Florida State University Libraries

Electronic Theses, Treatises and Dissertations The Graduate School

2014 Examining the Impact of Exposure to a Bipolar Disorder Storyline Using the Entertainment Overcoming Resistance Model Danyang Zhao

Follow this and additional works at the FSU Digital Library. For more information, please contact [email protected]

FLORIDA STATE UNIVERSITY

COLLEGE OF COMMUNICATION AND INFORMATION

EXAMINING THE IMPACT OF EXPOSURE TO A BIPOLAR DISORDER STORYLINE

USING THE ENTERTAINMENT OVERCOMING RESISTANCE MODEL

By

DANYANG ZHAO

A Thesis submitted to the School of Communication in partial fulfillment of the requirements for the degree of Master of Arts

Degree Awarded: Spring Semester, 2014

Danyang Zhao defended this thesis on January 16, 2014.

The members of the supervisory committee were:

Laura M. Arpan

Professor Directing Thesis

Arthur A. Raney

Committee Member

Juliann Cortese

Committee Member

The Graduate School has verified and approved the above-named committee members, and certifies that the thesis has been approved in accordance with university requirements.

ii

ACKNOWLEDGEMENT

I would like to express my sincere gratitude to my committee chair Dr. Laura Arpan for her patient guidance and kind encouragement throughout my thesis, which greatly improved the quality of my work. I’m also grateful to my other committee members, Dr. Arthur Raney and Dr.

Juliann Cortese, for their insight and suggestions on this project. I want to thank Dr. Raney for making the thesis defense a challenging, rewarding and enjoyable experience. Also, I would like to express my appreciation to Sophie Janicke, Paul Han and other fellow graduate students and communication researchers for their support. I want to extend my gratitude to the School of

Communication for providing a great educational experience through my two years at FSU. Last but not least, a very special thank you to my wonderful parents, Huichun Zhao and Juan Wu, for their love and support.

iii

TABLE OF CONTENTS

List of Tables ...... vi List of Figures ...... vii Abstract ...... viii 1. INTRODUCTION ...... 1 2. LITERATURE REVIEW ...... 6 2.1 Entertainment Education ...... 6 2.2 Narrative Persuasion ...... 8 2.3 Entertainment Overcoming Resistance Model ...... 9 2.4 Hypotheses ...... 20 3. METHODOLOGY ...... 28 3.1 Participants ...... 28 3.2 Study Design ...... 28 3.3 Stimulus Material ...... 29 3.4 Procedures ...... 30 3.5 Measures ...... 31 3.5.1 Perceived Persuasive Intent ...... 32 3.5.2 Transportation ...... 32 3.5.3 Identification ...... 33 3.5.4 Perceived Similarity ...... 33 3.5.5 Parasocial Interaction ...... 34 3.5.6 Perceived Invulnerability ...... 34 3.5.7 Counterarguing ...... 35 3.5.8 Reactance ...... 36 3.5.9 Pro-treatment Behavioral Intention ...... 37 3.5.10 Information-Seeking Behavior Regarding BPI at Follow-Up ...... 38 3.5.11 Control Variables ...... 39 3.5.11.1 Past BPI-related behaviors ...... 39 3.5.11.2 BPI-related experience ...... 39 3.5.11.3 Demographic information ...... 39 3.5.11.4 Perceived ages of characters ...... 39 3.5.11.5 Enjoyment ...... 39 4. RESULTS ...... 41 4.1 Preliminary Analyses ...... 41 4.1.1 Random Assignment Check Verifying Equivalence of BPI-related ...... Experience and Past Behaviors across Groups ...... 41 4.1.2 Enjoyment of Narrative Stimuli ...... 42 4.1.3 Tests for Gender Effects in Narrative Involvement ...... 42

iv

4.1.4 Summary of All Key Variables Measured ...... 43 4.2 Hypotheses Testing ...... 45 4.2.1 Comparing Message Effectiveness of Stimulus Programs ...... 45 4.2.2 Comparing Perceived Persuasive Intent ...... 47 4.2.3 Comparing Resistance between Two Message Conditions ...... 47 4.2.4 Examining the Relationship between Perceived Persuasive Intent ...... and Reactance ...... 49 4.2.5 Examining Narrative Involvement-Resistance Relationships in the BPI E-E ...... Narrative Condition ...... 50 4.2.6 Examining Narrative Resistance-Behavior Associations in the BPI E-E ...... Narrative Condition ...... 53 5. DISCUSSION ...... 57 5.1 Limitation ...... 65 APPENDICES ...... 67 A. IRB APPROVAL MEMORANDUM AND CONSENT FORM ...... 67 B. INSTRUCTIONS ...... 71 C. MEASURES ...... 72 REFERENCES ...... 99 BIOGRAPHICAL SKETCH ...... 107

v

LIST OF TABLES

1 Means and Standard Deviations for Variables Measured Across All Three Conditions ...... 43 2 Means and Standard Deviations for Variables Measured in BPI E-E Narrative Condition ...44 3 Means, Standard Deviations for BPI Pro-treatment Behavioral Intentions and BPI ...... Information Seeking Behaviors by Condition ...... 44 4 Summary of Hypotheses and Results ...... 48 5 Summary of Bivariate Correlations in E-E BPI Narrative Condition ...... 50

vi

LIST OF FIGURES

1 Hypotheses 7 – 12 developed based on the EORM ...... 26 2 Hypotheses 13 – 15 created based on the EORM ...... 27 3 Involvement-Resistance Associations ...... 51 4 Resistance-Behavioral Outcome Associations ...... 53

vii

ABSTRACT

The effects of entertainment-education narratives on pro-social, story-consistent beliefs

and behaviors have been discussed by many. Less research has looked into the underlying

mechanisms that facilitate these effects. Past research have attempted to explain the process through which persuasion occurs in narrative using a variety of theories, models and constructs.

Recently, a better-integrated and structured theoretical framework — the Entertainment

Overcoming Resistance Model — has been proposed to account for the persuasive effects. It clarifies the role of some similar constructs in narrative message processing. The present study examined several hypothesized relationships in the EORM by investigating the effects of

exposure to a major bipolar disorder storyline embedded in a popular serial teen drama — 90210

— compared to those of exposure to a PSA addressing bipolar I disorder. The role of

identification, transportation, parasocial interaction, perceived similarity, counterarguing,

reactance, and perceived invulnerability in producing persuasive outcomes were re-examined in

this bipolar I disorder E-E narrative. Participants were randomly assigned to one of three viewing

conditions, a Stormwater Runoff Public Service Announcement plus a video of 90210 excerpts

containing bipolar disorder storyline, a bipolar disorder PSA paired with a video of 90210

excerpts without bipolar disorder storyline, and the Stormwater Runoff PSA plus the video of

90210 excerpts without bipolar disorder storyline (control condition). They completed a questionnaire immediately after viewing each video and filled out another online follow-up

questionnaire two weeks later to assess their responses to the program. Results of the study

indicated the following: (1) Relative to the control condition, BPI E-E narrative exposure was

not associated with greater overall pro-treatment behavioral intention and behaviors regarding

BPI but did enhance the behavioral intention to seek help from mental health professional

viii

regarding BPI; (2) among the three resistance-related variables examined in this study, only

perceived invulnerability was negatively associated with behavioral intention; (3) identification with characters was the only construct that contributed to reduced perceived invulnerability; and

(4) the additional consideration of perceived invulnerability of “others” seemed helpful in understanding the effects of the bipolar I disorder narrative. Implications of findings, suggested

areas of future research, and limitations of the present study are discussed.

ix

CHAPTER ONE

INTRODUCTION

The debate of whether television has positive or negative effects on individuals and society has been going on for decades. Just as any other media form, television’s influence on humans and society is complicated and multifaceted. As suggested by social cognitive theory, humans are endowed with the abilities to gain knowledge not only from direct experience but also through observational learning, or vicarious experiences (Bandura, 2002). It is believed that if we can learn by observing others’ behaviors in the real world, we can also learn the behaviors

modeled by TV characters.

Television has been claimed to have potentially contributed to public stigmatization of mental disorders through stereotyping illnesses in various shows as well as influencing its viewers by way of cultivation, as a cumulative effect of the inaccurate depiction of mental illnesses in these shows over time (Diefenbach & West, 2007; Wahl, Hanrahan, Karl, Lasher, &

Swaye, 2007). Several content analyses have found inaccurate media portrayals of mental illness and frequent depictions associating mental illness with violence, aggression, suicide, rejection and isolation (Hoffner & Cohen, 2012; Pirkis, Blood, Francis, & McCallum, 2006; Wahl et al,

2007). These depictions may lead to viewers’ negative perceptions of mental disorders and of those living with mental illnesses.

However, it is important to note that television can also have an important, positive effect on the public by communicating significant social and health issues to foster pro-social attitudes and behaviors through storytelling (Kaiser Family Foundation, 2004; Piotrow & de Fossard,

2004; Singhal & Rogers, 2004). Being an important venue for Entertainment Education — a pro-

social communication strategy to entertain and to educate the audience by purposefully

1

incorporating educational message into media contents (Ritterfeld & Jin, 2006) — television can potentially serve as a useful tool to reduce stigma associated with mental illness and disseminate important mental health information. With the potential to impact millions of viewers on a daily basis and a capacity to carry an extremely powerful form of E-E— serial dramas — television is considered the primary medium for E-E interventions in the US and worldwide (Piotrow & de

Fossard, 2004).

Entertainment-education (E-E) as a strategy to bring about behavioral and social change has a history of more than six decades (Poindexter, 2004). It has been used to address various public health issues and social issues overseas as well as in the U.S. Studies suggest that the underlying pro-social messages of these E-E interventions can increase viewers’ awareness of the issues being addressed and influence their attitudes and behaviors related to these issues

(Moyer‐Gusé, 2008; Palmer, 2002).

Notwithstanding, rarely have E-E efforts been made to address mental health issues.

Mental disorders are common in the United States: the estimate of the adult population that

meets the criteria for one or more mental disorders is 26.2% in a given 12 month period, which is

approximately a quarter of U.S. adult population (Kessler, Chiu, Demler, & Walters, 2005;

NIMH, 2008). Among these diagnosed cases, 22.3% are classified as “severe” (Kessler et al,

2005). Mental disorders are one of the major predictors of suicide. Disorders characterized by anxiety and poor impulse-control predict which people with suicide ideation act on such thoughts

(Nock et al, 2009). Particularly, bipolar disorder (BPD) is one of the mental disorders that has been found to be a leading cause of premature mortality due to suicide and associated medical conditions (Kupfer, 2005; Tondo, Isacsson, & Baldessarini, 2003; Osby, Brandt, Correia,

Ekbom, & Sparen, 2001). Bipolar disorder affects approximately 2.6% of the American adult

2

population, but this number may underrepresent the prevalence because of mental illness sufferers’ possible reluctance to participate in mental health survey and the lack of participation from those hard-to-reach segments of population (Kessler et al, 2005). Bipolar disorder often develops in a person’s late teens or early adult years (NIMH, 2008). The average age at onset of the first episode of bipolar disorder is 18.2 to 20.3 years (Merikangas et al, 2007).

Generally, mental disorders are “characterized by abnormalities in cognition, emotion or mood, or the highest integrative aspects of behavior, such as social interactions or planning of future activities. These mental functions are all mediated by the brain.” (Satcher, 2000, p. 5). The public is less knowledgeable about mental illness than physical illness (Jorm, 2012). According to the World Health Organization’s World Mental Health Survey Initiative research report, only a minority of people with mental disorders received treatment for mental disorders in the year of onset, and for those who received treatment subsequently, the median delays in treatment seeking ranged from 1 to 14 years for mood disorders and from 3 to 30 years for anxiety disorders (Jorm, 2012; Wang et al., 2007). In the United States, the median duration of delay in treatment seeking for mood disorder after the first onset is four years (Wang et al., 2007). Past research on mental disorders suggests that the outcomes of treatment worsen with the length of the duration of untreated illness (Altamura et al, 2010; Jorm, 2012). For example, a study found that bipolar patients with a longer duration of untreated illness showed a higher number of suicide attempts and a higher frequency of depressive onset (Altamura et al, 2010). It is believed that lack of recognition of mental disorders is one of the factors that causes the delay in getting professional help. Another reason for the delay is that the first onset of mental disorders often occurs during adolescence or young adulthood—a period of life when people’s knowledge and experience are underdeveloped. As illustrated by a U.S. national survey, young people of 15 to

3

24 years old, particularly males, had less positive attitudes toward seeking help from mental health professionals than older age groups (Gonzalez, Alegria, & Prihoda, 2005). The author

believes that with foreseeable challenges, entertainment education strategy can help

communicate significant messages about mental illness to various age groups. E-E has the

potential to increase awareness of particular mental disorders and mental health issues in general and to facilitate adaptive and pro-health attitude and behavior change in the direction of the recommendation offered in the E-E narratives.

Recently, the results of a study examining the effects of viewing an accurate and empathetic movie portrayal of schizophrenia suggested that exposure increased knowledge about this mental health issue and contributed to attitude change and stigma reduction when combined with an educational trailer about the mental disorder after the movie exposure (Ritterfeld & Jin,

2006). Yet, the findings suggested that mere exposure to accurate portrayal of the mental disorder in the movie did not contribute to stigma reduction associated with schizophrenia, but

“resulted in an increase of negative attitudes toward mental illness” (Ritterfeld & Jin, 2006). This

boomerang effect among some individuals was attributed to the disturbing and conflicting ideas

produced by the dramatic complexity of this movie. The stimulus used in this study was a movie

neither originally designed to serve a part of an E-E strategy nor created with an intention to

communicate mental health information. It seems reasonable to conclude that addressing mental

health issues through entertainment narratives demands purposeful message design. Aside from

the necessary knowledge of the mental disorder, fruitful entertainment education follows a

thorough understanding of narrative information processing mechanisms and a purposeful

message designing along with collaborative efforts across important groups — from

governmental organizations, non-governmental organizations, communication researchers,

4

health professionals, to television scriptwriters, producers, and executives (Beck, 2004; Nahm et al, 2010).

A more recent, U.S. based, E-E intervention focused on bipolar disorder among youth has generated some desirable outcomes. Recognizing that the peak onset of bipolar disorder is during adolescence and young adulthood, the CW network and the writers and producers of the teen drama series 90210 collaborated with Hollywood, Health and Services (HH&S) to create a major

BPD storyline that was stretched over roughly 6 episodes including a pivotal episode in which a main character experiences bipolar I disorder. Episodes following the pivotal plot depict the

bipolar character as someone able to manage the mental illness (Nahm et al, 2010). This was not only a multi-media outreach but also a communication campaign combining a narrative and a

PSA with multiple access points for information seeking involving mental disorder. As a result

of this intervention, there was a dramatic increase in web traffic to bipolar disorder

organizations’ websites, which represents an increase in information seeking behavior after

exposure (Nahm et al, 2010).

The current study examined how the underlying messages embedded in the bipolar

storyline of this particular teen drama — 90210 —influenced viewers’ behavior towards bipolar

disorder using some of the constructs and associations suggested by the Entertainment

Overcoming Resistance Model (Moyer‐Gusé, 2008; Moyer‐Gusé & Nabi, 2010). Most

importantly, the study examined whether and how a televised narrative addressing bipolar

disorder contributes to positive behavioral intentions, and pro-treatment behaviors regarding bipolar I disorder. Since the focus of this study was to explore the independent persuasive effect of the narrative — bipolar disorder storyline in the television teen drama 90210 — the epilogue originally added to the end of the major bipolar disorder episode — “off the rails” was excluded.

5

CHAPTER TWO

LITERATURE REVIEW

2.1 Entertainment Education

Entertainment-education has been defined in different ways. A traditional definition

views E-E as a way of entertaining and educating audience by purposely designing and

incorporating an educational message into media content, in order to increase audience

members’ knowledge about an issue, create favorable attitudes, shift social norms, and ultimately

change their behavior (Singhal & Rogers, 2004, p. 5). A contrasting view defines E-E as a programming that contains some prosocial content, regardless of the purpose for which it was created (Greenberg, Salmon, Patel, Beck, & Cole, 2004). The current study uses the traditional definition of entertainment-education, which differs from other explanations in that it highlights a purposeful message designing process with certain communication goals in mind.

Entertainment-education can influence individuals at cognitive, attitudinal, and behavioral levels

toward a socially desirable end. For example, the Kaiser Family Foundation has worked with

many TV networks and producers to incorporate health messages into popular TV dramas, such

as NBC TV drama “ER”. They found, after a survey of 3,500 viewers, that there was a

significant increase in awareness about the health issues addressed on the show. Nearly 23%

viewers reported that they had sought additional information about a health issue by visiting

other sources (Kaiser Family Foundation, 2004). At the same time, entertainment-education can

also “influence audience’s external environment to help create the necessary conditions for social

change at the system level” (Singhal & Rogers, 2004). In order to achieve desired effects, E-E

narratives should be compelling enough to disguise the persuasive intent and engage the viewers,

yet educational enough to clearly communicate the advocated position (Busselle & Bilandzic,

6

2013). Recently, an increasing number of organizations, agencies and industries worldwide have

become involved in creating E-E interventions for various health and social issues (Kaiser

Family Foundation, 2004; Singhal & Rogers, 2004).

