commonplace that by 2005 placements are anticipated to occur in 75% of prime time broadcast network scripted shows (Consoli 2004).

Sitcom placements provide particularly strong evidence of the strategy's effectiveness in boosting sales across product categories. For example, in the apparel category, a 35% rise in sales of Nick & Nora pajamas occurred between 1995 and 1998 when Ally McBeal (The Ally

McBeal show) wore them as lounging outfits (Carter 2000; Stanley 1998). Similarly, in the non-

4 alcoholic beverage category, the 1990s growth in coffee bars such as Starbucks and New World

Coffee has been attributed to their importance in urban sitcoms such as Frasier or Friends (Tueth

2000); in the alcoholic beverage category, the popularity of the Cosmopolitan martini is often attributed to its association with the women in Sex and the City. Indeed, the impact of placements on product sales is so well accepted that the strategy has become part of the marketing mix of over 1,000 U.S brands (Marshall and Ayers 1998), including large Fortune 100 companies like Procter & Gamble, PepsiCo and Anheuser-Busch (Vranica 2004).

State of Research on Product Placement Effects

Over the last decade, a healthy stream of research has developed to test the effects of

product placements on audiences. Ever since product placement was identified as a hybrid

message (Balasubramanian 1994) consumer and advertising researchers have begun to study its

impact on memory (Babin and Carder 1996; Gupta and Lord 1998; Nelson 2002, Russell et al

2004), attitudes (d’Astous and Seguin 1999; Russell 2002) and behavior (Auty and Lewis 2004;

Russell and Puto 1999). Still little is known about the driving forces that affect higher order

effects, especially the attitudinal and behavioral consequences of product placement. Such

effects must take into account the contextual factors which contribute meaning to the placed

product, for example, the implied endorsement of a character or the way the product is used in

the program (Sherry 1995). The notions of liking and unit relations is implied in much placement

research to date but have not yet been integrated into a model of placement attitudes or tested

empirically.

One of the main challenges for product placement researchers is the difficult compromise

between internal and external validity. Controlled experiments allow the manipulation of unit

5 relations related to the placed product, such as plot connection (Russell 2002), but cannot readily manipulate relations that viewers establish with the characters because made-for-research stimuli do not allow the development of viewer-character relationships. Conversely, field studies relying on existing programs allow the testing of real audience factors such as involvement with the program or connectedness to the characters (Russell et al 2004) but provide less control over the stimulus and placements (Babin and Carder 1996).

THE THREE CS MODEL OF PLACEMENT EFFECTS

Our paper extends the research on product placement effects by proposing and testing a new model to describe the attitudinal effects of product placements on viewers. The 3Cs model presented in Figure 1 depicts the balance between three components of the triad (Heider 1958): the consumer, the consumed and the character. It presents the theoretical roots underlying the character-consumed dyad and the consumer-character dyad that allow us to predict attitudinal effects for the third dyad: consumer-consumed (the consumer’s attitude toward the placed product).

The theoretical root for studying the character-consumed dyad is genre theory, a form of classification found in literary criticism, here adapted to analyze product-character relationships in televised serial dramas. Genre studies reveal that different types of dramas (soaps, reality shows, Westerns, police procedurals) feature different types of characters and product roles, and sitcom (situation comedy) theory -- a new branch of comedy criticism -- is necessary to analyze the context in which characters and products appear. The justification is that the more we know about stimulus attributes, the better our understanding of their effect on consumer responses.

The theoretical root for studying the consumer-character dyad is parasocial interaction

6 theory, drawn from social psychology and mass communication. Since the balance theory framework (Heider 1958) assumes two components of relations, we rely on two constructs to capture the consumer-character relationships: attitude toward the character and parasocial interaction with television characters. Parasocial theory explains consumer-character relationships in television series and the affective attachment to characters that viewers develop over time. In this regard, the temporal longevity of sitcoms over a number of seasons (equivalent to years) -- for example, Friends lasted for nine seasons, and Frasier for eleven -- makes them especially suitable as loci of parasocial attachments in that viewers can get to know the characters and vicariously participate in their lives (Russell et al 2004).

These theoretical perspectives are integrated to produce a comprehensive and parsimonious model based on the extant balance theory literature to explain the relatively new and rapidly increasing phenomenon of product placements.

Analyzing Character-Consumed Relationships: A Turn To Genre Theory

The total structure of a dramatic work ... depends on a very delicate balance of a multitude of elements, all of which must contribute to the total pattern and all of which are wholly interdependent (Esslin 1976, p. 53).

Television texts are genre-bound (Fiske 1987), especially sitcoms -- the "signature" program type (Feuer 1992) invented for television -- which are essentially formulaic. From the outset, sitcoms have been created in conformance to conventions that dictate the deployment of dramatic elements, including, for the first time, products as part of the pattern (Neale and Krutnik

1990). Formulaic structure (Cawelti 1970) enables mass production of cookie-cutter programs that yield increased industry profits owing to low research and development costs, economies of

7 scale, and high volume. In addition, standardized entertainment products guide the meanings of programs for viewers, who "are influenced, even manipulated, by the genres they [the programs] are fitted into" (1987. p. 111; cf. Feuer 1992).

Serialization of sitcoms also guides viewer expectations of potentially long acquaintanceships with characters, for longevity is a built-in feature that goes hand in hand with standardization in impacting both profitability and viewer-constructed meanings. Just as manufacturing efficiency is increased when the same product is produced year after year, so too are viewer perceptions of meaning reinforced by repeated viewing of ongoing dramas (Esslin

1976). Each episode in a serial contains the same characters engaged in similar actions, which suggests that extended viewing contributes to audience understanding of what is going on. To examine the structure and content more fully, let us begin with a discussion of the generic conventions about the nature and role of characters, plots and products.

Characters: Familiar, Recognizable, Two-Dimensional

The dominant sitcom rule is that characters are "the most crucial element" (Wolff 1988, p. 14) in what has often been called "character comedy" (Feuer 1992, p. 154). The genre's popularity since the 1950s (I Love Lucy, The Honeymooners) is attributed to the presentation of types of people whom viewers recognize and with whom they become familiar over time.

Familiarity allows viewers to experience a "comforting feeling of security" in the presence of characters who have, owing to the intimacy of the medium, been invited into the viewers' homes

(Newcomb and Hirsch 1983). Viewer comfort is reaffirmed by the characters' predictability, for they are stereotypes or stock characters (Frye 1957 [1973]) unlikely to behave in threatening or disturbing fashion (Brown 1992). The comedic types are also familiar because they are part of

8 the sitcoms' literary heritage from the Old Comedy of Athens, a "live tradition, richly variegated and culturally interrelated" (Levin 1987, p. 4) and populated by stock characters such as lechers, sluts, fools, and liars. The same types reappear in contemporary sitcoms: Joey in Friends is lecherous; Samantha in Sex and the City is promiscuous; Jack in Will and Grace is a fool; and

George in Seinfeld is a liar.