A growing body of research has looked into E-E television programs as interventions

across a variety of topics related to social and public health issues from drunk driving (Moyer‐

Gusé, Jain and Chung, 2012; Winsten & DeJong, 2001), to unplanned teen pregnancy (Moyer‐

Gusé & Nabi, 2010), to organ donation (Bae & Kang, 2008; Movius, Cody, Huang, Berkowitz,

& Morgan, 2007), to teen obesity (Valente et al, 2007), to emergency contraception and HPV

(Brodie et al, 2001), to breast cancer (Hether, Huang, Beck, Murphy, & Valente, 2008; Wilkin et al, 2007), to HIV/AIDS prevention (Piotrow & de Fossard, 2004; Vaughan, Rogers, Singhal, &

Swalehe, 2000). In terms of the genres of programs into which health-related storylines were embedded, studies have examined medical dramas such as ER (Brodie et al, 2001; Valente et al,

2007; Whittier, Kennedy, Lawrence, Seeley, & Beck, 2005) and Grey’s Anatomy (Hether et al,

2008; Movius, 2007); nonmedical dramas like Numb3rs (Movius, 2007) and The OC (Moyer‐

Gusé & Nabi, 2010) as well as soap operas, such as The Young & the Restless (Beck, 2004); and

situation comedies such as Scrubs (Moyer-Gusé, Mahood & Brookes, 2011) to examine the association between exposure to health information and changes in topic-related knowledge,

attitudes, and behaviors.

Generally, these studies indicate the effectiveness of entertainment-education in

communicating health messages. However, a significant domain of public health — mental

health — has rarely been the focus of E-E efforts. To date, no known study has investigated the

influence of narrative persuasion brought about by EE television content related to mental health

issues. Therefore, the current study takes the first step to explore the possible persuasive

7

outcomes of an E-E television drama storyline featuring a mental health issue — bipolar disorder. In line with previous studies, this E-E narrative was created based on the assumptions

of established theories and models.

2.2 Narrative Persuasion

As demonstrated by research on E-E, the power of entertainment education comes from its unique way of delivering messages via a subtle integration of a persuasive content and an engaging narrative. The form of persuasion involved in E-E is often referred to as narrative

persuasion. Bilandzic and Busselle (2013) defined narrative persuasion as “any influence on

beliefs, attitudes, or actions brought about by a narrative message through processes associated

with narrative comprehension or engagement” (p. 201). Due to the narrative structure of E-E

messages, traditional models of persuasive information processing, such as the Elaboration

Likelihood Model (Petty & Cacioppo, 1981; Petty & Wegener, 1999) and the Heuristic-

Systematic Model (Chen & Chaiken, 1999) cannot fully capture the influence of E-E narrative

messages on viewers’ attitudes and behaviors. Rather, it is hypothesized that the underlying

persuasive message of narrative content influences viewers through a phenomena called

narrative involvement, which is a process of being immersed into the story world with intense

focus on the unfolding narrative (Green & Brock, 2000; Moyer‐Gusé, 2008). The narrative

experience of involvement has been described in different terms in the literature, such as

transportation (Gerrig, 1993; Green & Brock, 2000, 2002), flow (Csikszentmihalyi, 1990, 1997),

narrative engagement (Busselle & Bilandzic, 2008, 2009; Strange & Leung, 1999), and

absorption (Graesser, 1981; Slater & Rouner, 2002). It is believed that the narrative structure of

E-E messages can reduce counterargument, reactance, and resistance in recipients because it

makes the persuasive intent less overt, and, at the same time, it has the ability to engage the

8

audience with the story (Busselle & Bilandzic, 2013; Moyer‐Gusé, 2008; Moyer‐Gusé & Nabi,

2010). Reactance is identified as negative affect in response to a perceived threat to freedom — the perceived persuasive attempt—and a motivation to restore that freedom (Busselle &

Bilandzic, 2013; Knowles & Linn, 2004; Quick, Shen, & Dillard, 2013). Resistance is a more general motivational state associated with a desire to combat pressures to change one’s attitude

(Busselle & Bilandzic, 2013; Knowles & Linn, 2004). A wide range of theoretical constructs have been proposed to explain the process through which narrative persuasion facilitates attitude

and behavior change, including narrative-structure-related constructs such as identification, parasocial interaction, perceived similarity, liking, as well as viewer-reaction-related concepts such as reactance, counterarguing, perceived invulnerability, self-efficacy, outcome expectancies, and so forth. Recently, a better integrated theoretical framework — the

Entertainment Overcoming Resistance Model (EORM, Moyer‐Gusé, 2008; Moyer‐Gusé & Nabi,

2010) — has been developed to more fully explain E-E information processing. The model

proposes a set of relationships among relevant constructs and distinguishes among similar

concepts.

2.3 Entertainment Overcoming Resistance Model

In general, EORM predicts that the unique features of entertainment narratives can

encourage engagement with the narrative world and facilitate involvement with characters,

which, in turn suppresses various forms of resistance to attitude change in the direction of the

message recommendation and, ultimately, leads to story-consistent attitudes and behaviors

(Moyer‐Gusé, 2008). The overall theoretical framework of the EORM was developed based on

the relationships predicted by the E-ELM and SCT that discussed above and seven new

propositions below (for review, see Moyer‐Gusé, 2008):

9

“Proposition 1: The narrative structure of entertainment-education messages will overcome reactance by diminishing the viewer’s perception that the message is intended to persuade.” (p.

415)

“Proposition 2: PSI [Parasocial Interaction] with a central character can enhance the persuasive effects of entertainment-education content by reducing reactance.” (p. 416)

“Proposition 2b: Liking a central character can enhance the persuasive effects of entertainment-

education content by reducing reactance.” (p. 416)

“Proposition 3: PSI [Parasocial Interaction] with characters will enhance the persuasive effects

of embedded messages in entertainment content by reducing counterarguing.” (p. 416)

“Proposition 4: Identification with a character in an entertainment-education program will

overcome the selective avoidance due to inertia” (p. 417)

“Proposition 5: The enjoyment associated with viewing an entertainment-education narrative

will overcome selective avoidance due to fear.” (p. 418)

“Proposition 6: Perceived similarity and identification with a vulnerable character will enhance

the persuasive effects of entertainment-education content by increasing a viewer’s perceived

vulnerability.” (p. 419)

“Proposition 7: PSI [Parasocial Interaction] with a character displaying counternormative

behavior will enhance the persuasive effects of entertainment education content by changing

viewers’ perceived norms.” (p. 420)

The EORM mainly builds on the extended elaboration likelihood model (E-ELM) and social cognitive theory (SCT), with each addressing a different aspect of the narrative message processing - suppressing resistance to persuasion and motivating attitude and behavior change

(Moyer‐Gusé, 2008). SCT holds that people can learn through vicarious experience and from

10

“the extensive modeling in the symbolic environment of the mass media” (Bandura, 2001, p.

271). It also predicts under which circumstances people will perform the modeled behavior via

the motivational processes of observational learning (Bandura, 2001). Specifically, people are

motivated to follow successful actions carried out by others with whom they share similarities,

but are dissuaded from imitating the modeled behaviors that, as they perceive, have undesirable

consequences (Bandura, 2001, p. 274). In line with this notion, EORM (Moyer‐Gusé, 2008) holds that whether people are motivated to take the action they learned vicariously is affected by

outcome expectancies and self-efficacy which can be influenced by perceived similarity with an

efficacious character. Outcome expectancies refer to an individual’s perception of the positive or

negative consequences of a behavior (Bandura, 2004; Moyer‐Gusé, 2008). Self-efficacy refers to

an individual’s perception of his or her capability to produce desired effects by enacting certain

behavior (Bandura, 1995, 2004). In a televised narrative, perceived self-efficacy can be enhanced

by a character perceived as similar to viewers, a character with whom viewers have parasocial

relationship or identify with, and/or a character who possesses one or more of these features

successfully performing a rewarded behavior.

The other theoretical framework, E-ELM, highlights the unique power of narratives to

overcome resistance to persuasive messages. According to E-ELM, E-E narratives affect attitude

and behavior by suspending disbelief and minimizing counterarguing through transportation into

the narrative and identification with characters (Slater & Rouner, 2002). Counterarguing here

refers to the development of thoughts that refute the persuasive argument contained in the

message (Slater & Rouner, 2002). Simply put, viewers are less likely to counterargue with the

underlying message in the narrative if they are immersed into the story world and involved with

characters in the story. The more they are engaged with the narrative, the less motivated and less

11

able (because of increased demands on cognitive capacity) they are to critically evaluate the

underlying message (Green & Brook, 2000; Knowles & Linn, 2004; Moyer‐Gusé, 2008),

therefore, the weaker the potential for resistance to the message. The two forms of narrative engagement: transportation and identification with character play the role of suppressing a form of resistance (to the advocated position), counterarguing, when the underlying message tends to

be counterattitudinal.

Transportation. Transportation into the narrative world is described as a “convergent

process, where all mental systems and capacities become focused on events occurring in the

narrative” (Green & Brock, 2000, p. 701). Transportation is “a distinct mental process, an integrative melding of attention, imagery, and feelings” (Green & Brock, 2000, p. 701). It is explained that transportation facilitates attitude and behavior change through reducing counterarguing, fostering reality-like experiences, and facilitating strong feelings toward characters in the story (Green & Brock, 2000; Green & Clark, 2012). Numerous studies have supported the role of transportation in facilitating story-consistent persuasive outcomes (Appel &

Richter, 2010; Escalas, 2007; Green & Clark, 2012; Mazzocco, Green, Sasota, & Jones, 2010).

As predicted by E-ELM, transportation may reduce counterarguing. This proposition can also be explained by the limited capacity model of mediated information processing (Lang, 2000). In the state of transportation, more processing resources are allocated to keeping up with the unfolding storyline and to being concerned about the future of the character whose goals the viewer internalizes. As a result, fewer resources are allocated to critically evaluating the underlying message. Thus, the viewer is less able to counterargue with the advocated position. At the same time, the more viewers want to engage with the narrative, the less motivated they are to counterargue the underlying message, as the generation of rebuttals will interrupt the state of

12

engagement (Green & Brook, 2000; Knowles & Linn, 2004; Moyer‐Gusé, 2008). However, a recent study found that transportation into an E-E dramatic narrative related to unplanned teen pregnancy was associated with greater counterarguing (Moyer‐Gusé & Nabi, 2010). This unexpected finding was possibly due to the validity of the measurement of counterarguing they used, which might have made the target of the counterarguing unclear. In order to retest this relationship predicted by E-ELM, the current study examined transportation into an E-E dramatic narrative on a different topic.

The other form of narrative involvement in E-ELM, identification with characters, is defined as a process “in which an individual perceives another person as similar or at least as a person with whom they might have a social relationship’’ (Slater & Rouner, 2002, p. 178). This definition of identification does not differentiate three related constructs/variables as EORM does: identification, parasocial interaction, and similarity (Moyer‐Gusé, 2008). In the EORM, these three variables are proposed to explain how E-E helps overcome various forms of resistance to persuasive messages.

Identification. In the EORM, identification refers to “an imaginative process through which an audience member assumes the identity, goals and perspective of a character” (Cohen,

2001, p. 261). It is a cognitive and emotional process that involves loss of self-awareness and motivation to internalize the character’s goals (Cohen, 2001, p. 251). This temporary process is conceptually constituted by two components – affective empathy and cognitive empathy (Cohen,

2001; Tal-Or & Cohen, 2010, p. 404). Affective empathy involves feeling affinity towards the character, while cognitive empathy involves taking the character’s perspective in the narrative and internalizing the character’s goals (Tal-Or & Cohen, 2010, p. 404). Since identification and transportation are two constructs that both contribute to the immersion into narratives, it is

13

considered necessary to differentiate between them. Cohen (2010) found that, between the two, identification appears to be primarily affected by the valence of information about the character

(a character presented as good/positive vs. a character presented as bad/negative), which manipulates the emotional connection to the character; while transportation tends to be influenced by the time of deeds — information presented in the narrative about the character’s past versus information about the character’s future, which is related to suspense. This finding is in line with assumptions of existing theories that transportation is closely related to suspense

(Tal-Or & Cohen, 2010; Zillmann, 1991), while identification involves forming emotional connection with characters and internalizing their goals through perspective taking (Cohen,

2001; Tal-Or & Cohen, 2010).

Perceived similarity, or homophily, is another construct in EORM that is related to involvement with character. Perceived similarity with a character is hypothesized to enhance persuasive outcomes of the narrative by increasing perceived vulnerability to risks portrayed in narratives (Moyer‐Gusé, 2008). As defined by Moyer‐Gusé and Nabi (2010), perceived similarity refers to “a viewer’s judgment about the extent to which he or she and a character share common attributes, characteristics, beliefs, and/or values” (p. 30). It is considered an antecedent to identification (Cohen, 2001), but widely accepted as a distinct concept in existing research (Eyal & Rubin, 2003; Moyer‐Gusé, 2008; Moyer‐Gusé & Nabi, 2010). It is a distinct construct in EORM.

Parasocial interaction (PSI), or parasocial relationship (PSR), is a one-sided pseudorelationship a viewer forms with a mass communication “persona” (Horton & Wohl,

1956; Perse & Rubin, 1989). It is originally described as “the seeming face-to-face relationship between spectator and performer” (Horton & Wohl, 1956, p. 215). It resembles an interpersonal

14

relationship in that it develops over time. It may be enhanced through increased viewing and

similarity between media representations and interpersonal relationships (Horton & Wohl, 1956;

Perse & Rubin, 1989). PSI has been characterized as an affective viewer involvement, meaning it is a relationship based on a perceived affective bond with the character (Horton & Wohl, 1956;

Eyal & Rubin, 2003). PSI differs from identification in that it does not involve identity loss and it features an interactional aspect (Cohen, 2001). Moreover, it is thought to be relatively more lasting than identification, which typically takes place only while viewing (Cohen, 2001).

However, this is not to say that parasocial relationship cannot form upon a single encounter. This one-sided relationship that a viewer develops with characters in a narrative content is the other contributor to the reduction of reactance, as proposed by EORM. As the character that the viewer has a parasocial relationship with becomes a “peer” in his or her social life (Giles, 2002), it is less likely for the “peer” to be perceived as authoritative and manipulative by the viewer

(Burgoon, Alvaro, Grandpre, & Voloudakis, 2002; Moyer‐Gusé & Nabi, 2010). Thus, viewers are less likely to perceive the persuasive message as a “threat to freedom” and more likely to accept the message explicitly or implicitly sent by the “peer”.

Resistance to persuasion

The present study examined relative effects of exposure to an E-E program versus a PSA regarding bipolar disorder on three forms of resistance hypothesized by the EORM: counterarguing, perceived invulnerability, and reactance.

Counterarguing

As previously stated, theories in the realm of persuasion generally conceive

counterarguing as one of the obstacles to persuasive endeavors. Counterarguing is defined as

“generation of thoughts that dispute or are inconsistent with the persuasive argument” (Slater &

15

Rouner, 2002). The key premise of narrative persuasion is that various features of narratives can

hinder counterarguing. In narrative persuasion, counterarguing is considered incompatible with

narrative engagement because message scrutiny is less likely to happen if viewers are experiencing transportation and identification sufficiently (Slater, 1997; Slater & Rouner, 2002).

As posited by E-ELM, this form of resistance is negatively associated with narrative transportation and identification with characters (Slater & Rouner, 2002). In recent research, the relationship between these variables seems to vary depending upon how counterarguing is measured. Two methods have been used to operationalize counterarguing in the context of narrative persuasion. One technique asks participants to do a thought-listing task right after the exposure by writing down the thoughts they have when they are consuming a narrative. These thoughts are later coded into categories from favorable to unfavorable (Green & Brock, 2000).

The other technique uses summary measures to ask participants to estimate the extent to which, for instance, they found themselves looking for flaws in the way information was presented in the program (Moyer-Gusé, 2007). Studies using the thought-listing method to measure

counterarguing have had inconsistent findings (Green and Brock, 2000; Slater, Rouner & Long,

2006). A recent study examining identification with characters and discussion of sexually

transmitted infections after exposure to E-E narrative on STI related topic used summary

measures to operationalize counteraruging and concluded that findings were consistent with E-

ELM and EORM (Moyer‐Gusé, Chung & Jain, 2011).

Perceived invulnerability

Perceived invulnerability, also named “optimistic bias” in the literature, refers to people’s

tendency to believe that negative consequences of risky behaviors are less likely to happen to

themselves (Arnett, 2000; Moyer‐Gusé, 2008). It is also referred to as “personal fable” in

16

research on psychology of adolescence and described as “adolescents’ belief that they are special

in the sense of being unique, invulnerable, and omnipotent” (Goossens, Beyers, Emmen, &Van

Aken, 2002, p. 194). Studies suggest that individuals believe that they are less at risk than others

for many negative events (Klein & Helweg-Larsen, 2002; Weinstein, 1980). It seems reasonable

to postulate that people’s perception of others’ risk of getting mental illness would be higher than

their own. According to EORM, the perception of invulnerability will guide an individual to

resist prevention-related persuasive messages due to the false belief that the messages are not

relevant to him or her, and this tendency will be diminished by perceived similarity to and

identification with a vulnerable character (Moyer‐Gusé, 2008). In order words, the more viewers

identify with a character who is exposed to a threat, the more they will perceive themselves

vulnerable to the threat; the more viewers perceive that the vulnerable character is similar to

them, the higher level of vulnerability they will feel. Research testing these two assumptions on

physical-health-related topics has produced mixed findings. Moyer‐Gusé and Nabi (2010)

empirically tested the proposed relationships among identification, similarity and perceived

invulnerability on unplanned teen pregnancy and did not find strong support for the association

between identification and perceived invulnerability, however, this relationship was intensified

over time (two weeks after exposure). The hypothesized association between perceived similarity with character and perceived invulnerability was not supported. In a more recent study on a topic related to sexual health, identification with characters who modeled safe sex discussions was associated with viewers' motivation to engage in conversations about sexual health over the next

2 weeks, despite the lack of immediate effects on behavioral intentions. The results suggested

that identification improved viewers' self-efficacy and diminished their generation of

counterarguments. However, the association between identification with character and perceived

17

invulnerability proposed by EORM was less clear — identification did not reduce perceived

invulnerability (Moyer‐Gusé et al., 2011).

Reactance

The EORM posits that the narrative structure of E-E messages can reduce psychological

reactance by minimizing the perceived persuasive intent of the message. Psychological reactance

theory holds that individuals have a basic need for independence (Brehm, 1996; Brehm &

Brehm, 1981; Moyer‐Gusé & Nabi, 2010). Psychological reactance is defined as “the

motivational state that is hypothesized to occur when a freedom is eliminated or threatened with

elimination” (Brehm & Brehm, 1981, p. 37; Quick, Shen, & Dillard, 2013). The definition of

freedoms comprises affective, cognitive, and behavioral aspects (Brehm, 1966; Quick, Shen, &

Dillard, 2013), which includes freedom to hold an attitude or belief or the freedom to engage in a

particular behavior. In this sense, any intent to change others’ attitudes, beliefs, or behaviors can be perceived as a threat to their freedom. Put it in the persuasion context, when the persuasive intent of a message is explicit, people will likely feel threatened. Reactance has four components: freedom, threat to freedom, reactance, and restoration of freedom (Quick, Shen, & Dillard, 2013).

It is postulated that when a perceived freedom is inhibited or threatened, the individual will be motivated to rebuild that freedom in any of five ways: act the opposite of the advocated behavior, attack the source of threat, refuse to acknowledge the existence of the threat, practice a different freedom to regain sense of control, or increase liking for the risky choice (Quick, Shen,

& Dillard, 2013). Four of these coping methods are related to either ignoring or contradicting to the persuasive message, which can result in no effect or a boomerang effect—when message recipients move in the opposite direction of advocated position. Past research supports the relationship between perceived persuasive intent and reactance hypothesized by EORM (Moyer‐

18

Gusé, 2007; Moyer‐Gusé & Nabi, 2010). However, a more recent study manipulating the presence and absence of a drinking-and-driving PSA after exposure to narrative television content depicting the consequences of DUI found that an explicit persuasive message reminding the audience of the danger of drunk driving did not increase perceived persuasive intent or induce reactance. However, exposure to the E-E narrative about DUI alone was associated with greater perceived persuasive intent and reactance than exposure to the narrative along with the

PSA: “the inclusion of an explicit persuasive appeal led to less favorable attitudes toward drinking and driving”, which means the explicit appeal facilitated a prosocial outcome (Moyer-

Gusé et al., 2012). Two possible explanations were offered for this unpredicted finding. First, the message topic, being very familiar to college students, influenced the perceived persuasive intent of the narrative, and the addition of explicit appeal did not strengthen this perception. Second, audience members’ parasocial relationship with the spokesperson who delivered the explicit message (the PSA following the narrative) accounted for the lack of perceived persuasive intent, counterarguing or reactance in response to the explicit appeal (Moyer-Gusé et al., 2012). As previously stated, viewers are less likely to perceive the persuasive intent as a “threat to freedom” when the message is sent by the character they have parasocial relationship with.

Previous studies supported this negative association between parasocial relationship and reactance (Moyer‐Gusé & Nabi, 2010; Moyer‐Gusé et al., 2012). More interestingly, research data illustrated a new connection — PSR can contribute to the reduction of perceived persuasive intent in the first place (Moyer‐Gusé et al., 2012).

In general, some findings in the literature are inconsistent with the propositions in

EORM. The association between identification and perceived invulnerability is still not clear.

Additionally, research data does not always support the association between perceived similarity

19

and perceived invulnerability. The relationship between transportation and counterarguing is

generally supported despite some unexpected results potentially due to measurement issues of counterarguing. The association between perceived persuasive intent and reactance does not always hold up — sometimes it has been affected by the topic of the issue addressed by the E-E program. Finally, a new relationship between PSR and perceived persuasive intent was suggested by a recent study. These findings call for more research to test the propositions in EORM.

Additionally, as previously stated, mental illness has rarely been the focus of E-E endeavors despite much needed attention. The present study examined these relationships in a specific mental health context – bipolar I disorder.

2.4 Hypotheses

In an effort to better understand the effects of E-E on mental health, this study examined

how the underlying messages embedded in the dramatic storyline of a popular teen drama

affected viewers’ perceptions, intentions, and behaviors regarding bipolar disorder. Since the

focus of the study was on a television teen drama depicting the symptoms and treatments of

bipolar disorder, a relatively new topic to television programs, the persuasive outcomes in the

hypotheses were stated as pro-treatment behavioral intention and information-seeking behavior

involving bipolar disorder. Pro-treatment behavioral intention herein refers to the willingness to

search for information and to seek help for one’s bipolar disorder illness. It also involves helping

and encouraging people one cares about to search for information and seek help for bipolar

disorder. Information-seeking behavior involving bipolar disorder includes activities like looking

for information about bipolar disorder, contacting mental health professional to discuss bipolar

disorder, etc. This study did not examine the influence of this narrative on viewers’ attitudes towards people having this mental disorder or towards this mental disorder in general, mainly

20

because the attitudinal sphere of the narrative impact is closely related to mental illness stigma, which was not the focus of the current study. Moreover, stigma reduction typically requires multiple and sustained E-E endeavors to educate the public and change social norms, while this

study only examined outcomes associated with a single storyline in a particular TV drama in

order to test the relationships suggested by a current model of narrative persuasion. That being

said, the present study tested the effect of exposure to BPI stimuli on behavioral intention and

behaviors as well as several relationships within the model on a topic related to a mental illness –

bipolar I disorder using the constructs and associations postulated by EORM.

According to the propositions in EORM, reduced counterargument, reactance and

resistance would lead to story-consistent attitude and behaviors. Since traditional persuasive message processing does not have the set of mechanisms that facilitate reduction in resistance to change of attitude, it was assumed to be less effective. Additionally, it was assumed that the

presence of a persuasive message is more effective than absence of the persuasive message.

Therefore, the following set of hypotheses was proposed. As discussed earlier, the behavioral

intention tested in this study includes both first-person aspect (behavioral intention for oneself)

and third-person aspect (behavioral intention for people one cares about), so separate hypotheses

were developed for each aspect of pro-treatment behavioral intention.

H1a: Exposure to the E-E narrative program featuring bipolar disorder will be associated with

greater pro-treatment behavioral intention for “self” than exposure to the PSA about bipolar

disorder.

H1b: Exposure to the E-E narrative program featuring bipolar disorder will be associated with

greater pro-treatment behavioral intention in terms of helping people one cares about than

exposure to the PSA about bipolar disorder.

21

H1c: Exposure to the E-E narrative program featuring bipolar disorder will be associated with greater pro-treatment information seeking behavior for both oneself and someone they care about than exposure to the PSA about bipolar disorder.

H2a: Exposure to PSA about bipolar disorder is associated with greater pro-treatment behavioral intention for “self” than exposure to a narrative program and a PSA not relevant to bipolar disorder.

H2b: Exposure to PSA about bipolar disorder is associated with greater pro-treatment behavioral intention in terms of helping people one cares about than exposure to a narrative program and a

PSA not relevant to bipolar disorder.

H2c: Exposure to PSA about bipolar disorder is associated with greater pro-treatment information seeking behavior for both oneself and someone they care about than exposure to a narrative program and a PSA not relevant to bipolar disorder.

As discussed earlier, the narrative structure of E-E messages can reduce counterargument, reactance and resistance in recipients because it makes the persuasive intent less overt and it can draw the audience into the story world. In keeping with this assumption, the following hypotheses were tested:

H3: Exposure to the E-E narrative featuring bipolar disorder will be associated with less perceived persuasive intent than exposure to the PSA about bipolar disorder.

H4: Exposure to the E-E narrative featuring bipolar disorder will be associated with less counterargument than exposure to the PSA about bipolar disorder.

H5: Exposure to the E-E narrative featuring bipolar disorder will be associated with less reactance than exposure to the PSA about bipolar disorder.

22

H6: Exposure to the E-E narrative featuring bipolar disorder will be associated with less perceived invulnerability than exposure to the PSA about bipolar disorder.

As posited by EORM, the narrative structure of E-E programming should diminish reactance by disguising the message’s persuasive intent (Moyer‐Gusé, 2008). Thus, based on the

EORM, the following hypothesis was generated. [Note: all subsequent hypotheses were tested within the E-E stimulus condition only.]

H7: Lower estimates of the persuasive intent of a dramatic television narrative program featuring bipolar disorder will be associated with less reactance to the underlying message.

As stated earlier, another contributor to the reduction of reactance proposed by EORM is parasocial interaction with character. Consequently, the following hypothesis was tested.

H8a: Greater parasocial interaction with a character who experiences bipolar I disorder will be associated with less reactance.

Since there are two main characters in the narrative stimuli for the study, it is reasonable to expect that some viewers will identify and/or form parasocial relationship with the secondary, male character — the female character’s boyfriend whose life is affected by the female character’s various bipolar disorder symptoms. Therefore, a set of sub-hypotheses were proposed for those hypotheses involving character-related narrative involvement constructs including PSI, identification, and similarity.

H8b: Greater parasocial interaction with a character whose significant other experiences bipolar I disorder will be associated with less reactance.

Based on the postulations of E-ELM and EORM, identification with a character will lead to less counterarguing. As mentioned before, the process of identification involves loss of self- awareness, taking on the character’s perspective and an internalization of character’s goals, thus

23

people who experience this mental process will be less motivated to counterargue with the

underlying message of the narrative. In keeping with proposition, the following hypotheses were tested.

H9a: Identification with the main character in a dramatic narrative who experiences bipolar I

disorder will be negatively associated with counterarguing.

H9b: Identification with a main character in a dramatic narrative whose significant other is

experiencing bipolar I disorder will be negatively associated with counterarguing.

As suggested by the E-ELM and the EORM, the experience of being transported into a

dramatic narrative will inhibit message scrutiny and suspend disbelief. Therefore, viewers

experiencing transportation into the narrative storyline are less motivated to critically process the

underlying message of the story, which means they should be less likely to counterargue with it.

In keeping with this proposition, the following hypotheses were developed.

H10: Greater transportation into a dramatic television narrative about bipolar I disorder will be negatively associated with counterarguing.

In addition to reactance and counterarguing, perceived invulnerability is another form of resistance posited by the EORM that will diminish persuasive outcomes. The EORM proposes that perceived similarity and identification with a character who is vulnerable to a specific situation will increase the perceived vulnerability of viewers’ to the similar situation and as a result, they tend to be susceptible to the proposed coping method to prevent or remedy that situation. In line with this proposition, the following hypotheses were tested.

H11a: Identification with a character who experiences bipolar disorder will be negatively associated with perceived invulnerability.

24

H11b: Identification with a character whose significant other experiences bipolar disorder will be

negatively associated with perceived invulnerability.

H12a: Perceived similarity to a character who experiences bipolar disorder will be associated

with lower level of perceived invulnerability.

H12b: Perceived similarity to a character whose significant other experiences BP will be negatively associated with perceived invulnerability.

Keeping with EORM propositions, persuasive outcomes are reinforced by reduced resistance to narrative persuasion. Therefore, the following hypotheses were developed.

H13a: Less reactance will be associated with greater pro-treatment behavioral intention for “self” regarding bipolar I disorder.

H13b: Less reactance will be associated with greater pro-treatment behavioral intention involving bipolar I disorder in terms of helping people one cares about.

H13c: Less reactance will be associated with greater information seeking behavior for both

“self” and someone they care about involving bipolar I disorder.

H14a: Less counterarguing will be associated with greater pro-treatment behavioral intention for

“self” involving bipolar I disorder.

H14b: Less counterarguing will be associated with greater pro-treatment behavioral intention for people one cares about involving bipolar I disorder.

H14c: Less counterarguing will be associated with greater information seeking behavior involving bipolar I disorder for both oneself and someone they care about.

H15a: Lower perceived invulnerability will be associated with greater pro-treatment behavioral intention for “self” related to bipolar I disorder.

25

H15b: Lower perceived invulnerability will be associated with greater pro-treatment behavioral intention in terms of helping people one cares about involving bipolar I disorder.

H15c: Lower perceived invulnerability will be associated with greater information seeking behavior in regards to bipolar I disorder for both “self” and someone they care about.

Since perceived invulnerability of “others” is a new concept created in this study and it is a different aspect of perceived risk. A separate set of hypotheses were generated based on the idea of the original construct – perceived vulnerability.

H16a: There is a negative association between perceived invulnerability of “others” and pro- treatment behavioral intention regarding BPI for “self”.

H16b: There is a negative association between perceived invulnerability of “others” and pro-

treatment behavioral intention regarding BPI in terms of helping “others”.

H16c: A negative association exists between perceived invulnerability of “others” and

information seeking behavior concerning bipolar disorder.

Figure 1 and Figure 2 demonstrate all of the hypotheses developed based on the

Entertainment Overcoming Resistance Model.

PPI H7 (+) Reactance PSR with Characters H8a & H8b (-)

Transportation H10 (-) Counterarguing

H9a & H9b (-)

Identification H11a & 11b (-) with Characters Perceived

Invulnerability Perceived Similarity H12a & H12b (-)

Figure 1 Hypotheses 7 – 12 developed based on the EORM.

26

Reactance H13a - H13c (-)

H14a - H14c (-) Pro-treatment Behavioral Intention Counterarguing and Information Seeking Behavior

H15a - H15c (-)

Perceived Invulnerability

Figure 2 Hypotheses 13 – 15 created based on the EORM.

27

CHAPTER THREE

METHODOLOGY

3.1 Participants

This study employed a convenience sample at Florida State University. Undergraduates in four large lecture classes, one online course, and four smaller sized classes offered in School of Communication were recruited in exchange for extra course credit or required research credit

(N=180). A total of 180 participants between the ages of 17 and 36 attended the lab session of this study. Two participants over the age of 25 were dropped because they are not considered young adults. The current study defines young adult as “people in their late teens and early 20s”.

Another two cases were dropped due to invalid inputs for multiple questions. Of the 176 participants, 77.3% were female and 22.7% were male. In terms of ethnicity, 68.8% were

Caucasian, 19.3% were Hispanic, 7.4% were African-American, 2.3% were Asian-American,

1.1% were Asian. Data from 176 participants were used to analyze the posttest and the follow-up questionnaire.

3.2 Study Design

The study used a random group 3 (information condition) x 2 (time) mixed design with information condition as the between subjects variable and time as the within subjects variable.

This study was composed of a laboratory session and an online questionnaire two weeks after. In the laboratory session, participants were randomly assigned to one of three message conditions:

(PSA condition) a bipolar disorder PSA plus a video of 90210 excerpts without a bipolar disorder storyline, (E-E condition) an EPA Stormwater PSA with a video of 90210 excerpts with a bipolar disorder storyline, and (control condition) the EPA Stormwater PSA followed by the video of 90210 excerpts without the bipolar disorder storyline. Participants filled out a

28

questionnaire after being exposed to each stimulus program and completed a follow-up

questionnaire two weeks later. The following variables were measured in order right after the exposure to each PSA: reactance, perceived persuasive intent, counterarguing, perceived invulnerability and behavioral intention. These variables were also measured in the same order

after the exposure to each video of 90210 excerpts, and they were followed by another set of variables in the following order: transportation, perceived similarity, identification, parasocial interaction (or parasocial relationship) and control variables.

3.3 Stimulus Material

The E-E stimulus in the experimental condition was comprised of several scenes from four episodes of the primetime television drama 90210. In these episodes, Erin Silver, one of the main characters of this TV drama, displays various bipolar symptoms, such as hypersexuality, erratic behavior, racing speech pattern, euphoria, lack of sleep, confusion, sadness and poor

judgment. In the paramount episode, she made some destructive decisions, which lead to her

risky behavior (running towards a speeding train) at a rail station. Dixon, Silver’s boyfriend,

previously thought Silver was just “a drama queen” and was irritated by her unpredictable

behaviors, yet later saved her from future destructive acts by realizing that she has the same mental illness as his birth mother and showed his support. Dixon described his birth mother’s bipolar symptoms when he tried to communicate with Silver and to show his understanding.

Silver was later diagnosed with bipolar disorder and received treatment. She talked to Dixon

about her frustration at having to maintain a strict treatment regimen. Her sister, Kelly, being

very supportive of her treatment, tries everything she can to keep her highly regimented. The

following narrative depicts Silver as someone who can successfully manage her mental illness

and live a rewarding and productive life. Irrelevant storylines were removed in the editing

29

process. The control narrative is a different episode of 90210 without a bipolar storyline but

mainly about the relationship between Erin Silver and her boyfriend Dixon. These two dramatic narrative clips are both approximately 30 minutes in length. The PSA stimulus is a PSA

addressing bipolar I disorder. It is under 1.5 minutes long and delivered by a practicing

psychiatrist. She talks about what bipolar I disorder (manic depressive disorder) is, how it can

affect a person’s life, some of the symptoms, and the age when it most often developed. The

psychiatrist states at the end of the PSA “with proper diagnosis and treatment, most people with

bipolar I disorder can lead healthy and productive lives. If you think you may be suffering from

bipolar I disorder, PLEASE see a mental health processional!” The psychiatrist mentioned the

mood swings that characterize bipolar I disorder and some of the symptoms of manic and

depressive phases. For example, she stated that sudden euphoria along with hyperactivity,

decreased need for sleep, reckless behavior, inflated self-esteem, racing thoughts, risky sexual

behavior or foolish investment are some of the characteristics in manic phase; feelings of

hopelessness and worthlessness and thoughts of death or suicide are characteristics of the

depressive phase. These symptoms were portrayed in the 90210 bipolar disorder narrative and

demonstrated by the character Erin Silver’s behaviors and by her boyfriend Dixon’s description

of his birth mother. The other control video material is a PSA about stormwater runoff produced

by US Environmental Protection Agency. It is of the same length as the bipolar disorder PSA.