However, even though the characters are stereotypical, they are no longer one- dimensional figures such as Archie Bunker or Ralph Kramden, dominated by a single quality and knee-jerk behavior (Bergson 1900 [1956]). Rather, the 1970s Mary Tyler Moore show introduced characters who are more two-dimensional and show some capacity for development and change. The decrease in stereotyping came about as a consequence of cultural changes set in motion by the 1960s liberation movements, which ushered in "new wave" programs that "most markedly altered the 'grammar' of the formula" (Grote 1983, p. 153). Nonetheless, sitcom characters remain recognizable as players in a comic form that by its very nature eschews fully rounded complex heroes and heroines (Forster 1954).

The sitcom formula locates characters within a group -- the focus of "ensemble playlets"

(Brown 1992, p. 511) representing a lifestyle community held together by shared values such as

friendship, emotional interdependency, loyalty, and, since the 1970s, liberalism (Eaton 1978/9).

Emphasis on communal values demarcates the group as a model of middle-class stability, first

defined as the domestic family, but later extended to include office mates (Mary Tyler Moore) or

military comrades (M*A*S*H, McHale's Navy). Most recently, domestic groups are composed

of friends functioning as a surrogate family, a response to cultural changes in the nature of

family composition that now include unmarried singles living together, gay partners, live-at-

home adult children and single-parent households, all of which appear in programs such as "Sex

9 and the City," "Friends" and "Will and Grace."

Products: Quasi-Characters

Perhaps the most unique element in sitcoms is the role of products as quasi-characters in consumption scenarios that pervade the genre. Prior research (Hirschman, Scott, and Wells 1998;

Solomon and Greenberg 1993) on product-character relationships indicates that products enhance the reality of a consumption scenario by serving as "psycho-cultural" cues to audience construction of meanings about characters and groups (Sherry 1995). Meaning-construction is said to emanate from the characters' "narcissistic binding of ego to product" (Wernick 1991, p.

34) that symbolizes the link between social values and mass produced goods in a consumer society. Like other television genres, sitcoms are "popular when their conventions bear a close relationship to the dominant ideology of the time" (Fiske 1987, p. 112), one in which popular slogans such as 'you are what you buy' (or wear, or eat) reveal the importance of products such as clothing, make-up, home furnishings, food, and beverages as signifiers of character (Cornwell and Keillor 1996). The artistic rationale for placing such products is that they "represent consumption practices in a fairly accurate and naturalistic way, so that the settings and actions will seem real to the average American" (Hirschman, Scott, and Wells 1998, p. 32).

In sum, formulaic sitcom rules guide the creation of programs in which characters, plots, and products are interwoven in a typical generic pattern marked by a prominent display of consumption habits (Hirschman, Scott, and Wells 1998; Solomon and Greenberg 1993). The realism of a sitcom's socio-cultural milieu is enhanced by the inclusion of products associated with the characters who communicate cultural meanings to the audience (Fiske 1987, Wolff

1998).

10

Hypothesis 1: Character-Consumed Relationships: Valence And Intensity

The categorization of products as quasi-characters underlies the prediction that placed products will function importantly in sitcom character revelation (Wolff 1988), for the genre depends on a carefully drawn socio-cultural scenario filled with products that provide information about the characters who use them. These statements about preferences vary in valence, for characters cannot automatically be assumed to be favorably inclined toward products seen and discussed (Butler 2001). That is, they may display positive or negative attitudes. For example, in Will and Grace, the main characters make fun of their boring friends who want to eat at Olive Garden, a mass-market chain restaurant (Chang and Roth 1999) instead of a more upscale innovative place. Similarly, in Frasier, he and his brother Niles, both affluent food and wine snobs, mock their father's simple food preferences for beer, beef jerky and surf n' turf dinners (Chang and Roth 1999).

In addition to valence, characters' attitudes to products have been found to show weak/strong intensity depending on the strength of character-product associations. Weak associations refer to products that are merely placed in a character's home to enhance the realism of the set -- a form of "reality engineering" more implicit than product use or discussion

(Solomon and Englis 1994). Classic examples of backgrounding a product as something that is

"just there" are abundant in Friends, where the characters’ apartments contained items ranging from Heineken beer to Barilla spaghetti sauce. In contrast, strong associations refer to products used more explicitly to convey ideological information about a character (Hirschman 1988) that enriches his or her depiction (Holbrook and Grayson 1986). In this situation, the product contributes to the character's identity and/or strongly reflects a character's values (Hirschman

11 1988) and the associations may be relatively strong (Hirschman 1988; DeLorme and Reid 1999).

In the last season of Sex and the City, for instance, Miranda was defined by her pregnancy and motherhood and products related to that experience were strongly associated with this life transition, from the Bellini accessories and Kiehls baby lotions she purchased for the newborn to her Weight Watchers’ experience to shed baby weight.

Thus, our first propositions about the nature of placements in situation comedies relate to these two dimensions of character-product relationships:

H1A: Character-Consumed relationships in situation comedies can be positive or negative

(VALENCE).

H1B: Character-Consumed relationships in situation comedies can be weak or strong

(INTENSITY).

Analyzing Consumer-Character Relationships: Implications From Parasocial Interaction Theory

There is a long tradition of research of viewer-character relationships in the television context, spanning the fields of communications (Fiske 1987; Jenkins 1988; von Feilitzen and

Linné 1975), psychology (Horton and Whol 1956; Maccoby and Wilson 1957), and marketing

(Churchill and Moschis 1979; Kozinets 2001; Lavin 1995; McCracken 1986; Russell et al.

2004). This research informs the processes whereby liking and unit relations develop between viewers and television characters. Following a single episode, viewers are likely to develop attitudes toward characters. These attitudes reflect overall evaluations of the characters.

However, over the course of watching multiple episodes of television series, viewers can become actively vested in the characters whose lives they closely follow and care about, and begin to interact with them as if they were real, in a parasocial way. Like traditional interpersonal

12 relationships, these relationships develop and strengthen over time. Such relationships are captured by the construct of parasocial interaction, defined as audience interaction with fictional characters conceived of as real (Horton and Whol 1956; Levy 1962). Whereas attitude toward a character resembles Heider’s liking relation, parasocial interaction emphasizes the depth of the interpersonal relationship, and thus better captures the unit relation between consumer and television character. It represents the degree to which the character is viewed as a meaningful referent other, a concept closely related to unit-forming factors such as “similarity, proximity, common fate (…) or past experience” (Heider 1958, p. 177).

Parasocial relationships are expected to be especially important in situation comedies because of the genre’s characteristics. The stronger the audience identification, the more committed it is to re-experience the emotional reward of relationships with characters and the cognitive reward of increased information about what the characters are up to this time (Pfister

1991, p. 99).

Hypothesis 2: Consumer- Character Relationships: Attitude And Parasocial Interaction

Based on the literature, we propose that parasocial interaction and attitude are two distinct constructs. Attitude toward a character, which reflect Heider’s liking relation between a viewer and a character, is expected to vary in valence, from negative to positive. Parasocial interaction, which reflects the closeness of the relationship between a viewer and a character

(Russell et al. 2004), is expected to vary in intensity. This rationale leads to the following hypotheses:

H2A: Consumers’ attitudes toward characters (AC) can be positive or negative.