3.4 Procedures

Participants came to one of two chosen computer labs in School of Communication to

attend a laboratory experiment. After greeting them in the classroom connected to the lab, the

experimenter (the author) asked them to read the consent form and sign it if they agreed to

participate. The experimenter let the participants know that they could ask any question when

30

they were reading the informed consent. Participants were told in the consent form that they were going to view a public service announcement and a 30-minute video of excerpts from a TV drama and that they will fill out a questionnaire after viewing each video. Similar instruction was shown on each computer screen at the beginning of the lab session. Before leading each participant to the computer lab, the experimenter told him/her that the videos would automatically start and stop, so that he/she would not need to adjust anything when viewing the videos and all the instructions would show up on the computer screen. After submitting the signed consent form, each participant was told to pick a computer before the experimenter signed him/her into the MediaLab software. There were a set of headphones next to the keyboard of each computer for the participant to use. Participants were randomly assigned to one of the three

viewing conditions mentioned previously. At the beginning of the session, they were asked to

create a confidential code following the instruction on screen. They would need to provide the

code they created for the follow-up questionnaire in two weeks. They viewed the PSA first and

completed a short questionnaire about their reaction to the PSA. Right after that, they were

shown a 30-minute edited video of television excerpts taken from the primetime teen drama -

90210. They completed another questionnaire after viewing the clip; this questionnaire took

about 10-15 minutes to complete. On average, the laboratory session took about 50 minutes. Two

weeks after the lab session, participants were sent an email instruction containing a link to a very

brief online questionnaire to measure behavioral intention and pro-treatment behaviors.

3.5 Measures

At the beginning of this section, variables measured only in the posttest are presented.

Some of these variables are independent variables in some hypotheses and dependent variables

31

in other hypotheses. Variables assessed in the follow-up test are introduced right after. Lastly,

measures of control variables or background variables are described.

3.5.1 Perceived Persuasive Intent

Perceived persuasive intent refers to an individual’s perception in regards to the level of

persuasiveness of a message. It was measured with a single question item adapted from previous studies conducted by Moyer‐Gusé (2007) and Moyer‐Gusé and Nabi (2010). It asks participants

the degree to which they think the video was created to entertain versus to persuade, the options range from 1 (to persuade) to 7 (to entertain). This item was measured for both the bipolar disorder PSA (M = 5.78, SD = 1.19) and the 90210 excerpts featuring bipolar disorder (M = 1.79,

SD = 1.08) for comparison. It was reverse coded so that greater value represents stronger perceived persuasive intent.

3.5.2 Transportation

As previously noted, transportation refers to the “convergent process, where all mental systems and capacities become focused on events occurring in the narrative” (Green & Brock,

2000, p. 701). It was evaluated using eight question items adapted from the transportation scale used by Green and Brock (2000) and Moyer‐Gusé (2007). For example, one of the items states

“While I was watching, I was interested to learn how the show would end” with the answer ranges from 1 “strongly disagree” to 5 “strongly agree”. Seven items in the original transportation scale were excluded due to the tele-presence nature of the stimuli and the nature of the topic issue. The eight items were later summed and averaged (M = 3.35, SD = .64, α = .76), with higher scores indicating greater transportation. Four of the distraction-related negatively worded items were reverse coded. The reliability of the scale increases to α = .82 after dropping

32

one item – “I found myself thinking of ways the show could have turned out differently” (M =

3.40, SD = .73, α = .82).

3.5.3 Identification

Identification refers to “an imaginative process through which an audience member

assumes the identity, goals and perspective of a character” (Cohen, 2001, p. 261). Identification with characters was assessed using the method proposed by Cohen (2001), refined by Eyal and

Rubin (2003) and adapted by Moyer‐Gusé (2007). Eight items were included. Sample items

including, “At key moments in the show, I felt I knew exactly what [character’s name] was going

through” (1 = strongly disagree, 5 = strongly agree). Participants in all three conditions

responded to the same eight items for both female main characters and male main characters in

the video they viewed. For the BPI narrative group, these items were averaged to create a

measure of identification with the female main character (M = 3.03, SD = .73, α = .84), and

identification with the male main character (M = 3.49, SD = .66, α =. 88), with higher scores

indicating greater identification.

3.5.4 Perceived Similarity

Perceived similarity refers to “a viewer’s judgment about the extent to which he or she

and a character share common attributes, characteristics, beliefs, and/or values” (Moyer‐Gusé &

Nabi, 2010, p. 30 ). The attitude homophily scale developed by McCroskey, Richmond and Daly

(1975) was adopted to measure perceived similarity with characters. Participants in all conditions

were asked to indicate their degree of similarity to the main female and male characters in the

way they act, think, and behave like them on a Likert-type scale. For the BPI narrative group,

these items were summed and averaged to measure perceived similarity to the female main

33

character (M = 1.93, SD = .88, α = .94) and to the male main character (M = 3.07, SD = .90, α =

.91), with higher scores indicating greater perceived similarity.

3.5.5 Parasocial Interaction

Parasocial Interaction or PSI refers to a one-sided pseudorelationship a viewer forms with a mass communication “persona” (Horton & Wohl, 1956; Perse & Rubin, 1989). It was assessed with Rubin and Perse’s PSI scale (Rubin & Perse, 1987). Sample items include, “[name of the character] makes me feel comfortable, like I’m with a friend.” “If I could, I would like to meet

[name of the character] in person.” An item developed by Levy (1979) and refined by Moyer‐

Gusé (2007) — “I like to compare my ideas with what [name of the character] says” was also included. Another item used by Moyer‐Gusé (2007) was also included — “When [name of the character] shows me how she feels about an issue, it helps me make up my own mind about the issue”. All these items were averaged to create a measure of parasocial relationship with the female main character (M = 2.53, SD = .75, α = .92) and with the male main character (M = 3.03,

SD = .75, α = .93), with higher scores indicating greater parasocial interaction.

3.5.6 Perceived Invulnerability

Perceived invulnerability refers to people’s tendency to believe that negative consequences of risky behaviors are less likely to happen to themselves (Arnett, 2000; Moyer‐

Gusé, 2008). To measure perceived invulnerability, the current study adapted four items from previous research (Gibbons et al, 2003; Moyer‐Gusé, 2007). Due to the different nature of the topic issue in this study from those of the studies in which this measure was usually used, the question items were changed to adapt to the mental health issue being addressed. Participants were asked to indicate their perceived likelihood of getting worse and doing something self- destructive if they had bipolar disorder and did not seek treatment. Response options range from

34

1 (no chance) to 7 (definitely would happen). Another item asking their overall perception of risk without treatment was also included. Participants were also asked to estimate their chance of developing bipolar disorder in their lifetime, however, this item was dropped later in the data analysis because it did not correlate with other items. After all, bipolar I disorder is not an illness

that everyone can develop. Moreover, as suggested by the definition of perceived invulnerability

and past research on optimistic bias (Klein & Helweg-Larsen, 2002; Lapsley & Hill, 2010;

Weinstein, 1980), people’s perception of risk for others is typically higher than for themselves,

especially adolescent and young adults who often believe that “others” would be at higher risk

than themselves for negative events. So, it seems reasonable to expect a generally pro-treatment

attitude among young adults in terms of helping “others” to be influenced by the third-person

perspective of perceived vulnerability. Therefore, perceived vulnerability of “others” — people

they care about — was also measured using the same set of questions. Items for each target (self

vs. others) were summed and averaged to create a scale of perceived invulnerability of “self” (M

= 2.52, SD = 1.10, α = .66) and perceived invulnerability of “someone you care about” (M =

2.56, SD = 1.10, α = .73), with higher scores indicating stronger perceived invulnerability1.

3.5.7 Counterarguing

Counterarguing refers to “generation of thoughts that dispute or are inconsistent with the

persuasive argument” (Slater & Rouner, 2002). As previously stated, previous studies usually

use either summary measures in the form of closed-ended questions (Moyer‐Gusé, 2007; Moyer‐

Gusé & Nabi, 2010; Moyer-Gusé, Mahood & Brookes, 2011; Nabi, Moyer‐Gusé, & Byrne,

2007) or thought-listing (Green & Brock, 2000) to assess counterarguing with the underlying

message. However, each type of measure has its limitations. Because it has been suggested that

1Previous studies have reported somewhat low reliability of these variables. For example, r = .57 in posttest (Moyer-Gusé & Nabi, 2010), and r = .55 in pretest (Moyer-Gusé, 2007). 35

thought-listing often evokes responses not relevant to the underlying persuasive message and

does not adapt to narratives well (Green & Brock, 2000; Moyer‐Gusé, 2007), the current study

used four closed-ended self-report questions from previous studies (Moyer‐Gusé, 2007; Moyer‐

Gusé & Nabi, 2010). Sample items include: “While watching the program, I couldn’t help

thinking about ways that the information being presented was inaccurate or misleading” (1=

strongly disagree 5= strongly agree). These four items were averaged to form a scale of

counterarguing in the data analyses (M = 2.55, SD = 1.02, α = .89), with higher scores indicating

greater counterarguing.

3.5.8 Reactance

Psychological reactance is defined as “the motivational state that is hypothesized to occur

when a freedom is eliminated or threatened with elimination” (Brehm & Brehm, 1981, p. 37;

Quick, Shen, & Dillard, 2013). Reactance was measured through a combination of its emotional and cognitive components – anger and negative cognitions (Dillard & Shen, 2005). The four

closed-ended items developed in previous research (Dillard & Shen, 2005) were used to measure the emotional component of reactance - anger. On a scale of 1 (not at all) to 7 (very much), participants were asked to report how angry, irritated, annoyed, aggravated they were while watching the program (M = 2.01, SD = 1.44, α = .96). There are two ways to measure the cognitive component of reactance — negative cognitions — the one often used is a thought- listing method, which asks participants to write out thoughts that they had while consuming the message. Their listed thoughts are then coded as positive, neutral, or negative through a four-step

process by coders. Only negative cognitions are used in the data analysis (Quick, Shen, &

Dillard, 2013). The other technique is a summary measure, which uses closed-ended questions to capture negative cognitions. In the current study, closed-ended items measuring negative

36

cognitions were adapted from the “cognitive evaluations” measure in a reactance study conducted by Miller, Lane, Deatrick, Young, and Potts (2007). Items measuring attention and

enjoyment were dropped because they overlap with some of the items in the narrative

involvement measures. The remaining items included “The video was reasonable/interesting/not

exaggerated/ pleasant/reflected my attitudes” and “The content of the video mattered to me”.

They were measured with Likert-type items with response categories from “1 = strongly

disagree” to “5 = strongly agree” (M = 2.77, SD = .60, α = .68). These six items were

transformed to 7-point Likert scale items so that they could be combined with the affective

component of reactance to form an overall measure of reactance (M = 3.00, SD = .95, α = .84),

with higher scores indicating greater reactance.

3.5.9 Pro-treatment Behavioral Intention

Pro-treatment behavioral intention refers to the likelihood that people will seek

information regarding bipolar disorder and seek help from mental health professional for

themselves and/or someone they care about if they or people they care about experience bipolar

disorder symptoms. It was measured at the posttest and in the follow-up test two weeks later

across three conditions. Specifically, participants were asked to report their intention toward

searching for information about BPI and seeking help from mental health professionals if they

thought they had bipolar disorder, as well as their likelihood to help someone they care about

who suffers from this mood disorder seek help from mental health professionals. The two items

were “How likely is it that you would search for information about bipolar disorder if you

thought you had bipolar disorder?” (M = 5.95, SD = 1.53) and “How likely do you think it is that

you would seek help from mental health professionals if you experienced bipolar disorder

symptom (s)?” (M = 4.87, SD = 2.15, 1 = very unlikely, 7 = very likely). The items measuring

37

behavioral intention related to pro-treatment information searching and seeking help from mental

health professionals for “self” were not strongly correlated at the posttest, Pearson’s r = .23, p =

.001. However, conceptually, these two items were considered two aspects of the behavioral

intention tested in this study, so they were still combined as a measure of behavioral intention for

the posttest when testing relevant hypotheses to make results clearer (M = 5.40, SD = 1.46). The

Spearman-Brown split-half reliability coefficient for this two-item measure was rsb = .38. There

was a strong correlation between the two items measuring pro-treatment behavioral intention

related to helping “someone you care about” to get help (Pearson’s r = .74, p < .001, Spearman-

Brown Coefficient rsb = .85) at the posttest. Consequently, they were averaged to form a measure

of pro-treatment behavioral intention in terms of “others” (M = 5.63, SD = 1.35). For both

indexes, a higher score represents greater pro-treatment behavioral intention.

3.5.10 Information-Seeking Behavior Involving Bipolar Disorder at Follow-up

Information-seeking behavior involving bipolar disorder includes activities like looking

for information about bipolar disorder, talking to mental health professionals about bipolar

disorder, visiting bipolar disorder’s organizational websites etc. To examine the behavioral

effects of the E-E narrative in the two weeks after exposure, information-seeking behavior was

measured in the follow-up test. Participants were asked whether they had performed one or more

of the information-seeking behaviors mentioned earlier during the past two weeks. Sample items in the follow-up questionnaire include, “During the last two weeks have you contacted a mental health professional for information about bipolar disorder for yourself or someone else?” (0 = no;

1 = yes). These items were summed to form a measure of information–seeking behavior(s), with higher scores indicating more information-seeking behaviors.

38

3.5.11 Control Variables

3.5.11.1 Past BPI-related behaviors. In the posttest questionnaire administered after exposure to the bipolar disorder or control message, participants were asked their past behaviors related to seeking information or seeking help involving BPI for themselves or someone else.

They also estimated the number of times they searched for information regarding BPI and the number of people with whom they discussed BPI.

3.5.11.2 BPI-related experience. As bipolar disorder related experience will likely influence participants’ attitude towards the stimuli. Participants were asked whether they had ever been diagnosed as bipolar disorder. They were also asked whether or not they know someone around them suffers from bipolar disorder. This was asked at the end of the post-test questionnaire to avoid hypothesis guessing.

3.5.11.3 Demographic information. Participants were asked to indicate their gender, ethnicity and age in posttest questionnaires.

3.5.11.4 Perceived ages of characters. Two items measuring participants’ perception of the two main characters’ ages were included in the bipolar disorder narrative post-test

questionnaire. The mean perceived ages participants reported were both 17 and they were at

normal range (15-22 for Silver, SD = .89, 15-22 for Dixon, SD = .93).

3.5.11.5 Enjoyment. In order to examine unpredictable technical influences on the

effects of the narrative, two enjoyment items adapted from Tal-Or & Cohen (2010)’s study were used to measure viewer’s overall enjoyment of the videos of 90210 excerpts. Also, at the beginning of the follow-up survey, participants were asked to rate how favorable were their

impressions of the show on a scale from 1 (= very unfavorable) to 7 (= very favorable).

39

3.5.11.6 Perceived social norms. In the EORM, perceived social norms are considered

another form of resistance. Previous studies on the Focus Theory of Normative Conduct suggest that perceived social norms can be a strong predictor of health and related behaviors (Kallgren,

Reno & Cialdini, 2000; Rhodes & Ewoldsen, 2009). As proposed in the EORM, when a risky behavior becomes normative in people’s perception, they will often ignore the negative

consequences associated with that behavior, which means, persuasive messages discouraging

that act will be resisted. The current study included nine Likert-type scale items testing the

perceived social norms involving bipolar disorder for a future study. For example, one of the

items asking participants’ level of agreement with this statement —“many people like me have bipolar disorder”.

40

CHAPTER FOUR

RESULTS

The probability of Type I error was set at .05 for all tests of significance reported in this thesis. Unless otherwise specified, all p-values reported in this study were one-tailed. Table 4 shows a summary of hypotheses tested in this study and results of those tests. Table 5 reports bivariate correlations between variables tested in the E-E BPI narrative condition. Figure 3 and

Figure 4 capture all the associations tested in the E-E BPI narrative condition.

4.1 Preliminary Analyses

4.1.1 Random Assignment Check Verifying Equivalence of BPI-related Experience and

Past Behaviors across Groups

Among the 176 participants, 31.3% reported that they or someone they care about have been diagnosed with bipolar disorder. These participants were evenly distributed among the three conditions, χ2(4) = 4.32, p = .37. Also, there was no difference in the number of participants in the three conditions who know someone suffering from bipolar disorder, χ2(6) = 6.04, p = .42.

Nearly 14% of the 176 participants had looked for information about treatment of bipolar disorder for themselves or someone else before they participated in this study, and that percentage did not differ among conditions, χ2(2) = .19, p = .91. Only 2.8% of all participants had made an appointment with a mental health professional to discuss bipolar disorder for themselves or someone else, and there was no difference in this percentage among groups, χ2(2)

= 2.95, p = .23. Only 3.4% of the participants had visited bipolar disorder organizations’ websites. No difference was found among the three groups in regard to this past behavior, χ2(2) =

1.13, p = .57. The percentage of people who reported that they have thought about preventing

41

bipolar disorder for themselves or others was 14.8%. There was no group difference in this

variable either, χ2(2) = 1.38, p = .50.

4.1.2 Enjoyment of Narrative Stimuli

In order to ensure the two narrative stimuli were equally interesting, two enjoyment items

were included in the posttest, they were summed and averaged to create an overall enjoyment

measure. Enjoyment did not differ among groups (BPI narrative condition: M = 3.51, SD = 1.08;

BPI PSA condition: M = 3.53, SD = 1.05; and control condition: M = 3.82, SD = 1.00), F (2) =

3.45, p = .21, with higher score indicating greater enjoyment (possible range: 1~5). In the follow-

up questionnaire two weeks after, participants were also asked to rate how favorable their

impressions of the 90210 video were. No significant difference was found between the group

exposed to the narrative stimuli containing the BPI storyline and the group that viewed the

narrative stimuli without the BPI storyline (BPI narrative condition: M = 4.16, SD = 1.49; BPI

PSA condition: M = 4.60, SD = 1.54; control condition: M = 4.56, SD = 1.79), F (2) = 1.36, p =

.26, with higher score reflecting greater favorable impression (possible range: 1~7). Therefore, participants of these two groups reported the same level of enjoyment. In general, as the mean

scores slightly leaning toward agreement to the statements indicating enjoyment, participants seemed to have experienced low-to-medium level of enjoyment.