H2B: Consumers’ level of parasocial interaction with characters (PARA) can be weak or strong.

13 H2C: Consumers’ attitude toward a character and level of parasocial interaction with that character are distinct constructs.

The Three C’s Model: Implications from Balance Theory.

Heider’s balance theory and its heirs (extensions?) stipulate that people prefer balanced triadic attitudinal structures. This theory has direct consequences on predicting the evaluation of products associated with a character in a program. In the context of our proposed consumer- character-consumed triad, balance exists when all three relations among elements are positive, or when two relations are negative and one positive. When consumers are exposed to a product placement associated with a character, their drive to establish attitudinal balance implies that their evaluation of the placed product will be directly affected by: the relationship and attitude they have with the character and the relationship that exists between the character and the product.

Hypothesis 3: Consumer-Consumed Relationships: Deriving Placement Effects based on the

Three C’s Model

The final set of hypotheses specify the linkages between consumer-character variables

(attitude and parasocial interaction) and character-product variables (valence and intensity of the associations). Per Heider’s balance theory, attitudes toward the placed product will be a function of all four variables: AP = f(AC, PARA, VALENCE, INTENSITY)

According to hypothesis 1, product-character relations vary in valence from negative to positive. Based upon Heider’s theory, if the product-character relation is positive, a balanced inter-attitudinal state implies that AP would be positively related to AC and PARA. Conversely,

14 if the product-character relation is negative, a balanced state implies that AP would be negatively related to AC and PARA. Following H1A's prediction that product-character relations can be positive versus negative we thus posit that:

H3A: if the product-character relation is positive, AP is positively related to AC and PARA

H3B: if the product-character relation is negative, AP is negatively related to AC and PARA

Based on hypothesis 1B, product-character associations not only vary in valence (H1A) but also in intensity (H1B). In Heider’s framework, the intensity of the association reflects the unit relation between product and character and it is expected that a stronger unit relation results in a greater need for attitudinal balance. Therefore we predict that the intensity of the product- character relation will moderate the relationship between AP and the character variables: AC and

PARA. In other words,

H3C: the effects of AC and PARA on AP are moderated by the intensity of the product-character relation.

Finally, based on the extant research, we expect that parasocial interaction levels will be better predictors of attitudinal balance, and thus AP, than the more general attitudes toward character. [NOTE: this is indeed what we find but I do not know whether it should be a formal hypothesis or simply a statement in this section]

H3D: Parasocial interaction with characters (PARA) will be a stronger predictor of attitudinal effects than attitude toward characters (AC).

THE STUDY

In order to maximize external validity, the hypotheses were tested in a “live” study using actual episodes of situation comedies as stimuli.

15

Pre-Test

Stimulus. The pretest stimuli were actual sitcom episodes--the season finales of Will and

Grace and Friends--watched in real-time by "real" viewers.

Sample Recruitment. The method of recruiting a sample of respondents under these conditions was pre-tested, using either Will and Grace or Friends as the stimulus, with pre- recruitment selection performed by sending email invitations to a group of male and female viewers of a season's worth of one or the other series. In the selection process, potential respondents were asked if they were willing to watch the season’s finale episode of the series, and later on that evening log on to a secure web site to complete a questionnaire about their viewing experience. A $50 lottery cash prize was offered as an incentive to all of those who participated in the study. A professional online survey website (surveymonkey.com) was used to program the survey and collect the data. After watching the episode, the respondents received an email message with the link to the website. The time when each respondent logged on and the amount of time it took to complete the self-paced questionnaire were recorded.

Of the 38 viewers initially recruited, 26 completed the survey and did so right after the viewing of the episode (n Will and Grace = 16, and n Friends = 10). Respondents' ages ranged from 16 to 39, and females outnumbered males (20 females and 6 males). Findings indicate that the recruiting methodology achieved a 70% timely response rate, which we view as successful.

Findings. Attitudes toward the program (mean on all three attitude items) were strongly positive (M = 6.42 (SD = .95) on seven-point scale). On average, respondents listed 2.38 (SD =

1.96) products or services and a total of 2.66 (SD = 2.04) characters, with no statistically significant differences between the two programs (t(24) = .104 and -.949 respectively, p > .05).

16 Reported levels of parasocial interaction with each character varied from 1.22 to 5.00, as measured by a nine-item version of Rubin, Perse and Powell (1985)'s parasocial attachment scale

(see Appendix 2). Results indicate that the recruitment method was suitable for testing the hypotheses about the relationships between AP, AC, PARA, and character-product associations using real programs as stimuli, real viewers as respondents, and same-day recall.

Main Study

As in the pre-test, the stimuli were actual sitcom episodes--the season premieres of Will and Grace, Friends, Everybody Loves Raymond, and Frasier--watched in real-time by "real" viewers.

Sample. The procedure replicated that in the pre-test, with pre-recruitment of a panel of male and female viewers who indicated that they were regular viewers of at least one series.

Recruitment was conducted in three waves. First, email contact was initiated with 45,833 U.S. consumers listed in a purchased database, and the consumers were asked to participate in a general survey on television viewing. Second, among those who initially responded to the invitation (n = 10,329), we contacted those who indicated that they regularly watched one of the four sitcoms (n = 1,628), inviting them to participate in the study. During the selection process, the respondents were told that they would have to watch the first episode of a series, and then later on that evening log on to a secure web site to complete a questionnaire about their viewing experience. A total of 690 agreed to participate, and the same professional online survey website was used to program the survey, record the time when each respondent logged on, record the amount of time it took to complete the self-paced questionnaire, and collect the data. The response rate was 37.8%, with a total of 261 respondents completing the questionnaire (n

17 Everybody Loves Raymond = 76, n Frasier = 32, n Will and Grace = 33, and n Friends = 123).

Respondents' ages ranged from 16 to 84 and the majority were female (79.5%).

Research Instrument. The questionnaire included three parts. Part 1 contained a three- item semantic differential measure of the respondents' attitude toward the series, with each item consisting of a seven-point attitude toward the show measure (I like it/I dislike it, good/bad, unpleasant/pleasant). The television connectedness scale was also administered as an indicator of the general intensity of the relationship between the respondent and the program and its characters (Russell et al. 2004).

After the respondents completed part 1, part 2 began with an explanation that "in the show, sometimes people are seen interacting with certain products or services (eating certain foods, drinking certain things, using certain modes of transportation, traveling to certain places, shopping at certain places, consuming certain forms of entertainment, using certain types of services, etc)." In part 2, the respondents were asked to remember up to five instances of similar consumption-related events, and for each instance they were asked to type in the product or service, complete three seven-point attitude items about the consumption event (I liked it/I disliked it, good/bad, unpleasant/pleasant). Then they were asked to fill in open-ended comments about why they thought the product was there, what they thought it contributed to the show, and which character(s) it was most closely associated with and why. An additional three-item scale measured the intensity of the product-character association ([product] was strongly associated with this character; the character interacted with [product]; the character expressed like/dislike for [product]). Finally, the respondent indicated the valence of this association (the association between [product] and this character was positive). The procedure was repeated for each instance mentioned.