4.1.3 Tests for Gender Effects in Narrative Involvement

In past research, gender has been considered as a factor for narrative involvement constructs such as identification with characters, perceived similarity with characters, and parasocial relationship with characters, and individuals typically identify more with same-gender characters (Hoffner & Buchanan, 2005; Steinke et al, 2012). Therefore, gender differences were examined for these variables in the present study. However, t-tests revealed that gender was not a

42

factor in any of these variables. There was no difference between females and males in

identification with the female main character, t(59) = .87, p =.19, in identification with the male

main character, t(59) = 1.139, p =.13, in perceived similarity with the female main character, t(59) = -.74, p =.23, in perceived similarity with the male main character, t(59) = -.329, p =.37, in parasocial relationship with the female main character, t(59) = .772, p =.22, or in parasocial relationship with the male main character, t(59) = .644, p > .26. As a result, participant gender was not included as a factor in the following hypothesis tests.

4.1.4 Summary of All Key Variables Measured

Table 1 and Table 2 report means and standard deviations of variables measured in this study. Some of the variables are measured only within the two groups exposed to BPI information and the rest of the variables are measured across two or all three research conditions.

Table 1 shows means and standard deviations for variables measured in all three conditions.

Table 2 presents all the variables measured only within the E-E narrative group. Table 3 shows the means of all persuasive outcomes assessed in this study by condition.

Table 1

Means and Standard Deviations for Variables Measured Across All Three Conditions

Variable n M SD Skew Possible Range Perceived persuasive intent 1~7 BPI Narrative 61 1.79 1.08 1.50 BPI PSA 58 5.78 1.19 -.60 Non-BPI Narrative 57 1.33 .81 3.92 Behavioral intention for self 172 5.40 1.46 -.83 1~7 Behavioral intention for others 176 5.63 1.35 -1.07 1~7 Counterarguing 119 2.55 1.02 .37 1~5 Reactance 119 3.00 .95 1.06 1~7 Perceived invulnerability of self 176 2.52 1.10 .81 1~7 Perceived invulnerability of others 176 2.56 1.10 .37 1~7 Information seeking behavior 174 .15 .43 3.00 0~5

43

Table 2

Means and Standard Deviations for Variables Measured in BPI E-E Narrative Condition

Possible Variable n M SD Skew Range Identification with Silver 61 3.03 .73 .15 1~5 Identification with Dixon 61 3.49 .66 -.64 1~5 Transportation 61 3.40 .73 -.19 1~5 Perceived similarity with Silver 61 1.93 .88 1.22 1~5 Perceived similarity with Dixon 61 3.07 .90 -.82 1~5 Parasocial relationship with Silver 61 2.53 .75 -.22 1~5 Parasocial relationship with Dixon 61 3.03 .75 -.62 1~5 Reactance 61 3.39 1.08 .45 1~7 Counterarguing 61 3.04 .99 -.19 1~5 Perceived invulnerability of self 59 2.28 .95 .65 1~7 Perceived invulnerability of others 61 2.40 1.07 .46 1~7 Behavioral intention for self 60 6.01 1.03 -1.22 1~7 Behavioral intention for others 61 5.74 1.36 -1.16 1~7 Information-seeking behavior 61 .16 .45 2.88 0~5

Table 3

Means, Standard Deviations for BPI Pro-treatment Behavioral Intentions and BPI Information Seeking Behaviors by Condition

Condition Behavioral Intention Behavioral Intention Information-Seeking for “Self” for “Others” Behavior

BPI E-E narrative n 60 61 61 M (SD) 6.01 (1.03) 5.74 (1.36) .164 (.45) BPI PSA n 55 58 57 M (SD) 4.40 (1.23) 5.45 (1.27) .158 (.45) Control n 57 57 56 M (SD) 5.73 (1.56) 5.71 (1.42) .125 (.38) Total n 172 176 174 M (SD) 5.40 (1.46) 5.63 (1.35) .149 (.43)

44

4.2 Hypotheses Testing

4.2.1 Comparing Message Effectiveness of Stimulus Programs

Hypotheses H1a – H1c predicted greater pro-treatment behavioral intention and

information seeking behavior involving bipolar disorder after exposure to the E-E narrative video

addressing bipolar disorder than after exposure to the PSA about bipolar disorder. Specifically,

Hypothesis 1a predicted greater pro-treatment behavioral intention for “self” in the E-E group.

Consistent with this hypothesis, a t-test showed that participants in the E-E group reported higher behavioral intention (M = 6.01, SD = 1.03) at the posttest than participants in the BPI PSA group

(M = 4.40, SD = 1.23) and the difference was statistically significant, t(113) = 7.62, p < .001.

Thus, hypothesis 1a was supported. When it comes to behavioral intention in regards to helping someone participants care about (H1b), there was no significant difference between the BPI E-E

narrative group (M = 5.74, SD = 1.36) and the BPI PSA group (M = 5.45, SD = 1.27), t(117) =

1.199, p = .12. So, hypothesis 1b was not supported. When comparing effects on information- seeking behavior reported at time of follow-up (H1c) between these two conditions, an

independent samples t-test did not find a significant difference between the BPI E-E narrative group (M = .16, SD = .45) and the BPI PSA group (M = .16, SD = .45), t(116) = .072, p = .47.

Therefore, Hypothesis 1c was not supported. In sum, among these three sub-hypotheses, the BPI narrative was associated with more pro-treatment behavioral intention for “self” than the BPI

PSA but equal behavioral intention for helping others after viewing and information-seeking

behavior two weeks later.

Hypothesis H2a – H2c predicted that exposure to the BPI PSA would yield a stronger

behavioral effect than watching videos without bipolar disorder content. Hypothesis 2a predicted

that exposure to BPI PSA would lead to stronger BPI pro-treatment behavioral intention than

45

exposure to videos that were irrelevant to bipolar disorder. Conversely, participants in the control

group (M = 5.73, SD = 1.56) reported significantly higher pro-treatment behavioral intention

than the BPI PSA group (M = 4.40, SD = 1.23), t(110) = -5.00, p < .001. Therefore, this

hypothesis was not supported. Because of this unexpected result, behavioral intention was also

compared between the E-E group and the control group. The mean difference in behavioral

intention between these two groups did not reach statistical significance, t(96.544) = 1.142, p =

.13. Hypothesis 2b predicted greater pro-treatment behavioral intention in terms of helping

someone they care about in BPI PSA group than in the control group. However, there was no

significant difference between the BPI PSA group (M = 5.45, SD = 1.27) and the control group

(M = 5.71, SD = 1.42) at the posttest. Thus, hypothesis 2b was not supported. With regard to

self-reported information-seeking behavioral at time of follow-up, there was no significant mean

difference between the BPI PSA group (M = .16, SD = .45) and the control group (M = .13, SD =

.38), t(111) = .415, p < .34. Consequently, H2c was not supported. Taken together, this set of

hypotheses predicting more behavioral intention and behaviors in BPI PSA condition was not

supported. Unexpectedly, participants in control group yielded more pro-treatment behavioral

intention than those in the BPI PSA group. Also, control group and the E-E group have equal pro-treatment behavioral intention.

To further explore the odd results from the test of H2a, the two aspects of behavioral intention for “self” (searching for more information and seeking help from a medical professional) were compared across three conditions, as a supplemental test. While there was no difference in behavioral intention of information-searching across three conditions, F(2) = .018, p = .98, significant differences were found in behavioral intention to seek professional help, with participants in BPI E-E narrative condition (M = 6.05, SD = 1.31) reporting greater intention than

46

participants in both BPI PSA condition (M = 2.87, SD = 1.78), t(98.622) = 10.871, p < .01, and the non-BPI/control condition (M = 5.53, SD = 1.84), t(100.427) = 1.766, p < .05.

4.2.2 Comparing Perceived Persuasive Intent

Hypothesis 3 predicted that the perceived persuasive intent of the BPI E-E narrative would be less than the perceived persuasive intent of the BPI PSA. The result of an independent samples t-test suggested that perceived persuasive intent of BPI PSA (M = 5.78, SD = 1.19) was significantly higher than BPI E-E narrative (M = 1.79, SD = 1.08), t(117) = 1.209, p < .001.

Thus, H3 was supported.

4.2.3 Comparing Resistance between Two Message Conditions

Hypothesis 4 predicted that exposure to the bipolar disorder E-E narrative would be associated with less counterarguing than exposure to the BPI PSA. In opposition to this hypothesis, participants in the E-E group (M = 3.04, SD = .99) reported more counterarguing than participants in the PSA group (M = 2.03, SD = .78). The mean difference between two groups was statistically significant, t (112.84) = 6.211, p < .001. As a result, H4 was not supported.

Hypothesis 5 predicted less reactance associated with exposure to the BPI E-E narrative than reactance to BPI PSA. However, counter to this hypothesis, participants in the BPI E-E narrative group reported significantly more reactance (M = 3.39, SD = 1.08) than those in the

BPI PSA group (M = 2.58, SD = .54), t (88.975) = 5.22, p < .001. Therefore, hypothesis 5 was not supported.

Hypothesis 6 predicted that exposure to the BPI E-E narrative would be associated with less perceived invulnerability than exposure to the BPI PSA. The independent samples t-test showed significantly lower perceived invulnerability of “self” among participants in E-E narrative group (M = 2.28, SD = .95) than participants in the BPI PSA group (M = 2.72, SD =

47

Table 4

Summary of Hypotheses and Results

Supported Exposure to BPI E-E narrative is associated with greater pro-treatment behavioral intention than exposure to the BPI PSA(H1a) Yes Exposure to BPI E-E narrative is associated with greater pro-treatment behavioral intention regarding helping “others” than exposure to BPI PSA (H1b) No Exposure to BPI E-E narrative is associated with greater BPI information seeking behavior than exposure to BPI PSA (H1c) No Exposure to BPI PSA is associated with greater pro-treatment behavioral intention than exposure to non-BPI programs(H2a) No1 Exposure to BPI PSA is associated with greater pro-treatment behavioral intention regarding helping “others” than exposure to non-BPI programs(H2b) No Exposure to BPI PSA is associated with greater BPI information seeking behavior than exposure to non-BPI programs(H2c) No Exposure to BPI E-E narrative is associated with less PPI than exposure to the BPI PSA (H3) Yes Exposure to BPI E-E narrative is associated with less counterarguing than exposure to the BPI PSA (H4) No1 Exposure to BPI E-E narrative is associated with less reactance than exposure to the BPI PSA (H5) No1 Exposure to BPI E-E narrative is associated with less perceived invulnerability than exposure to the BPI PSA (H6) Yes2 In BPI E-E narrative, the less the perceived persuasive intent, the less the reactance (H7) No Greater parasocial relationship with BPI character - Silver, less reactance (H8a) Yes Greater parasocial relationship with BPI character’s close friend – Dixon, less reactance(H8b) Yes More identification with Silver, less counterarguing (H9a) No More identification with Dixon, less counterarguing (H9b) Yes More transportation into BPI E-E narrative, less counterarguing (H10) Yes More identification with Silver, less perceived invulnerability (H11a) Yes2 More identification with Dixon, less perceived invulnerability (H11b) Yes2 More perceived similarity to Silver, less perceived invulnerability (H12a) No More perceived similarity to Dixon, less perceived invulnerability (H12b) No Less reactance, greater pro-treatment behavioral intention (H13a) No Less reactance, greater pro-treatment behavioral intention regarding “others” (H13b) No Less reactance, greater BPI information seeking behavior (H13c) No Less counterarguing, greater pro-treatment behavioral intention for “self” (H14a) No Less counterarguing, greater pro-treatment behavioral intention regarding “others” (H14b) No1 Less counterarguing, greater BPI information seeking behavior (H14c) No Lower perceived invulnerability of “self”, greater pro-treatment behavioral intention for “self” (H15a) Yes3 Lower perceived invulnerability of “self”, greater pro-treatment behavioral intention regarding “others” (H15b) Yes Lower perceived invulnerability of “self”, greater BPI information seeking behavior (H15c) No Lower perceived invulnerability of “others”, greater pro-treatment behavioral intention for “self” (H16a) Yes Lower perceived invulnerability of “others”, greater pro-treatment behavioral intention concerning “others” (H16b) Yes Lower perceived invulnerability of “others”, greater BPI information seeking behavior (H16c) Yes3 Note. 1: Opposite result significant. 2: Both perceived invulnerability and perceived invulnerability of “others” were supported. 3: Results in predicted direction, hypothesis supported at p < .10.

48

.90), and the difference was significant, t(115) = - 2.57, p < .01. Moreover, perceived

invulnerability of “others” was also significantly lower among those in the BPI narrative group

(M = 2.40, SD = 1.07) than in BPI PSA group (M = 2.83, SD = 1.03), t(117) = - 2.221, p < .05.

Hence, H6 was supported. Also, because of the unanticipated findings on behavioral intentions among message conditions, a supplemental test was conducted to compare perceived invulnerability between the BPI E-E narrative condition and the non-BPI/control condition as well as between the BPI PSA condition and the non-BPI condition. There was no significant difference in perceived invulnerability of “self” between BPI narrative condition (M = 2.28, SD

= .95) and the control condition (M = 2.56, SD = 1.37), t(95.501) = -1.233, p = .11. Likewise, no appreciable difference was found in perceived invulnerability of “others” between the BPI

narrative (M = 2.40, SD = 1.07) and the control (M = 2.45, SD = 1.16) conditions, t(116) = -.25,

p = .40. There was no difference in perceived invulnerability of “self” between BPI PSA

condition (M = 2.72, SD = .90) and the control condition, t(92.559) = .757, p = .23. However,

perceived invulnerability of “others” was significantly greater in the BPI PSA condition (M =

2.83, SD = 1.03) than in the control condition, t (113) = 1.841, p <.05.Perceived invulnerability

of “others” was also significantly greater in BPI PSA condition than in E-E narrative condition.

So, the BPI PSA condition was associated with the greatest perceived invulnerability of “others”

in three conditions.

4.2.4 Examining the Relationship between Perceived Persuasive Intent and Reactance

Hypothesis 7 predicted a positive association between perceived persuasive intent and

reactance in bipolar disorder E-E narrative condition. However, no significant positive

correlation was found between these two variables r(59) = -.11, p = .21, thus, H7 was not

supported.

49

4.2.5 Examining Narrative Involvement-Resistance Relationships in the BPI E-E Narrative

Condition

Hypothesis 8a posits that greater parasocial interaction with a main character who

experiences bipolar disorder would be associated with less reactance. Simply put, this hypothesis

predicted a negative association between parasocial interaction with the BPI character and

reactance to the underlying message in the narrative. Pearson’s r revealed moderate, negative

relationship between these two variables, r(59)= -.54, p <.01. Therefore, H8a was supported.

Hypothesis 8b predicted that parasocial interaction with a character whose significant

other experiences bipolar I disorder would be associated with less reactance to the underlying

Table 5

Summary of Bivariate Correlations in E-E BPI Narrative Condition

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1.PSR with Silver -- 2.PSR with Dixon .67** -- 3.Identification(Silver) .62** .44** -- 4.Identification(Dixon) .55** .78** .56** -- 5.Similarity(Silver) .47** .27* .65** .22* -- 6.Similarity(Dixon) .12 .38** .04 .28* -.04 -- 7.Transportation .48** .36** .46** .37** .29* .14 -- 8.Reactance -.54** -.43** -.47** -.54** -.26* -.14 -.48** -- 9.Counterarguing -.34** -.24* -.11 -.27* -.03 -.06 -.26* .54** -- 10.Invulnerability(self) -.28* -.34** -.25* -.30* -.02 .02 -.32** .09 .10 -- 11.Invulnerability(others) -.32** -.41** -.26* -.26* -.14 .08 -.21† .06 -.04 .78** -- 12.Intention(self) .06 .19† .12 .17† -.01 .07 .06 .01 .13 -.20† -.14 -- 13.Intention(others) .06 .04 .19† -.01 .03 -.18† -.04 .03 .21† -.25* -.47** .25* -- 14.Behavior .23* .25* .15 .20† .08 -.20† -.07 .15 .00 -.16 -.19† .03 -.04 -- 15.Persuasive Intent .14 .21† -.05 -.00 -.03 -.22* .01 -.11 -.13 .16 .07 .05 -.06 .14 -- Note. † p <.10, * p < .05, ** p < .01 (one-tailed), n = 59~61.

message of the narrative. This negative association was identified by a Pearson’s correlation test and was statistically significant, r(59) = -.43, p < .01. Accordingly, H8b was supported.

Hypothesis 9a predicted that identification with a main character who experiences bipolar

I disorder in a dramatic narrative program would be negatively associated with counterarguing.

50

However, Pearson correlation analysis showed no appreciable relationship between these two

variables, r(59) = -.11, p = .21. Therefore, H9a was not supported.

Hypothesis 9b posited that identification with a main character whose loved one is

experiencing BPI would be associated with less counterarguing. Pearson correlation showed a

weak negative relationship between these two variables, r(59) = -.27, p < .05. Thus, H9b was

supported.

Hypothesis 10 predicted that transportation into a dramatic narrative involving BPI would

be associated with less counterarguing. There was a weak negative relationship between

transportation and conterarguing, r(59) = -.26, p <.05. Therefore, H10 was supported.

Perceived -.11 Persuasive Intent

Reactance PSR with Silver/Dixon -.54**/-.43**

-.26* Transportation Counterarguing

-.11/-.27* Identification with Silver/Dixon -.25*/-.30*

Perceived Invulnerability of “self” Perceived -.02/.02 Similarity with

Silver/Dixon -.14/.08 Perceived Invulnerability of -.26*/-.26* “others” Figure 3. Involvement-Resistance Associations

Note. These data represent results of Pearson correlation tests, not results of structural equation or path model tests. Results are depicted in this way to facilitate comparisons across multiple hypotheses and ease of reading. Perceived Invulnerability of “others” is a variable created in study, it is not in the Entertainment Overcoming Resistance Model. * p < .05, ** p < .01.