18 When part 2 was completed, each respondent turned to part 3, which contained scales measuring responses to the characters he or she had listed. The first scale was a three-item seven- point attitude to the character measure (I like him or her / I dislike him or her, s/he is good / bad, s/he is pleasant / unpleasant), and the second was the nine-item parasocial attachment scale

(Rubin et al. 1985, see Appendix 2). The scales were followed by an open-ended question asking the respondent to describe the character in as much detail as possible (what is she/he like? What does she/he enjoy, and so forth). The procedure was repeated for each character listed in part 2.

At the end of part 3, demographic information was collected.

DATA ANALYSIS AND FINDINGS

Overview of Data. Attitudes toward each episode (mean on all three attitude items) were strongly positive and did not significantly differ by series (MRaymond = 6.81, MFrasier = 6.73,

MFriends = 6.78 and MWill and Grace = 6.84, F(3, 257) = .252, p > .05). Connectedness levels did not significantly differ either (MRaymond = 3.10, MFrasier = 3.02, MFriends = 3.26, and

MWill and Grace = 3.02 (F(3, 260) = 3.197, p = .02). On average, respondents listed 2.97 (SD =

1.64) products or services, with the number of products significantly lower for Friends (M =

2.52) than for the other programs (MRaymond = 3.22, MFrasier = 3.69, and MWill and Grace =

3.40; F(3, 250) = 6.497, p < .01). The most likely explanation is that the Friends episode took place in a hospital/maternity room, which affords less opportunity for a wide array of products.

Characters associated with the products appear in Table 1.

INSERT TABLE 1 ABOUT HERE

When appropriate, our hypotheses were tested using the full dataset. However, to take into account the differing number of products listed by each respondent, the analyses related to the complete 3Cs model used only data from the first product listed by each respondent and only

19 those products mentioned by at least 10% of the respondents. These first-listed products and the characters associated with them (n = 226) are depicted in Table 2.

INSERT TABLE 2 ABOUT HERE

Hypothesis 1. Across all products listed, the intensity and valence of the character- product relationships varied, thus supporting hypotheses 1A and 1B. The intensity of the character-product relationships (mean across the three five-point scale items, α = .84) varied from 1.0 to 5.0, with a grand mean was 3.84. The valence also reflected variance in these associations, with a grand mean of 3.88. Because this valence variable was a Likert-scale going from negative to positive, we divided the observations based on the mid-scale point: for the first product listed: 31.8 % of the ratings for first-listed products were at or below the mid-scale point and, across the total of 914 products listed, 36.9% were at or below the mid-scale point.

Hypothesis 2. Attitude toward individual characters were generally positive (grand mean on all three attitude items = 6.47) and did not differ significantly by series (F(3, 237) = 1.396, p >

.05). We note that the skewness of these attitudinal scores limits our ability to truly assess hypothesis 2A. Reported levels of parasocial interaction with each character varied from 1.00 to

5.00, with an overall mean of 3.68 and were not significantly different by program (F(3, 239) =

.464, p > .05). The range of parasocial interaction scores provides support for hypothesis 2B.

Finally, to test hypothesis 2C, we computed the correlation coefficient between AC and PARA.

This correlation was .214 (p < . 01), thereby suggesting that only a minor correlation exists between the two constructs and that they are indeed distinct constructs, in support of hypothesis

2C.

Hypothesis 3. To test hypothesis 3, we conducted a series of regressions of AP as the dependent variable on the different predictor variables. Hypotheses 3A and 3B predict that, for

20 balance to occur, AC and PARA would be positively associated with AP if the valence of the product-character association is positive (H3A) and negatively associated with AP if the valence of the product-character association is negative (H3B). Furthermore, hypothesis 3C predicts that the intensity of the association would moderate these effects.

Since the hypotheses predict different effects depending on the valence of the product- character association, we conducted separate regression analyses on that basis: the dataset was split between the negative (n = 135) and the positive (n = 90) product-character associations. For each dataset, a first regression was conducted, including simple main effects only (attitude toward the character - AC - and parasocial interaction with the character - PARA). The second regression model included these main effects as well as the interaction terms with the dichotomized intensity variable (AC X INTENSITY and PARA X INTENSITY) to test hypothesis 3C. The intensity variable was operationalized as weak (INTENSITY = 0) or strong

(INTENSITY = 1) based on the mean on the three-item product-character association measure.

INSERT TABLE 3 ABOUT HERE

The regression results appear in table 3. Together, these regression analyses show that parasocial interaction affects attitudes toward associated products but the effects differ depending on both the intensity and valence of the product character associations, inline with our hypotheses. Notably, we find that parasocial interaction is the only character variable that consistently affects attitudes to placed products. Specifically, the results for positive product- character associations show that parasocial interaction is positively related to attitude toward the placed product (supporting H3B) regardless of the intensity of the product-character association

(no moderating effect of intensity). The results for negative product-character associations show

21 that parasocial interaction is negatively related to attitude toward the placed product (supporting

H3B) but only when the association is strong (moderating effect of intensity).

Overall, the findings are supportive of the 3Cs model as it relates to parasocial interaction

between consumers and sitcom characters. When the product-character association is negative,

PARA significantly affects AP if the intensity of the product-character association is strong (i.e.

if consumers perceive a strong association between the character and the product, PARA affects

AP negatively). When the product-character association is positive, PARA affects AP regardless

of the intensity of the product-character association (if consumers perceive a positive association

between the character and the product, PARA affects AP positively). Support for hypotheses

related to AC’s effects is limited. In fact, the only significant effect of AC on AP was found for strong negative product-character associations. This provides support for hypothesis 3D: parasocial interaction is a stronger predictor of AP than attitude toward the character. However, as acknowledged earlier, this could be due to the positively skewed distribution of attitude scores.

LIMITATIONS

Nonetheless, even though this study provides support for the influence of characters and product-character associations on attitudes to placed products, the generalizability across media, vehicles and genres is not possible. The focus on situation comedies alone prevents application of the findings across other popular television genres such as soap operas or reality shows, the newest, most popular, and least-studied television format – as well as across other media such as books or music videos. Given the naturalistic and consumption-oriented milieu of sitcoms, it may be that the role of products is more formulaic than it is in other genres.

22 Another limitation is the methodological trade-off, as the real television stimuli provide more realism and enhanced external validity but diminished opportunity to manipulate product- character associations. Although it allowed the study of the phenomenon in real viewing conditions, the methodology lacked the full control over the scheduling of stimuli (Campbell and

Stanley 1966) and, unlike experimental data, correlational data does not allow tests of causality.

The complexity and abundance of product-character associations in real sitcoms necessitated that model testing be limited to the characters most clearly associated with a product. Still, we feel that the benefits of externally valid stimuli, a real-world sample of viewers, and naturalistic settings, outweigh the disadvantages of the methodology. We suggest that these limitations can be overcome in future research based on examination of media other than television, genres other than sitcoms, and inclusion of different character types.