51

Hypothesis 11a posited a negative association between identification with a main character experiencing bipolar disorder and perceived invulnerability. When testing this hypothesis, perceived invulnerability was examined separately for both the first-person aspect

(perceived invulnerability of “self”) and the third-person aspect (perceived invulnerability of someone one cares about). Pearson correlation tests revealed weak-to-moderate relationships.

Participant’s perceived invulnerability of “self” was negatively associated with identification with the main female character having bipolar disorder – Silver, r(59) = -.25, p < .05. When it comes to perceived invulnerability of “others”, there was a negative association between these two variables as well, r(59) = -.26, p < .05. Hence, H11a was supported.

Hypothesis 11b predicted that identification with a main character whose significant other experiences bipolar disorder would be negatively associated with perceived invulnerability.

The Pearson correlation analysis showed that identification with Dixon had a moderate negative relationship with perceived invulnerability of “self”, r(59) = -.30, p <.05. Identification with

Dixon also had a negative association with perceived invulnerability of “others” r(59) = -.26, p

<.05. Therefore, H11b was supported.

Hypothesis 12a predicted that perceived similarity to a character who experiences bipolar disorder would be associated with reduced perceived invulnerability. Using Pearson correlation tests, no significant relationships were found either between perceived similarity to the female main character with BPI and perceived invulnerability of “self”, r(59) = -.02, p = .44, or between perceived similarity to the female main character with BPI and perceived invulnerability of

“others”, r(59) = - .14, p = .15. Therefore, H12a was not supported.

Similarly, Hypothesis 12b predicted that perceived similarity to a character whose loved one experiences BPI would be negatively associated with perceived invulnerability. Again,

52

Pearson correlation tests did not find a significant relationship either between perceived

similarity to the supportive male main character and perceived invulnerability of “self”, r(59) =

.02, p = .44, or between perceived similarity to the male character and perceived invulnerability

of someone participants cared about r(59) = .08, p = .28. Hence, H12b was not supported.

4.2.6 Examining Resistance-Behavior Associations in the BPI E-E Narrative Condition

Hypothesis 13a predicted that reactance would be negatively associated with pro-

treatment behavioral intention involving bipolar disorder for “self”. However, no appreciable

relationship was found at the posttest between reactance and BPI-related pro-treatment behavioral intention for “self”, r(59) = .01, p = .48. As a result, H13a was not supported.

.01 Behavioral intention for .13 “self”

Behavioral intention for † Reactance .03 .21 Counterarguing “others” .15 .00 Information-seeking behavior concerning BPI at follow-up

Behavioral intention for “self” -.20† -.14 Perceived Perceived -.25* Behavioral intention for -.47** Invulnerability Invulnerability “others” of “self” -.16 of “others”

Information - seeking -.19† behavior concerning BPI at follow-up Figure 4. Resistance-Behavioral Outcome Associations

Note. These data represent results of Pearson correlation tests, not results of structural equation or path model tests. Results are depicted in this way to facilitate comparisons across multiple hypotheses and ease of reading. Perceived Invulnerability of “others” and behavioral intention for “others” are variables created in study, they are not in the Entertainment Overcoming Resistance Model. † p < .10, * p < .05, ** p < .01.

53

Hypothesis 13b predicted a negative association between reactance and pro-treatment behavioral intention for “others". Yet, no relationship was found between these two variables at the posttest, r(59) = .03, p = .40. Thus, H13b was not supported.

Hypothesis 13c posited that reactance would be negatively associated with self-reported information seeking behavior involving bipolar I disorder at the time of follow-up (two weeks after the posttest). A Pearson correlation test did not find a significant relationship between these two variables, r(59) = .15, p = .13. Thus, H13c was not supported.

Hypothesis 14a predicted that less counterarguing would be associated with greater pro- treatment behavioral intention for “self”. However, no appreciable relationship was found between counterarguing and pro-treatment behavioral intention at the posttest, r(59) = .13, p =

.16. Therefore, H14a was not supported.

Hypothesis 14b predicted a negative association between counterarguing and pro- treatment behavioral intention for “others”. However, no significant, negative correlation was found between counterarguing and pro-treatment behavioral intention regarding “others” r(59) =

.21, p = .05. Therefore, H14b was not supported.

H14c posited a negative association between counterarguing and BPI-related self- reported information-seeking behavior at the time of follow-up. However, a Pearson correlation test showed no appreciable relationship between these two variables, r(59) = .003, p = .49. Thus,

H14c was not supported.

Hypothesis 15a predicted that lower perceived invulnerability of “self” would be associated with greater BPI-related pro-treatment behavioral intention for “self”. At the posttest, a weak, negative correlation was found between perceived invulnerability of “self” and BPI- related pro-treatment behavioral intention for “self” (r = -.20. p = .06). The correlation was in

54

predicted direction, approaching the traditional level of statistical significance, p < .10. As a

result, H15a received limited support.

Hypothesis 15b predicted a negative association between perceived invulnerability of

“self” and pro-treatment behavioral intention regarding helping someone participants cared

about. A negative association between perceived invulnerability of “self” and pro-treatment

behavioral intention concerning “others” was found, r(59) = -.25, p < .05. Therefore, H15b was

supported.

Hypothesis 15c predicted that less perceived invulnerability of “self” would be associated

with greater BPI-related information-seeking behavior at the time of follow-up. A Pearson’s correlation test found no significant relationship between perceived invulnerability and information-seeking behavior, r(59) = -.16, p = .11. As a result, H15c was not supported.

H16a predicted a negative association between perceived invulnerability of “others” and pro-treatment behavioral intention for “self”. At the posttest, no statistically significant negative

relationship was found between perceived invulnerability of “others” and pro-treatment

behavioral intention for “self”, r(59) = -.14, p = .14. Consequently, H16a was not supported.

H16b posited a negative association between perceived invulnerability of “others” and

pro-treatment behavioral intention regarding BPI in terms of helping “others”. A Pearson’s

correlation test revealed a fair degree of negative association between perceived invulnerability

of “others” and pro-treatment behavioral intention concerning “others”, r(59) = -.47, p < .01.

H16b was, therefore, supported.

H16c predicted a negative association between perceived invulnerability of “others” and

information seeking reported by participants at the time of follow-up concerning bipolar

disorder. Perceived invulnerability of “others” and BPI-related information-seeking behavior

55

were slightly and negatively associated, r(59) = -.19, p = .07, with the correlation approaching the traditional level of statistical significance, p < .10. Hence, H16c received limited support.

56

CHAPTER FIVE

DISCUSSION

This study sought to examine several hypothesized relationships in the Entertainment

Overcoming Resistance Model through investigating the effects of exposure to an Entertainment-

Education narrative addressing bipolar I disorder compared to those of exposure a BPI public

service announcement on individual’s BPI-related behavioral intentions and behaviors.

Specifically, the persuasive outcomes measured were BPI-related information searching and professional help-seeking behavioral intentions immediately after exposure and bipolar disorder information seeking behaviors two weeks after exposure to the messages. For the first time, the

EORM was employed to explain the effects of an E-E television drama addressing a mental

illness. Given that the percentage of people who suffer from this mental illness is relatively small

in the population and that most people are still in the learning stage regarding this topic,

resistance to the underlying message of the narrative and to the explicit message of the PSA

could reasonably have been expected to be fairly weak in general, specifically for the two main

resistance constructs - reactance and counterarguing. Data analyses did show a fairly low level of

reactance (M = 3.00, SD = .95) and counterarguing (M = 2.55, SD = 1.02) across the two BPI

message conditions. Moreover, most participants in this sample did not have either direct or

indirect experience with bipolar I disorder, so it could have been foreseen that the behavioral

effects would be considerably small. Despite these somewhat predictable outcomes, this study

did have some interesting findings.

The E-E narrative addressing bipolar I disorder had stronger effects than the BPI PSA on pro-treatment behavioral intentions for “self”. This is in line with the EORM proposition.

Unexpectedly, exposure to non-BPI information (the control condition) was associated with

57

higher level of behavioral intention than BPI PSA exposure, which appears to indicate a boomerang effect of the explicit persuasive message. More interestingly, exposure to the BPI narrative and exposure to the non-BPI stimuli (the control condition) were associated with a same level of behavioral intentions for “self”, which suggests that this particular BPI E-E narrative had no appreciable effect on participants’ behavioral intention for “self”. However, in the supplemental test comparing the two aspects of behavioral intention for “self” among three conditions, behavioral intention of seeking professional help was significantly higher in the BPI

E-E narrative condition than in either one of the other two conditions. This result seems to suggest that this BPI E-E narrative was effective in increasing behavioral intention to seek help from mental health professionals when participants suspected that they were experiencing some

BPI symptoms, which is consistent with the underlying message of the E-E narrative and arguably the most important intention effect in E-E interventions addressing mental illnesses.

Given recent research has suggested that young people (especially males) held less positive attitudes toward seeking help from mental health professionals than order age groups (Gonzalez,

Alegria, & Prihoda, 2005), the finding here is especially noteworthy.

Since the three conditions were associated with the same level of behavioral intention for

“others” and information seeking behaviors, it can be inferred that the BPI E-E narrative did not affect participants’ behavioral intentions to help others or impact their information seeking behavior regarding BPI for both “self” and “others”. Of note, the least amount of behavioral intention for “self” in the BPI PSA condition seems to suggest that attempting to increase BPI knowledge and intentions via PSAs might be counterproductive.

The overall lack of effect from the E-E narrative could be attributed to the dramatic structure of the storyline and the lack of relevance to the participants in general. Even though the

58

BPI E-E narrative storyline showed a relatively accurate portrayal of BPI, it could have been

confusing to those participants who did not have precise knowledge about this mental illness to

view a character exhibiting many erratic behaviors within 20 minutes. For these participants, the

storyline could have been considered overly dramatic. On the other hand, it is also likely that

these individuals might not have been strongly affected by this narrative because they perceived

BPI as a matter irrelevant to them. After all, most of the participants did not have direct or

indirect experience with BPI.

Additionally, contrary to the predictions of the E-ELM and the EORM, reactance and

counterarguing were found to be higher in the BPI narrative condition than in the explicit

persuasive message condition (BPI PSA condition). There are some possible explanations for

this unexpected finding. First, the use of summary measures to test these two constructs in a

narrative context could have made the target of counterarguing and reactance unclear. Previous

studies have discussed this problem with closed-ended items when measuring these two

resistance constructs (Moyer-Gusé, 2010; Busselle & Bilandzic, 2013). With these measures, it

was difficult to identify what exactly were respondents counterarguing with or reacting to. For this particular narrative, rather than counterarguing with and reacting to the underlying

persuasive message of the narrative, participants could have been responding to reckless

behaviors performed by the main character or the irrational decisions made by her before the narrative revealed that she was living with BPI. Moreover, the narrative was much longer in time than the BPI PSA, and so it could have provided more targets for participants to respond negatively to than the short BPI PSA did.

Because reactance and counterarguing were both higher in the narrative condition than the PSA condition, a lower behavioral intention in the narrative condition could have been

59

expected, due to the seemingly high resistance to persuasive outcomes. Yet, as previously discussed, pro-treatment behavioral intention for “self” was actually higher in the narrative

condition. This conflicting finding made it necessary to consider another resistance-related

construct tested in this study: perceived invulnerability. Perceived invulnerability of “self” was

significantly lower in the E-E narrative condition than in the PSA condition. Perceived

invulnerability of “others” was also found to be lower in the BPI narrative condition than the BPI

PSA condition. In keeping with the EORM, less perceived invulnerability is associated with

higher behavioral outcomes. So, it appears that lower perceived invulnerability could account for the greater behavioral intention in the BPI narrative condition. At the same time, since reactance and counterarguing were both lower in the BPI PSA condition, the relatively high perceived invulnerability could also be an explanation for the low behavioral intention in the BPI PSA condition.

Given that BPI PSA condition was associated with the lowest pro-treatment behavioral

intention for “self” (significantly lower than control condition at p < .01 level) and the highest

perceived invulnerability of “others” among three conditions, exposure to the PSA seemed to

lead to the worst outcomes.

Another intriguing finding is that despite the lack of significant differences in both

perceived invulnerability of “self” and perceived invulnerability of “others” between the BPI

narrative condition and the non-BPI (control) condition, BPI narrative exposure was associated

with higher behavioral intention for seeking professional help for “self” than the exposure to

non-BPI message was. As discussed earlier, there was no significant difference in overall pro-

treatment behavioral intention (the measure included intention to seek more information and seek

help from a mental health professional) between the BPI narrative condition and the control

60

condition, which suggested that exposure to the BPI narrative had no effect on individuals’

behavioral intentions involving BPI. But a significant difference of behavioral intention to seeking help from mental health professional was found between these two conditions when the two aspects of behavioral intention for “self” were analyzed separately. However, this was in conflict with the lack of difference in perceived invulnerability between these two conditions.

Further scrutiny of the results of t-tests on perceived invulnerability of self between the BPI

narrative and the control conditions reveals that the mean differences identified by the one-tailed t-test was in the predicted direction (BPI narrative < Control) with significance value a little bit over the traditional level of statistical significance (p = .11). This could actually be a significant difference if more intended message recipients had been included in the sample or if there had been a bigger sample size. On the contrary, it is also likely that the higher behavioral intention of seeking professional help in the E-E narrative condition was not predicted by reduced perceived invulnerability, but by other factors such as increased knowledge of BPI or enhanced self- efficacy that were not hypothesized in this study.

In the BPI narrative condition, among the three resistance-related variables, only perceived invulnerability had a clear and significant negative association with pro-treatment behavioral intention. Specifically, both perceived invulnerability of “self” and perceived invulnerability of “others” were significantly associated with pro-treatment behavioral intention for “others”. The relationship between perceived invulnerability of “others” and behavioral intention for “others” was stronger than the association between perceived invulnerability of

“self” and behavioral intention for “others”. Perceived invulnerability of “self” had a slight negative association with behavioral intention for “self”, however, no association was found between perceived invulnerability of “self” and information-seeking behavior regarding BPI two

61

weeks after the E-E narrative exposure. Reduced perceived invulnerability of “others” was found

to be marginally associated (p = .07) with information-seeking behavior at the time of follow-up.

In summary, because the mental illness addressed in the E-E narrative only affects a small part of

the population, it seems that the additional consideration of a third-person aspect of perceived

invulnerability and behavioral intention was helpful in understanding the effect of this BPI E-E

narrative.

Perceived persuasive intent was not associated with reactance to the underlying message

in this study. This result is likely due to the consistently low reactance score in general and the

measure of perceived persuasive intent. The item measuring persuasive intent was a bipolar scale

ranging from “to persuade” to “to entertain”, and it was likely to generate polarized responses to

this measure. Since participants were aware of the type of program (PSA and TV drama) they

were going to view, it was likely very apparent to these undergraduates that a PSA is not created

to entertain, yet TV drama is commonly accepted as a form of entertainment media content.

Therefore, recognizing a drama as a form of entertainment might not be associated with less

reactance in this context.

Among the four narrative-involvement-related constructs – parasocial interaction with

characters, transportation, perceived similarity with characters, and identification with

characters, the negative association between parasocial interaction with characters and reactance

was supported, transportation was found to be negatively correlated with counterarguing (see

Table 4); and identification with characters predicted both perceived invulnerability of “self” and

perceived invulnerability of “others”. However, perceived similarity with characters failed to

predict either perceived vulnerability of “self” or perceived vulnerability of “others”. Participants

reported significantly more perceived similarity to “Dixon” than to “Silver”, which is probably

62

mainly because “Silver”, the BPI sufferer, was exhibiting various abnormal behaviors in the drama, and participants found it difficult to perceive similarity to her.

The negative relationship between identification with the character having BPI (“Silver”) and counterarguing was not supported. But identification with the main character’s boyfriend —

Dixon — was negatively associated with counterarguing. Similar to the explanation for low perceived similarity to “Silver”, the consistently low identification with “Silver”, the BPI sufferer, could account for this result.

By and large, the EORM hypothesized relationships between parasocial interaction and reactance, between transportation and counterarguing, and between identification and invulnerability were supported. The E-E narrative addressing bipolar I disorder did not affect individuals’ overall pro-treatment behavioral intention, but it enhanced one aspect of the behavioral intention – intention to seek help from mental health professional if they had BPI. The

BPI E-E narrative was more effective than the short BPI PSA in influencing perceived invulnerability of oneself and perceived invulnerability of someone they care about. Within the narrative condition, identification with characters was the only construct that contributed to the reduced perceived invulnerability. Contradicting to a hypothesized relationship in EORM, perceived similarity failed to predict reduced perceived invulnerability. This result could be attributed to the low perceived similarity to the main character who exhibited various BPI symptoms in the drama. On the other hand, the result that perceived similarity failed to predict perceived vulnerability was consistent with the findings of a previous study on EORM (Moyer-

Gusé & Nabi, 2010), in which the authors concluded that identification was more important than perceived similarity in general for persuasive effects (Moyer-Gusé & Nabi, 2010; Slater &

Rouner, 2002; Stephenson, 2003).

63

Another question of significance is whether the EORM is adaptable to mental illness contexts in general. The majority of the studies from which EORM was developed were conducted using issues more controversial and relatable than the particular mental illness addressed in the present study. Individuals are more likely to have pre-existing opinions and attitudes on those issues, for example, safe sex (Moyer-Gusé, 2007), unplanned teen pregnancy

(Moyer‐Gusé & Nabi, 2010), and drinking-and-driving (Moyer-Gusé et al., 2012). These issues addressed in past studies typically affect a larger population in a more direct manner than bipolar

I disorder does. Therefore, it is a possibility that the nature of the topic or issue affected the

results of this study. More research should be conducted to test the EORM on mental illness or

less familiar topics to find out whether the relationships in the model are influenced by the nature

of health issue.