Finally, the studies’ reliance on explicit recall measures inherently limited the analyses to high plot placements, those that viewers are aware of (Russell 2002). The persuasive processes underlying low plot placements, which viewers do not necessarily remember, are likely to differ and merit additional research.

DISCUSSION AND RESEARCH IMPLICATIONS

Notwithstanding correctable limitations in future research, the findings reveal support for the 3Cs model of attitude toward placed products. Support is also found for the superior predictive ability of parasocial interaction over attitude toward a character. By integrating the influences received from television referent others, and the genre implications of product- character associations, the 3Cs model contributes to a richer understanding of product placement effects.

23 Further, the study’s attempt to “confront rather than evade external validity” (Wells 2001, p. 496) paves the way for the use of methodology that allows for collection of real-world data from non-student samples. This advances product placement research, which has been limited by the lack of either external validity stemming from the use of controlled experiments to manipulate the placements (Russell 2002), or internal validity stemming from the use of real- world stimuli that respondents may have been exposed to previously.

There are practical public policy implications of the study’s findings as well as numerous research implications. On the practical level, the finding that viewers’ attitudes are influenced by characters portrayed as consumers and/or non-consumers of bad-for-you products is significant and extends research on the placement of ethically charged products (Gould, Gupta, and

Grabner-Kräuter 2000). For example, television, radio, film, and especially music videos have been identified as potential sources of excessive drug use and drinking (Gerbner 1995; Hundley

1995) to such an extent that as of 2003, MTV is “cracking down on product placement,” particularly those featured prominently in songs such as Pass the Courvoisier (Arango 2001, p.

31). On the government level, since 1998, the Office of National Drug Control Policy has shown interest in increasing negative portrayals of drug use in television programming and associating destructive consumption behaviors with negative role models (Jurith 2001).

In addition, a full plate of research implications can guide future researchers in the extension of product placement research to other media and other generic vehicles, which can be studied in a systematic manner based on the premise that "context is all" (Esslin 1976, p. 53). To date, contexts other than films and television shows have barely been the objects of research, perhaps because research on placements is at an early stage. This creates abundant opportunity

24 for investigation in a wide variety of contexts where product-character associations and parasocial relationships seem likely to occur. Among the many contexts are placements in books

(Fay Weldon's The Bulgari Connection ; Lauren Weisberger's The Devil Wears Prada ); hip-hop music videos and songs (Busta Rhymes' songs Pass the Courvoisier ; Pass the Courvoisier 2 ;

Ludicris's Southern Hospitality , which mentions Cadillac repeatedly); online video games ( The

Sims Online ); and Broadway shows (Baz Luhrmann's La Boheme ). Yet despite the use of placements in so many media, and advertisers' apparent faith in their effectiveness, there is still little research on the relationship between a placement's role in a specific media exemplar or on the precise consumer effects it is assumed to achieve. Issues to be investigated in future research include the purpose of media placements in context, the means whereby they are introduced into the vehicle, the consumer-character relationship (if any), and the attitude changes that are anticipated.

Just as the media contexts require future research, so too do the program genres within a single medium, notably television. Here, even though the history of placements in sitcoms, soaps, and game shows is as long as that of television itself (Lavin 1995), new program types do crop up--for example mob dramas ( The Sopranos ) and reality shows ( Joe Millionaire , American

Idol ). Indeed, the burgeoning of reality shows has been considered the fuel that powers "the

apotheosis of recent product placements" (Carter 2001, p. A1), which in 2003 will occupy 15 of

18 the half-hour time periods on Monday, Tuesday, and Wednesday nights, and half of the slots

on other nights (Carter 2003). However, reality shows differ fundamentally from any other

television generic half-hour program in that whereas other programs are scripted and populated

by professional actors playing the characters, reality shows are unscripted and populated by real

people who ARE the characters. A real person such as Richard Hatch in Survivor is said to have

25 "introduced a striking new character type to the contemporary culture: the Machiavellian hero" whose nature comes from "the mutations made possible by reality TV" (Thompson 2001, p. 23).

Furthermore, the participants in these shows appear for only a limited time period (eight to 10 weeks), and then vanish completely. These special characteristics may very well affect the parasocial relationships, but until further research is conducted, we cannot speak knowledgeably about placement effects in this new genre. In conclusion, the findings that consumption scenarios influence one type of television program suggests the need for future investigation of progressive new empirical methodologies as well as the consideration of new theories that draw from other paradigms (Kuhn 1962; Laudan 1984).

26

Figure 1: The Three Cs Model of Product Placement Effects

The Consumed (Placed Product or Brand)

Attitude toward placed Valence of product-character association (H1A) product (H3) Level of association (H1B)

The Consumer The Character (Viewer) (Endorser)

Attitude toward character (H2A) Parasocial interaction w/ character (H2B)

27 TABLE 1: PRODUCT-CHARACTER ASSOCIATIONS BY PROGRAM

Everybody Debra Raymond Robert Marie Frank Children Loves 84 82 71 67 53 2 Raymond Ms. Niles Daphne Frasier Martin Roz Alice Judge Frasier Moon 35 34 32 27 16 4 2 1 Mr. Rachel Ross Joey Phoebe Monica Chandler Friends Geller Emma Nurse 189 126 117 111 109 92 9 3 1

Will and Grace Will Karen Jack Leo Rosario

Grace 31 29 21 20 7 2

28 TABLE 2: PRODUCTS SELECTED FOR ANALYSIS

Everybody Product / Character Raymond Marie Robert Debra Frank Loves Cake 1 14 10 1 3 Raymond Chips 4 2 Ginger ale 4 Pickles 14 Hawaiian shirt 4 Tums 3 TV 6 1 Vacuum Cleaner 7 Frasier Product / Character Frasier Martin Niles Daphne Cell phone 1 2 Potato chips 3 Pudding cups 5 Wedding ring 2 Wine 7 1 Friends Product / Character Rachel Ross Phoebe Joey Balloons 6 Engagement ring 36 9 1 Flowers 3 11 Game boy 21 Plastic fruit 7 Sandwich 1 9 Will and Product / Character Grace Karen Will Grace Blueberry muffin 10 Frozen yogurt 8 1 Twizzlers 3 2 Wine 3

29 TABLE 3: PARAMETER ESTIMATES FOR REGRESSION MODELS

Positive Product-Character Negative Product-Character Associations Associations Step 1 Step 2 Step 1 Step 2 Beta Estimate Beta Estimate Beta Estimate Beta Estimate Intercept 4.428 * 4.498 * 3.785 * 3.785 *

AC .065 ns -.091 ns .231 ns .125 ns PARA .311 * .566 * .067 ns .224 ns AC x INTENSITY .203 ns .428 *

PARA x INTENSITY -.348 ns -.633 *

R 2 5.8% 7.2% 2.9% 9.6%

Note -- DV = AP (attitude toward the product); *: p<.05 Intensity is a dummy variable where 1 = high, 0 = low.

30 APPENDIX 1: SERIES AND EPISODE ANALYSES

Raymond – Series Description Characters The Barone family Robert Ray's brother, a policeman, lives with his parents Ray Robert's brother, a sportswriter for Newsday (Long Island daily paper) Debra Ray's wife, housewife, she and Ray live across the street from Marie and Frank. They have 3 children – Ali is 6 or 7, twin boys are 3 Marie Ray and Robert's mother Frank Ray and Robert's father Setting Long Island suburbs Situation The group is the Barone family, middle-class Italian, where men work and women stay at home. Conflict between Deborah, who is a terrible housekeeper and cook, and Marie, who is a critical mother- in-law. Raymond – Episode Description “ The Cult” Ray is eating PICKLES, and talking to Debra about them, as they are eating dinner and drinking GINGER ALE. Robert, who is depressed and eating TUMS in bed, gets up and goes across the street to visit. He enters wearing a HAWAIIAN SHIRT, and says that is distraught about ongoing feud between Marie and Debra, and decides to join a support group called "Inner Path" for therapy that will enable him to achieve inner peace. The family thinks that it is a cult, and Ray accepts Robert's invitation to go to the next meeting. Ray disapproves of the group, but says that he is depressed by the unending feud. Ray wants to get Robert away from the group, and Robert's partner - Judy - suggests an intervention. Robert is invited to come over to Ray's house for some CAKE, and the whole family is there. Robert tries to resist their intervention, but the family tries to express their feelings about Robert: Ray expresses his love for Robert; Marie wants to feed everyone some CAKE; and Frank calls Robert a "lughead" for not understanding how his father feels. Debra and Marie are so worried about Robert that they decide to make up. Afterwards, Robert says that he and Ray planned the support group "therapy" to get Marie and Debra to make up.

31

Raymond – Episode Description “ Counseling” The second episode is also based on therapy, but this time for Debra and Ray. It begins with an argument when she asks him to help her clean the house before a dinner party, and he refuses to help because he "spent the whole day with the kids" and wants to relax. He is watching TV and eating CHIPS, and drinking GINGER ALE, while Debra is trying to VACUUM up the crumbs, and she eventually vacuums up the remote. At the dinner party, their friends talk about the help they've gotten from their therapist, Pamela, who has enabled them to discover the root of their arguments. Debra is fascinated, but Ray is not, and that night, they have an argument about starting therapy. Debra wins, and they make an appointment with Pamela, where Debra cries because she works hard at home and feels "unappreciated"; Ray's strategy is to agree with Debra, admitting that he's been selfish and calling her a "saint" for putting up with him. Pamela applauds his honesty, and Debra is furious because Ray tried to manipulate Pamela. When they return home, his parents are there, and they all discuss his failure to help Debra clean the house. Frank tells Ray to help out a little, and Marie says that Debra needs all the help she can get because she is a terrible housekeeper. Ray admits that he really wants to be taken care of, and agrees that he would rather have a mother than a wife; Robert laughs; Marie smiles. When Robert comes over the next day to apologize for laughing at Ray, he brings a gift: a life size cardboard figure of Marie in her wedding dress.

Note -- The season premiere of Raymond consisted of two back-to-back episodes.

32

Frasier – Series Description Characters Frasier Niles’s brother, psychiatrist, host of a radio talk show, snob Niles Frasier’s brother, psychiatrist, divorced, also a snob Martin Niles and Frasier’s father, ex-cop, disabled (walks with cane), dislikes his sons’ snobbery Daphne Martin’s caretaker, a physical therapist. Martin and Daphne live with Frasier Roz The producer of Frasier’s radio talk show, and a personal friend of Frasier and Daphne. Setting Seattle, Frasier’s apartment – very deluxe, with a view of Seattle, expensive art and furniture, full wine rack Situation Family group – Niles and Frasier are brother psychiatrists, whose snobbishness irritates their father, Martin, an ex-policeman. Daphne is Martin’s physical therapist, and they both live with Frasier. Frasier – Episode Description “ The Ring Cycle” The episode is about Niles and Daphne’s three successive weddings, complete with complications. The first takes place in Reno at the Lucky 7 chapel, where Niles has forgotten to buy a WEDDING RING. Roz, the producer of Frasier’s show, calls and Niles’s CELL PHONE rings; Niles and Daphne then realize that they can’t say that they eloped because Frasier would be hurt at not being invited. They decide to get married again the next day at City Hall, and en route to the wedding, Frasier investigates Martin’s lunch bag (prepared in Daphne’s absence), and finds CHIPS and PUDDING CUPS. When Daphne’s mother gets to City Hall, she says that no daughter of hers can get married without a minister, and she walks out. But the newlyweds get married again, and Daphne’s mother comes to Frasier’s apartment to apologize for demanding that a minister be present. Frasier brings in a bottle of WINE, and Niles and Daphne tell her that they will go back to City Hall to get married again (number 3).

33

Friends – Series Description Characters Late 20s - early 30's, white Joey Actor in "Days of Our Lives" (soap opera), Italian, a womanizer Monica Ross's sister, wealthy background, Jewish, married to Chandler, works as a chef, is food-obsessed, neurotic, and controlling Phoebe Massage therapist, new age hippie, nurturing, loves animals Rachel Monica's high school best friend, works at Ralph Lauren, on and off romance with Ross, has just had his baby Ross Monica's brother, wealthy background, Jewish, works as a paleontologist, university professor, married 3 times, one child Chandler Married to Monica, works at unknown job in finance, Ross's college room-mate Setting All characters live in NY - the West village - and hang out at "Central Perk," an upscale coffee house. Situation A "pack of pals" show, comparable to Cheers, in which 6 close friends get together at each others' apartments or at Central Perk to discuss jobs, relationships, clothing, and other issues. Friends – Episode Description “ The One Where No One Proposes” "Who-proposed-exactly-what-to-Rachel" showdown. All of the friends gather in Rachel's hospital room, filled with BALLOONS and FLOWERS, to celebrate her giving birth to Ross's baby. Joey, who is in love with her, finds an ENGAGEMENT RING that has fallen out of Ross's coat. When he bends down to retrieve it, Rachel thinks that he is proposing to her--she accepts. Joey realizes that Ross was about to propose, and Phoebe leaves the room to get her GAME BOY back from a child to whom she lent it. Monica and Chandler come to visit next, and since they are trying to have a baby, when she announces that she is ovulating they duck into a closet to have sex, unfortunately interrupted by Mr. Geller (Rachel's father), who gives them advice on the best positions for conceiving. Back in Rachel's room, Phoebe congratulates Ross on his engagement, and Ross, who is eating a SANDWICH outside, is confused; he mentions that Joey ate a piece of PLASTIC FRUIT earlier. When Ross returns to Rachel's room, he thinks that Rachel is hallucinating until Phoebe points out her RING. Joey explains that he didn't propose; tells Rachel that the ring is Ross's; Ross explains that he was not going to propose, and was carrying the ring because his mother had just given it to him. However, he tells Rachel that he wants to start dating again.