Additionally, future research should develop more valid measures for reactance and

counterarguing to the underlying message of an E-E narrative. Perceived persuasive intent also

needs a better measure when it is compared between an E-E narrative and a traditional

persuasive message so that polarized responses could be avoided. A measure of perceived threat from reactance literature should be used to better predict reactance in future studies. Due to the

novelty of the mental illness topic issue and its inapplicability to the majority of the population, more research is needed to investigate the third person aspect of perceived invulnerability and persuasive outcomes in mental health contexts. The current study did not examine the attitudinal aspect of persuasive outcomes produced by the exposure to BPI messages. However, attitude change could be a meaningful outcome for E-E narrative addressing a mental illness considering its possible contributions to stigma reduction and increase in willingness to accept treatment among mental illness sufferers.

64

5.1 Limitations

This study is affected by several limitations. The employment of a convenience sample

— undergraduate students taking classes in school of communication — makes the findings of this study unable to be generalized to the greater population. College students may have different living environments, attitudes, and behaviors towards mental disorder than non-college students.

However, the age range of this sample did match the average age at onset of first bipolar I disorder episode.

The experimental setting was different from the natural setting in which participants watch television dramas and PSAs. They were confined in a small computer lab viewing videos with computers and headphones, which does not resemble the setting in which most people

watch a television program. This affects the external validity of the study.

Another limitation with this study is the small sample size. For a study examining effects

of an E-E narrative addressing an issue not relevant to the majority of the population, the effect

size was expected to be very small. A considerably larger sample size should have been utilized

to increase statistical power. Moreover, all of the narrative persuasion related hypotheses of this

study were tested in the E-E narrative group only (n = 61), so the statistical power suffered from

the small sample size.

Additionally, there was a drastic difference in the length of time for the two stimuli being

compared – the E-E narrative and the BPI PSA. Because of this difference, some measures such

as reactance and counterarguing could have been affected. Participants may have formed more

reactance to the lengthy narrative and counterarguing as opposed to the short public service

announcement.

Finally, the present study used a single exposure to intended message. It is possible that

E-E effects on perceptions and behaviors most likely to be produced by sustained presence of a

65

message over a long period of time or multiple exposures to the same message in different programs rendered by sustained E-E efforts (Hether et al, 2008; Piotrow & de Fossard, 2004).

66

APPENDIX A

IRB APPROVAL MEMORANDUM AND CONSENT FORM

67

INFORMED CONSENT STATEMENT Date of Approval: 09/13/2013 Assurance Number: IRB00000446

I freely and voluntarily and without element of force or coercion, consent to be a participant in the research project entitled Processing of TV Drama Content.

This research is being conducted by Danyang Zhao, a graduate student in the School of communication at Florida State University. I understand that the purpose of the research project is to better understand how viewers are affected by the content of television shows that they view. I understand that this study consists of two parts. The first part of the study is a laboratory session that will take me no more than 70 minutes to complete. The second part of the study is an online survey that will be sent to me in two weeks after I attend the first session, which will take no more than 10 minutes to complete. During the first part of the study I will view a public service announcement and answer a short questionnaire, then view a video of excerpts from a selected prime time TV drama series and answer questions about that experience and the characters involved in the narratives. I also understand I will be asked for demographic information. The demographic information will be used for classification purposes only. I will be asked some sensitive questions. I may find some of the questions being asked upsetting.

However, there is no penalty associated with failure to answer the questions about my family members’ or my friends’ mental health history. I understand that the media content may contain violence, sexual themes, or offensive language, but that it will be equivalent to TV series shown on prime time television that are TVMA rated. I understand that my participation in this study is voluntary, and I may decline to participate or choose leave the study without penalty (including no negative effect on your grade) at any time. However, in order to receive full amount of research credit, I will need to participate for both parts of the study and the full length of the research session. If I choose to not participate in this study, I will be offered an alternative

68

project that will award me the same amount of credit. My name will not appear on any of the results, nor will the ID code that I develop myself at the beginning of the study be in any way relatable to my identity. No individual responses will be reported. Only group findings will be reported. The information in the study records will be kept confidential to the extent allowed by law. Data will be stored securely in a locked desk where only the researcher has the key; this form will be stored in a location separate from the questionnaire data. Data will be destroyed within five years of publication of the results. I understand there are benefits for participating in this research project. For participating I will receive either (a) credit toward a class requirement or (b) extra credit for my class, the amount of which is determined by the instructor of the course from which I was recruited. I understand that different instructors give different types and amounts of credit for participating in this study. I acknowledge that if I have questions about the exact type or amount of credit that I can receive, then I can postpone and reschedule my participation without penalty. I understand that the researchers will communicate this information to my instructor, in plenty of time to have that credit count toward my class this semester. I also acknowledge that as a student at a research university, my benefit in participating in such a project is to get hands on understanding of how to investigate entertainment media in a scientific way. With my participation I will assist communication scholars in better understanding the phenomenon of entertainment media.

I understand that I may contact Ms. Danyang Zhao at [phone number and email address] for answers to questions about this research or my rights. Also, if I have any questions about my rights as a subject/participant in this research, or if I feel I have been placed at risk, I can contact the Chair of the Human Subjects Committee, Institutional Review Board, through the Vice

President for the Office of Research at (850) 644-8633.

69

Before you can continue with the survey you need to respond whether you understand and

agree with the consent form by ticking on one of the options below.

☐Yes, I understand and agree

☐No, I do not understand (please contact the researcher before you continue at [email

address])

☐No, I do not agree with the consent form (you will not be able to continue. Please

contact the researcher if you have any questions at [email address])

If you agree to participate in the survey described above, please sign your name.

______

Participant Signature Date

70

APPENDIX B

INSTRUCTIONS

Instructions (On Screen)

• Before you begin, please make sure your cell phones are turned off.

• You are going to watch a public service announcement (1.5 minutes) and a 30-minute

video of excerpts taken from a TV drama.

• You will be guided to answer a short questionnaire on your computer screen after

viewing the public service announcement and another questionnaire after watching the

30-minute video clip.

• Please read all the instructions as you proceed and keep in mind that your answers are

anonymous and there are no right or wrong answers, so please provide honest answers.

• When you have finished the second questionnaire, you are free to go.

• There is still one more part of this study that you must complete in order to secure the

extra or research requirement credit. You will receive an email in about 2 weeks

containing a link to an online survey. Upon completing the 5 minute online survey you

will then have completed your participation of this study and will be able to get your

credit.

71

APPENDIX C

MEASURES

POSTTEST QUESTIONNAIRE FOR EXPERIMENTAL GROUP (EPA PSA + BPI Narrative) QUESTIONNAIRE - PART I Before you begin, we need you to provide us a confidential code so that we can match this questionnaire with the one you will fill out in two weeks following an email instruction. PLEASE NOTE: IT IS IMPORTANT THAT YOU USE THE SAME CODE EACH TIME. To make it easier to remember, please use: a. The LAST TWO digits of your current cell phone number e.g.: (850) 537-2821, the last two digits would be 21 b. The FIRST letter of your FIRST name (NOT nickname) e.g.: Stacey, the first letter would be s c. The two-digit MONTH of your birthday e.g.: 09/06/1992, the two-digit month would be 09 d. The last TWO letters of your LAST name e.g.: Young, the last two letters would be ng As an example, if your phone number was (850) 537-2821, your first name was Stacey, you were born in September, and your last name was Young, your confidential code would be 21s09ng. Please type your code below (PLEASE ASK IF YOU HAVE ANY QUESTIONS). Your Code: ______Please provide your official FSU e-mail address (e.g.: [email protected]) below for the online questionnaire two weeks later (PLEASE NOTE: Your email address will be separated from the current questionnaire to make sure you are anonymous as a research participant). ______INSTRUCTIONS. Please think about the program you just watched and answer the following questions. 1. While watching the video, how much did you feel each of the following? Not at all Very much Angry 1 2 3 4 5 6 7 Irritated 1 2 3 4 5 6 7 Annoyed 1 2 3 4 5 6 7 Aggravated 1 2 3 4 5 6 7 2. The video was reasonable. SD D N A SA 3. The video was interesting. SD D N A SA 4. The video was pleasant. SD D N A SA 5. The video was not exaggerated. SD D N A SA 6. The video reflected my attitudes. SD D N A SA

72

7. The content of the video mattered to me. SD D N A SA 8. Do you think the video you just watched was created more to entertain or more to persuade? To persuade 1 2 3 4 5 6 7 To entertain INSTRUCTIONS. Using the scale below, please respond by clicking on the phrase that best reflects your feeling about each statement. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree

9. While watching the video, I sometimes felt like I wanted to “argue back” to what was going on onscreen. SD D N A SA 10. While watching the video, I sometimes found myself thinking of ways I disagreed with what was being presented. SD D N A SA 11. While watching the video, I couldn’t help thinking about ways that the information being presented was inaccurate or misleading. SD D N A SA 12. I found myself looking for flaws in the way information was presented in the video. SD D N A SA INSTRUCTIONS. Now, we would like to ask you some questions about your perception of risk involving stormwater runoff. 1. In general, how dangerous do you think stormwater runoff would be for you if you do not do anything about it? Extremely 1 2 3 4 5 6 7 Not at all Dangerous Dangerous  2. In general, what do you think are your chances of suffering from the consequences of stormwater runoff during your lifetime? No Chance 1 2 3 4 5 6 7 Definitely would happen  3. What do you think are the chances that you would contract any type of illnesses caused by stormwater runoff during your lifetime? No Chance 1 2 3 4 5 6 7 Definitely would happen  4. If your local area had water pollution because of stormwater runoff and no one had done anything to reduce it, what do you think the chances are that it would get worse? No Chance 1 2 3 4 5 6 7 Definitely would happen  5. How likely do you think it is that you would look for more information about stormwater runoff? Very unlikely 1 2 3 4 5 6 7 Very likely 6. How likely do you think it is that you would go to the website to download the stormwater calculator shown in the video clip? Very unlikely 1 2 3 4 5 6 7 Very likely

73

7. How likely do you think it is that you would go to the website shown in the video to download the stormwater calculator if you experienced a water pollution problem caused by stormwater runoff? Very unlikely 1 2 3 4 5 6 7 Very likely 8. How likely is that you would refer your friends and family members to the website shown in the video to download the stormwater calculator? Very unlikely 1 2 3 4 5 6 7 Very likely 9. How likely is it that you would talk to your family members about adding green features to your home to help prevent pollution by controlling stormwater runoff? Very unlikely 1 2 3 4 5 6 7 Very likely 10. How likely do you think you would be to encourage your friends and neighbors to add green features to their homes to control stormwater runoff? Very unlikely 1 2 3 4 5 6 7 Very likely POSTTEST QUESTIONNAIRE - PART II INSTRUCTIONS. Please think about the program you just watched and answer the following questions. 1. While watching the video, how much did you feel each of the following? Not at all Very much Angry 1 2 3 4 5 6 7 Irritated 1 2 3 4 5 6 7 Annoyed 1 2 3 4 5 6 7 Aggravated 1 2 3 4 5 6 7 2. The video was reasonable. SD D N A SA 3. The video was interesting. SD D N A SA 4. The video was pleasant. SD D N A SA 5. The video was not exaggerated. SD D N A SA 6. The video reflected my attitudes. SD D N A SA 7. The content of the video mattered to me. SD D N A SA INSTRUCTIONS. Using the scale below, please respond by clicking on the phrase that best reflects your feeling about each statement. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree

1. When I was watching this program, I felt like I was really one of the people in the story. SD D N A SA  2. While I was watching this program, activity going on in the room around me was on my mind.

74

SD D N A SA 3. I was mentally involved in the show while watching it. SD D N A SA  4. After the show was over, I found it easy to put it out of my mind. SD D N A SA 5. While I was watching, I was interested to learn how the show would end. SD D N A SA 6. The show affected me emotionally. SD D N A SA  7. I found myself thinking of ways the show could have turned out differently. SD D N A SA  8. I found my mind wandering while watching the show. SD D N A SA 9. I enjoyed the television show I watched in the experiment very much. SD D N A SA 10. This is a TV drama that I can enjoy. SD D N A SA  11. Do you think the video you just watched was created more to entertain or more to persuade? To persuade 1 2 3 4 5 6 7 To entertain INSTRUCTIONS. Think about the characters Silver and Dixon from the video you saw. Click on the phrase that best represents your answer. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree 1. Silver… is like me SD D N A SA

is similar to me SD D N A SA

thinks like me SD D N A SA

behaves like me SD D N A SA

2. Dixon… is like me SD D N A SA

is similar to me SD D N A SA

thinks like me SD D N A SA

behaves like me SD D N A SA Now think just about the character Silver, from the video you just watched. For each statement, please use the following scale and click on the phrase that best expresses your own feelings about Silver.

75

SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree

1. I think I have a good understanding of Silver. SD D N A SA 2. I tend to understand the reasons why Silver did what she did. SD D N A SA 3. While viewing the show, I could feel the emotions Silver portrayed SD D N A SA 4. At key moments in the show, I felt I knew exactly what Silver was going through. SD D N A SA 5. When I watched Silver on the program, I felt I understood the way she felt. SD D N A SA 6. When I watched Silver on the program, I imagined myself doing the same thing she was doing. SD D N A SA 7. When I watched Silver on the program, I really felt as if I were one of the people taking part in the action. SD D N A SA 8. When I watched Silver, I felt like I was a part of her social world. SD D N A SA 9. Silver makes me feel comfortable, like I’m with a friend. SD D N A SA 10. I look forward to watching Silver when her show is on. SD D N A SA 11. If Silver appeared on another show, I would want to watch it. SD D N A SA 12. I see Silver as a natural, down-to-earth person. SD D N A SA 13. Silver seems to understand the kinds of things I want to know. SD D N A SA 14. If I saw a story about Silver in a newspaper or magazine, I would want to read it. SD D N A SA 15. I miss seeing Silver when her show isn’t on for some reason. SD D N A SA 16. If I could, I would like to meet Silver in person. SD D N A SA 17. When Silver shows me how she feels about an issue, it helps me make up my own mind about the issue. SD D N A SA 18. I like to compare my ideas with what Silver says. SD D N A SA 19. I feel sorry for Silver when she makes a mistake.

76

SD D N A SA Now, consider the character Dixon, from the video you just watched. For each statement, please click on the phrase that best expresses your own feelings about this Dixon. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree

1. I think I have a good understanding of Dixon. SD D N A SA 2. I tend to understand the reasons why Dixon did what he did. SD D N A SA 3. While viewing the show, I could feel the emotions Dixon portrayed SD D N A SA 4. At key moments in the show, I felt I knew exactly what Dixon was going through. SD D N A SA 5. When I watched Dixon on the program, I felt I understood the way he felt. SD D N A SA 6. When I watched Dixon on the program, I imagined myself doing the same thing he was doing. SD D N A SA 7. When I watched Dixon on the program, I really felt as if I were one of the people taking part in the drama. SD D N A SA 8. When I watched Dixon, I felt like I was a part of his social world. SD D N A SA 9. Dixon makes me feel comfortable, like I’m with a friend. SD D N A SA 10. I look forward to watching Dixon when his show is on. SD D N A SA 11. If Dixon appeared on another show, I would want to watch it. SD D N A SA 12. I see Dixon as a natural, down-to-earth person. SD D N A SA 13. Dixon seems to understand the kinds of things I want to know. SD D N A SA 14. If I saw a story about Dixon in a newspaper or magazine, I would want to read it. SD D N A SA 15. I miss seeing Dixon when his show isn’t on for some reason. SD D N A SA 16. If I could, I would like to meet Dixon in person. SD D N A SA 17. When Dixon shows me how he feels about an issue, it helps me make up my own mind about the issue. SD D N A SA

77

18. I like to compare my ideas with what Dixon says. SD D N A SA 19. I feel sorry for Dixon when he makes a mistake. SD D N A SA INSTRUCTIONS. Using the scale below, please respond by clicking on the phrase that best reflects your feeling about each statement. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree

1. While watching the video, I sometimes felt like I wanted to “argue back” to what was going on onscreen. SD D N A SA 2. While watching the video, I sometimes found myself thinking of ways I disagreed with what was being presented. SD D N A SA 3. While watching the video, I couldn’t help thinking about ways that the information being presented was inaccurate or misleading. SD D N A SA 4. I found myself looking for flaws in the way information was presented in the video. SD D N A SA INSTRUCTIONS. Now, we would like to ask you some questions about your perception of risk involving bipolar disorder. 1. In general, how dangerous do you think bipolar disorder would be for you if you did not get treatment? Extremely Dangerous 1 2 3 4 5 6 7 Not at all dangerous 8 Prefer not to answer  2. In general, what do you think are your chances of developing bipolar disorder during your lifetime? No Chance 1 2 3 4 5 6 7 Definitely would happen 8 Prefer not to answer  3. If you had bipolar disorder and did not seek treatment, what do you think the chances are that you would get worse? No Chance 1 2 3 4 5 6 7 Definitely would happen 8 Prefer not to answer  4. If you had bipolar disorder and did not receive treatment, what do you think the chances are that you would end up doing something self-destructive? No Chance 1 2 3 4 5 6 7 Definitely would happen 8 Prefer not to answer 5. How likely is it that you would search for information about bipolar disorder if you thought you had bipolar disorder? Very unlikely 1 2 3 4 5 6 7 Very likely 8 Prefer not to answer

78

6. How likely do you think it is that you would seek help from mental health professionals if you experienced bipolar disorder symptom (s)? Very unlikely 1 2 3 4 5 6 7 Very likely 8 Prefer not to answer Now think about people you care about and their risk involving bipolar disorder. 1. In general, how dangerous do you think bipolar disorder would be for him/her/them if he/she/they did not get treatment? Extremely 1 2 3 4 5 6 7 Not at all Dangerous Dangerous  2. In general, what do you think are his/her/their chances of developing bipolar disorder during his/her/their lifetime? No Chance 1 2 3 4 5 6 7 Definitely would happen  3. If he/she/they had bipolar disorder and did not seek treatment, what do you think the chances are that he/she/they would get worse? No Chance 1 2 3 4 5 6 7 Definitely would happen  4. If he/she/they had bipolar disorder and did not receive treatment, what do you think the chances are that he/she/they would end up doing something self-destructive? No Chance 1 2 3 4 5 6 7 Definitely would happen 5. How likely do you think you would help him/her/them search for information about bipolar disorder if you thought he/she/they had bipolar disorder? Very unlikely 1 2 3 4 5 6 7 Very likely 6. How likely do you think you would be to encourage him/her/them to seek help by contacting mental health professionals if you thought she/he/they had bipolar disorder? Very unlikely 1 2 3 4 5 6 7 Very likely INSTRUCTIONS. Now, please click on the phrase that best represents how you feel. 1. Many people like me have bipolar disorder. SD D N A SA 2. Many college students have bipolar disorder. SD D N A SA 3. Many young people like me have bipolar disorder. SD D N A SA 4. Many Americans have bipolar disorder. SD D N A SA 5. Most people with bipolar disorder seek treatment. SD D N A SA 6. Most people with bipolar disorder have friends who support them. SD D N A SA 7. Most people with bipolar disorder have family members who support them. SD D N A SA 8. Bipolar disorder is a common disease. SD D N A SA 9. Bipolar disorder affects many people. SD D N A SA Now, please respond to the following questions about the television drama – 90210. 1. Can you recall ever seeing this episode before you saw it today? YES NO 2. How often do you watch 90210?