34

Will and Grace – Series Description Characters 30-something, white Will Gay lawyer, has 2 bedroom apartment on upper west side (Lincoln Towers - 155 West End Avenue), not currently in a relationship Grace Jewish, lives with Will, single interior designer who dated Will at Columbia, when both were undergraduates Leo Markus Jewish doctor, helps Grace after a fall in the park, interested in dating Grace Karen Walker Grace's receptionist, married to a wealthy man (Stan), whom we never see, talks about taking drugs and drinking, sexually ambiguous - flirts with Jack and with women Jack Younger than Will, Will's best friend since college, an actor who works in retail (Barney's and Banana Republic), fashion-conscious, and flirts with men to find dates Setting Will and Grace's apartment, nearby places (Central Park, coffee shop) Situation A "pack of pals" show with gay men and straight women forming a new family group. Characters are engaged in shifting relationships (Will and Grace; Jack and Will; Will, Grace, and Leo), with gay/straight issues as the dominant theme. Will and Grace – Episode Description “ ... And the Horse He Rode in On” Suspense about whether Will and Grace will actually have a baby together via in vitro. The episode begins in Central Park, which Grace is running through to retrieve Will's forgotten sperm sample. She runs into a lamppost, becomes unconscious, and a man on horseback--a Jewish doctor named Leo Markus--stops to help her. When Will and Grace meet at the gynecologist's office, the insemination cannot take place because Grace dropped the sperm when she fell. They return home. Karen tells Will that she was drinking WINE at a bar where she met Mr. Banks. When he invited her to his hotel room for some more WINE, they “fooled around afterwards,” and did “everything but.” When Will goes to buy groceries, he brings back TWIZZLERS as a present for Grace. Will and Grace go to a coffee shop, and Grace is eating a BLUEBERRY MUFFIN when Leo walks in by chance. She refuses to go out on a date with Leo, but when she forgets her wallet, Leo finds it and goes to Will and Grace's apartment to return it. Grace is happy because her frequent FROZEN YOGURT card is in it. Leo asks Will who he is, and when Will tells him that he is having a baby with Grace, Leo leaves. Will is jealous and angry at Grace for her supposed violation of the no dating promise, but admits that he kissed a guy at the Gay Gay Gay Gay Club. When Grace looks through her wallet afterwards, she finds that Leo has slipped in his card, and she decides to keep it.

NOTE -- website: www.tvguide.com/showguide/episodeguide

35 APPENDIX 2: PARASOCIAL INTERACTION SCALE

I think _____ is like an old friend; ______makes me feel comfortable, as if I am with friends; ______seems to understand the things I want to know; I find ______to be attractive; I would like to meet ____ in person; I feel sorry when _____ makes a mistake; I like to compare my ideas with what _____ says; When I'm watching the program ____ is on, I feel as if I am part of the group; I miss seeing ____ when his or her program is not on.

36 REFERENCES

Abrams, Morris. H. (1993), A Glossary of Literary Terms , 6th Ed., New York: Harcourt Brace.

Arango, Tim (2001), “Videos Unplugged: MTV Cracking Down On Product Placement,” New

York Post (December 18, 2002), 31.

Balasubramanian, Siva K. (1994) "Beyond Advertising And Publicity: Hybrid Messages And

Public Policy Issues," Journal Of Advertising, 23 (4): 29-46.

Barker, David (1985), "Television Production Techniques as Communication," Critical Studies

in Mass Communication , 2 (June), 234-246.

Bergson, Henri (1900, repr. 1956), "Laughter," in Comedy , Wylie Sypher, intro. and appendix,

Garden City, NY: Doubleday and Company, Inc., 61-146.

Brown, Stephen P. and Douglas M. Stayman (1992) “Antecedents and Consequences of Attitude

toward the Ad: A Meta-analysis,” Journal of Consumer Research 19 (June): 34-51.

Bruner and Hensel 1996

Butler, Robert W. (2001) “Calling on Hollywood Sprint rep keeps phones busy competing for

product placement in movies” in The Kansas City Star , 01/13/2001: E1.

Calder, Bobby J. and Alice M. Tybout (1987), "What Consumer Research is...," Journal of

Consumer Research , 14 (June), 136-140. Campbell, Donald T. and Julian C. Stanley

(1966), Experimental And Quasi-Experimental Designs for Research , Chicago: Rand

McNally.

Carter, Reon (2000), “PJs Have Their Days; No Longer Just A Sleepy Fashion, Cozy Pajamas

Get A Leg Up On Comfy Competitors,” The Cincinnati Enquirer , (January 13), E01.

Carter, Bill (2001), “New Reality Show Planning To Put Ads Between the Ads,” The New York

Times (April 30), A1, A17.

37 ______(2003), “Reality Shows Alter Way TV Does Business,” The New York Times

(January 25), C1, C14.

Chang, Jennifer E. and W. Edward Roth (2000), "When is Cranberry Sauce Shaped Like a Can?

An Investigation of Cultural Capital, Gender and Consumption in Television

Programming," in Gender, Marketing, and Consumer Behavior: Fifth Conference

Proceedings , J. Schroeder and C. Otnes (eds.), 107-123.

Consoli, John (2004), "Running in Place(ment)," Brandweek, 45 (28), 4. 6.

DeLorme, Denise E. and Leonard N. Reid (1999) “Moviegoers' Experiences and Interpretations

Of Brands In Films Revisited,” Journal Of Advertising, XXVIII (2): 71-96.

Davila, Arlene (1998), “El Kiosko Budweiser: The Making of A "National" Television Show in

Puerto Rico,” American Ethnologist , 25(3), 452-470.

Earnest, Leslie (2000), “Oakley Proves Sale Mission Possible,” Los Angeles Times , 07-20-00:

C10

Esslin, Martin (1976), An Anatomy of Drama , London: Temple Smith. von Feilitzen, Cecilia and Olga Linné (1975), “The Effects Of Television On Children And

Adolescents Identifying With Television Characters,” Journal of Communication , 4, 31-

55.

Fergusson, Francis (1961), Aristotle's Poetics , S. H. Butcher, transl. New York: Hill and Wang.

Feuer, Jane (1992), "Genre Study and Television," in Channels of Discourse, Reassembled , 2nd

ed., ed. Robert C. Allen, Chapel Hill: The University of North Carolina Press, 138-160.

Fiske, John (1987), Television Culture , London: Routledge.

Forster, E.M. (1954), Aspects of the Novel , New York: Harcourt, Brace and World, Inc.

38 Fowler, Alastair (1982), An Introduction to the Theory of Genres and Modes , Cambridge, MA:

Harvard University Press.

Frye, Northrop (1973 [1957]), Anatomy of Criticism: Four Essays , Princeton: Princeton

University Press.

Feilitzen and Linne 1975Gerbner, George (1995). Alcohol in American Culture. In S. E. Martin

(ed.), The Effects Of The Mass Media On Use And Abuse Of Alcohol (pp. 3-29).

Bethesda, MD: National Institute on Alcohol Abuse and Alcoholism.

Gould, Stephen J., Pola B. Gupta, and Sonja Grabner-Kräuter (2000), “Product Placements in

Movies: A Cross-Cultural Analysis of Austrian, French and American Consumers’

Attitudes Toward This Emerging, International Promotional Medium,” Journal of

Advertising , 29 (4), 41-58.

Jenkins 1988

Heider, E (1946). Attitudes and cognitive organization. Journal of Psychology, 21, 107-112. Heider, Fritz (1958), The Psychology of Interpersonal Relations, New York: John Wiley & Sons.

Hirschman, Elizabeth C., Linda Scott, and William B. Wells (1998), "A Model of Product

Discourse: Linking Consumer Practice to Cultural Texts," Journal of Advertising , 27

(Spring), 33-50.

______. and Craig J. Thompson (1997), “Why Media Matter: Toward a Richer

Understanding of Consumers' Relationship with Advertising and Mass Media,” Journal

of Advertising , 26 (Spring), 43-60.

______. (1988), “The Ideology of Consumption; A Structural-Syntactical Analysis of

“Dallas” and Dynasty”,” Journal of Consumer Research , 15 (December), 344-359.

39 Holbrook, Morris B. and Mark W. Grayson (1986), “The Semiology of Cinematic Consumption:

Symbolic Consumer Behavior in Out of Africa,” Journal of Consumer Research , 13

(December), 374-381.

Horton, Donald and R. Richard Whol (1956), “Mass Communication and Para-Social

Interaction,” Psychiatry , 19 (August), 215-229.

Hundley, Heather. L. (1995), “The Naturalization of Beer in Cheers,” Journal of Broadcasting

and Electronic Media , 39, 350-359.

Jurith, Edward, H. (2001), “National Youth Anti-Drug Media Campaign: How to Ensure the

Program Operates Efficiently and Effectively,” Office of National Drug Control Policy,

http://www.whitehousedrugpolicy.gov/news/testimony01/080101/content.html , August 1,

2001

Kozinets, Robert V. (2001), “Utopian Enterprise: Articulating the Meanings of Star Trek’s

Culture of Consumption,” Journal of Consumer Research, 28 (June), 67-88.

Kuhn, Thomas S. (1962), The Structure of Scientific Revolutions , Chicago: University of

Chicago Press.

Laudan, Larry (1984), Science and Values , Berkeley: University of California Press.

Lavin, Marilyn (1995), “Creating Consumers in the 1930s: Irna Phillips and the Radio Soap

Opera," Journal of Consumer Research , XXII (June), 75-89.

Leet, deBecker, and Giles 1995

Levy, Sidney (1962), “Phases in Changing Interpersonal Relations,” Merrill-Palmer Quarterly of

Behavior and Development , 8 (2), 121-128.

40 Marshall, Norm and Dean Ayers (1998), “Product Placement Worth More Than Its Weight,”

Brandweek , 02/09/98, 39 (6): 16-17.

______(2001), “ How Product Placement And Entertainment Promotions Make Sense For

Your Business ,” Retrieved January 15, 2001, from the World Wide Web:

http://www.normmarshall.com/home.html .

Maccoby, Eleanor E. and William Cody Wilson (1957), “Identification and Observational

Learning From Films,” Journal of Abnormal and Social Psychology, 55 (1), 76-87.

McNatt, Robert and Oleck, Joan (2000), "The Plot Thickens, Like Heinz Ketchup," Business

Week, 11/06/00, 3706: 14

Neale, Steve and Frank Krutnik (1990), Popular Film and Television Comedy , London:

Routledge.

Newton, Barbara J. and Elizabeth B. Buck (1985), “Television as a Significant Other,” Journal

of Cross-Cultural Psychology , 16 (3), 289-311.

Pfister, Manfred (1991), The Theory and Analysis of Drama , Cambridge, ENG: Cambridge

University Press.

Rogers, T. B., Kuipers, N. A., and Kirker, W. S. (1977), “Self-Reference And The Encoding Of

Personal Information,” Journal of Personality and Social Psychology, 35 , 677-688.

Rosenblatt, Louise (1979 [1937]), The Reader, The Text, The Poem: The Transactional Theory

of the Literary Work , Carbondale, IL: Southern Illinois University Press.

Rubin, Alan M., Elizabeth M. Perse, and R. Powell (1985), “Loneliness, Parasocial Interaction,

and Local Television News Viewing,” Human Communication Research , 12, 155-180.

______and Elizabeth M. Perse (1987), “Audience Activity and Soap Opera Involvement,”

Human Communication Research , 14, 246-268.

41 Russell, Cristel A. (2002), “Investigating The Effectiveness Of Product Placements In Television

Shows: The Role Of Modality And Plot Connection Congruence On Brand Memory And

Attitude,” Journal of Consumer Research, 29 (December), 306-318.

______Andrew T. Norman, and Susan E. Heckler (2004), “The Consumption of Television

Programming: Development and Validation of the Connectedness Scale,” Journal of

Consumer Research , 31 (2).

Schlosser, Joe (2002), “Where the Product is Co-Star,” Broadcasting and Cable, 132 (8), 24.

Sherry, John, F. (1995), “Bottomless Cup, Plug-In Drug: A Telethnography of Coffee,” Visual

Anthropology , 7, 351-370.

Solomon, Michael R. and Lawrence Greenberg (1993), "Setting the Stage: Collective Selection

in the Stylistic Context of Commercials," Journal of Advertising , 22 (March), 11-24.

______and Basil G. Englis (1994), "Reality Engineering: Blurring the Boundaries Between

Commercial Signification and Popular Culture," Journal of Current Issues and Research

in Advertising , 16 (2), 1-17.

Stanley, T. L. (1998), "Wanted: Ally McBeal," Brandweek , 08/10/98, 39 (31), 32-35.

Stern, Barbara B. (1994), "Classical and Vignette Television Advertising Dramas: Structural

Models, Formal Analysis, and Consumer Effects," Journal of Consumer Research , 20

(March), 601-615.

______(1993), "Feminist Literary Criticism and the Deconstruction of Ads: A Postmodern

View of Advertising and Consumer Responses," Journal of Consumer Research , 19

(March), 556-566.

42 ______(1991), "Literary Analysis of an Advertisement: The Commercial as Soap Opera,"

Advances in Consumer Research , Vol. 18, eds. Rebecca H. Holman and Michael

Solomon, Provo, UT: Association for Consumer Research, 164-171.

Thompson, Robert (2001), “Reality and the Future of Television,” Television Quarterly , 31

(Winter), 20-25.

Tueth, M. (2000), “Fun City: TV's Urban Situation Comedies Of The 1990s,” Journal of

Popular Film and Television , 28 (3), 98-107.

Tulloch, John C. and Henry Jenkins (1995), Science Fiction Audiences: Watching Doctor Who

and Star Trek , London: Routledge.

Wells, William, D. (2001), “The Perils of N = 1,” Journal of Consumer Research , 28

(December), 494-498.

Wernick, Andrew (1991), Promotional Culture: Advertising, Ideology And Symbolic Expression ,

London: Sage.

Wolff, Jurgen (1988), Successful Sitcom Writing , New York: St. Martin's Press.

43