79

Never Some weeks Most weeks Every week 3. Have you seen an episode of 90210 in the last month? YES NO 4. Have you seen the character, Silver in any episode of 90210 in the last month? YES NO 5. About how old is the character Silver in the episode you just watched today? ______6. About how old is the character Dixon in the episode you just watched today? ______INSTRUCTIONS. Now we would like to ask you some questions about your experience involving mental illness. 1. Have you or anyone you care about ever been diagnosed with bipolar disorder? Yes No Prefer not to answer 2. Approximately how many people do you know who suffer from bipolar disorder? ______3. Have you ever…(check all that apply) □ looked for information about treatment of Bipolar Disorder for yourself or someone else □ made an appointment with a mental health professional to discuss Bipolar Disorder for yourself or someone else □ visited some Bipolar Disorder organizations’ websites for yourself or someone else □ thought about preventing Bipolar Disorder for yourself or someone else □ None of the above 4. About how many TIMES have you searched for information about bipolar disorder for yourself or someone else?

Number of times______5. About how many PEOPLE have you discussed bipolar disorder with? ______6. Have you ever talked to any mental health professionals? ______Yes ______No Please answer the following questions about yourself. Age in years: ______Do you consider your ethnicity to be primarily: □ African-American □ Asian-American □ Caucasian □ Hispanic □ Native American □ Other (Please specify) ______Gender □ Male □ Female

Thank you! You have completed Step 1 of this two-part study. To complete your participation, please complete the next very brief online questionnaire. You will receive an email with a link to the next questionnaire in about 2 weeks.

80

POSTTEST QUESTIONNAIRE FOR BPI PSA GROUP (BPI PSA + Control Narrative) QUESTIONNAIRE - PART I Before you begin, we need you to provide us a confidential code so that we can match this questionnaire with the one you will fill out in two weeks following an email instruction. PLEASE NOTE: IT IS IMPORTANT THAT YOU USE THE SAME CODE EACH TIME. To make it easier to remember, please use: a. The LAST TWO digits of your current cell phone number e.g.: (850) 537-2821, the last two digits would be 21 b. The FIRST letter of your FIRST name (NOT nickname) e.g.: Stacey, the first letter would be s c. The two-digit MONTH of your birthday e.g.: 09/06/1992, the two-digit month would be 09 d. The last TWO letters of your LAST name e.g.: Young, the last two letters would be ng As an example, if your phone number was (850) 537-2821, your first name was Stacey, you were born in September, and your last name was Young, your confidential code would be 21s09ng. Please type your code below (PLEASE ASK IF YOU HAVE ANY QUESTIONS). Your Code: ______Please provide your official FSU e-mail address (e.g.: [email protected]) below for the online questionnaire two weeks later (PLEASE NOTE: Your email address will be separated from the current questionnaire to make sure you are anonymous as a research participant). ______INSTRUCTIONS. Please think about the program you just watched and answer the following questions. 1. While watching the video, how much did you feel each of the following? Not at all Very much Angry 1 2 3 4 5 6 7 Irritated 1 2 3 4 5 6 7 Annoyed 1 2 3 4 5 6 7 Aggravated 1 2 3 4 5 6 7 2. The video was reasonable. SD D N A SA 3. The video was interesting. SD D N A SA 4. The video was pleasant. SD D N A SA 5. The video was not exaggerated. SD D N A SA 6. The video reflected my attitudes. SD D N A SA 7. The content of the video mattered to me. SD D N A SA  Do you think the video you just watched was created more to entertain or more to persuade? To persuade 1 2 3 4 5 6 7 To entertain

81

INSTRUCTIONS. Using the scale below, please respond by clicking on the phrase that best reflects your feeling about each statement. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree

1. While watching the video, I sometimes felt like I wanted to “argue back” to what was going on onscreen. SD D N A SA 2. While watching the video, I sometimes found myself thinking of ways I disagreed with what was being presented. SD D N A SA 3. While watching the video, I couldn’t help thinking about ways that the information being presented was inaccurate or misleading. SD D N A SA 4. I found myself looking for flaws in the way information was presented in the video. SD D N A SA INSTRUCTIONS. Now, we would like to ask you some questions about your perception of risk involving bipolar disorder. 1. In general, how dangerous do you think bipolar disorder would be for you if you did not get treatment? Extremely dangerous 1 2 3 4 5 6 7 Not at all dangerous 8 Prefer not to answer  2. In general, what do you think are your chances of developing bipolar disorder during your lifetime? No Chance 1 2 3 4 5 6 7 Definitely would happen 8 Prefer not to answer  3. If you had bipolar disorder and did not seek treatment, what do you think the chances are that you would get worse? No Chance 1 2 3 4 5 6 7 Definitely would happen 8 Prefer not to answer  4. If you had bipolar disorder and did not receive treatment, what do you think the chances are that you would end up doing something self-destructive? No Chance 1 2 3 4 5 6 7 Definitely would happen 8 Prefer not to answer 5. How likely is it that you would search for information about bipolar disorder if you thought you had bipolar disorder? Very unlikely 1 2 3 4 5 6 7 Very likely 8 Prefer not to answer 6. How likely do you think it is that you would seek help from mental health professionals if you experienced bipolar disorder symptom (s)? Very unlikely 1 2 3 4 5 6 7 Very likely 8 Prefer not to answer Now think about people you care about and their risk involving bipolar disorder.

82

1. In general, how dangerous do you think bipolar disorder would be for him/her/them if he/she/they did not get treatment? Extremely 1 2 3 4 5 6 7 Not at all Dangerous Dangerous  2. In general, what do you think are his/her/their chances of developing bipolar disorder during his/her/their lifetime? No Chance 1 2 3 4 5 6 7 Definitely would happen  3. If he/she/they had bipolar disorder and did not seek treatment, what do you think the chances are that he/she/they would get worse? No Chance 1 2 3 4 5 6 7 Definitely would happen  4. If he/she/they had bipolar disorder and did not receive treatment, what do you think the chances are that he/she/they would end up doing something self-destructive? No Chance 1 2 3 4 5 6 7 Definitely would happen 5. How likely do you think you would help him/her/them search for information about bipolar disorder if you thought he/she/they had bipolar disorder? Very unlikely 1 2 3 4 5 6 7 Very likely 6. How likely do you think you would be to encourage him/her/them to seek help by contacting mental health professionals if you thought she/he/they had bipolar disorder? Very unlikely 1 2 3 4 5 6 7 Very likely INSTRUCTIONS. Now, please click on the phrase that best represents how you feel. 1. Many people like me have bipolar disorder. SD D N A SA 2. Many college students like me have bipolar disorder. SD D N A SA 3. Many young people like me have bipolar disorder. SD D N A SA 4. Many Americans like me have bipolar disorder. SD D N A SA 5. Most people with bipolar disorder seek treatment. SD D N A SA 6. Most people with bipolar disorder have friends who support them. SD D N A SA 7. Most people with bipolar disorder have family members who support them. SD D N A SA 8. Bipolar disorder is a common disease. SD D N A SA 9. Bipolar disorder affects many people. SD D N A SA INSTRUCTIONS. Now we would like to ask you some questions about your experience involving mental illness. 1. Have you or anyone you care about ever been diagnosed with bipolar disorder? Yes No Prefer not to answer 2. Approximately how many people do you know who suffer from bipolar disorder? ______3. Have you ever…(check all that apply)

83

□ looked for information about treatment of Bipolar Disorder for yourself or someone else □ made an appointment with a mental health professional to discuss Bipolar Disorder for yourself or someone else □ visited some Bipolar Disorder organizations’ websites for yourself or someone else □ thought about preventing Bipolar Disorder for yourself or someone else □ None of the above 4. About how many TIMES have you searched for information about bipolar disorder for yourself or someone else? Number of times______5. About how many PEOPLE have you discussed bipolar disorder with? ______6. Have you ever talked to any mental health professionals? ______Yes ______No POSTTEST QUESTIONNAIRE - PART II INSTRUCTIONS. Using the scale below, please respond by clicking on the phrase that best reflects your feeling about each statement. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree

1. When I was watching this program, I felt like I was really one of the people in the story. SD D N A SA  2. While I was watching this program, activity going on in the room around me was on my mind. SD D N A SA 3. I was mentally involved in the show while watching it. SD D N A SA  4. After the show was over, I found it easy to put it out of my mind. SD D N A SA 5. While I was watching, I was interested to learn how the show would end. SD D N A SA 6. The show affected me emotionally. SD D N A SA  7. I found myself thinking of ways the show could have turned out differently. SD D N A SA  8. I found my mind wandering while watching the show. SD D N A SA 9. I enjoyed the program I watched in the experiment very much. SD D N A SA 10. This is a TV drama that I can enjoy. SD D N A SA  11. Do you think the video you just watched was created more to entertain or more to persuade?

84

To persuade 1 2 3 4 5 6 7 To entertain INSTRUCTIONS. Think about the characters Silver and Dixon from the show you saw. Click on the phrase that best represents your answer. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree 1. Silver… is like me SD D N A SA

is similar to me SD D N A SA

thinks like me SD D N A SA

behaves like me SD D N A SA

2. Dixon… is like me SD D N A SA

is similar to me SD D N A SA

thinks like me SD D N A SA

behaves like me SD D N A SA Now think just about the character Silver, from the show you just watched. For each statement, please use the following scale and click on the phrase that best expresses your own feelings about Silver. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree 1. I think I have a good understanding of Silver. SD D N A SA 2. I tend to understand the reasons why Silver did what she did. SD D N A SA 3. While viewing the show, I could feel the emotions Silver portrayed SD D N A SA 4. At key moments in the show, I felt I knew exactly what Silver was going through. SD D N A SA 5. When I watched Silver on the program, I felt I understood the way she felt. SD D N A SA 6. When I watched Silver on the program, I imagined myself doing the same thing she was doing. SD D N A SA

85

7. When I watched Silver on the program, I really felt as if I were one of the people taking part in the action. SD D N A SA 8. When I watched Silver, I felt like I was a part of her social world. SD D N A SA 9. Silver makes me feel comfortable, like I’m with a friend. SD D N A SA 10. I look forward to watching Silver when her show is on. SD D N A SA 11. If Silver appeared on another show, I would want to watch it. SD D N A SA 12. I see Silver as a natural, down-to-earth person. SD D N A SA 13. Silver seems to understand the kinds of things I want to know. SD D N A SA 14. If I saw a story about Silver in a newspaper or magazine, I would want to read it. SD D N A SA 15. I miss seeing Silver when her show isn’t on for some reason. SD D N A SA 16. If I could, I would like to meet Silver in person. SD D N A SA 17. When Silver shows me how she feels about an issue, it helps me make up my own mind about the issue. SD D N A SA 18. I like to compare my ideas with what Silver says. SD D N A SA 19. I feel sorry for Silver when she makes a mistake. SD D N A SA Now, consider the character Dixon, from the show you just watched. For each statement, please click on the phrase that best expresses your own feelings about this Dixon. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree 1. I think I have a good understanding of Dixon. SD D N A SA 2. I tend to understand the reasons why Dixon did what he did. SD D N A SA 3. While viewing the show, I could feel the emotions Dixon portrayed SD D N A SA 4. At key moments in the show, I felt I knew exactly what Dixon was going through. SD D N A SA 5. When I watched Dixon on the program, I felt I understood the way he felt. SD D N A SA

86

6. When I watched Dixon on the program, I imagined myself doing the same thing he was doing. SD D N A SA 7. When I watched Dixon on the program, I really felt as if I were one of the people taking part in the drama. SD D N A SA 8. When I watched Dixon, I felt like I was a part of his social world. SD D N A SA 9. Dixon makes me feel comfortable, like I’m with a friend. SD D N A SA 10. I look forward to watching Dixon when his show is on. SD D N A SA 11. If Dixon appeared on another show, I would want to watch it. SD D N A SA 12. I see Dixon as a natural, down-to-earth person. SD D N A SA 13. Dixon seems to understand the kinds of things I want to know. SD D N A SA 14. If I saw a story about Dixon in a newspaper or magazine, I would want to read it. SD D N A SA 15. I miss seeing Dixon when his show isn’t on for some reason. SD D N A SA 16. If I could, I would like to meet Dixon in person. SD D N A SA 17. When Dixon shows me how he feels about an issue, it helps me make up my own mind about the issue. SD D N A SA 18. I like to compare my ideas with what Dixon says. SD D N A SA 19. I feel sorry for Dixon when he makes a mistake. SD D N A SA Now, please respond to the following questions about the television drama – 90210. 1. Can you recall ever seeing this episode before you saw it today? YES NO 2. How often do you watch 90210? Never Some weeks Most weeks Every week 3. Have you seen an episode of 90210 in the last month? YES NO 4. Have you seen the character, Silver in any episode of 90210 in the last month? YES NO 5. About how old is the character Silver in the episode you just watched today? ______6. About how old is the character Dixon in the episode you just watched today? ______INSTRUCTIONS. Now we would like to ask you some questions about your attitudes toward discussion of problems in a relationship. Read each item, then click on the phrase that best reflects how you feel about important relationships in your life (current, past, or future). SD = strongly disagree

87

D = disagree N = neither agree nor disagree A = agree SA = strongly agree 1. I believe that it is very important to a relationship that my partner share his/her worries with me. SD D N A SA 2. It is crucial that partners have similar values, morals and religious beliefs. SD D N A SA 3. It is very important to a relationship that partners agree on our plans for activities and entertainment. SD D N A SA 4. I believe the more affection and intimacy in a relationship, the healthier the relationship will be. SD D N A SA 5. My partner should always be supportive of things that I do. SD D N A SA 6. My partner should encourage me to try new things. SD D N A SA 7. I would be worried about my relationship if my partner did not listen when I have something on my mind. SD D N A SA 8. Most of my peers think waiting until marriage before having sexual intercourse is not a good idea. SD D N A SA 9. My partner and I should not engage in sexual intercourse until we are in a relationship with someone we feel we would like to marry. SD D N A SA 10. Sex is not fun if a contraceptive is used. SD D N A SA 11. Using birth control makes sex less romantic. SD D N A SA Now, please respond to the following questions about the television drama – 90210. 1. Can you recall ever seeing this episode before you saw it today? YES NO 2. How often do you watch 90210? Never Some weeks Most weeks Every week 3. Have you seen an episode of 90210 in the last month? YES NO 4. Have you seen the character, Silver in any episode of 90210 in the last month? YES NO 5. About how old is the character Silver in the episode you just watched today? ______6. About how old is the character Dixon in the episode you just watched today? ______Please answer the following questions about yourself. Age in years: ______Do you consider your ethnicity to be primarily: □ African-American

88

□ Asian-American □ Caucasian □ Hispanic □ Native American □ Other (Please specify) ______Gender □ Male □ Female

Thank you! You have completed Step 1 of this two-part study. To complete your participation, please complete a very brief online questionnaire. You will receive an email with this link in about 2 weeks.

POSTTEST QUESTIONNAIRE FOR ENTERTAINMENT GROUP (Stormwater PSA + Control Narrative) QUESTIONNAIRE - PART I Before you begin, we need you to provide us a confidential code so that we can match this questionnaire with the one you will fill out in two weeks following an email instruction. PLEASE NOTE: IT IS IMPORTANT THAT YOU USE THE SAME CODE EACH TIME. To make it easier to remember, please use: a. The LAST TWO digits of your current cell phone number e.g.: (850) 537-2821, the last two digits would be 21 b. The FIRST letter of your FIRST name (NOT nickname) e.g.: Stacey, the first letter would be s c. The two-digit MONTH of your birthday e.g.: 09/06/1992, the two-digit month would be 09 d. The last TWO letters of your LAST name e.g.: Young, the last two letters would be ng As an example, if your phone number was (850) 537-2821, your first name was Stacey, you were born in September, and your last name was Young, your confidential code would be 21s09ng. Please type your code below (PLEASE ASK IF YOU HAVE ANY QUESTIONS). Your Code: ______Please provide your official FSU e-mail address (e.g.: [email protected]) below for the online questionnaire two weeks later (PLEASE NOTE: Your email address will be separated from the current questionnaire to make sure you are anonymous as a research participant). ______INSTRUCTIONS. Please think about the program you just watched and answer the following questions. 1. While watching the video, how much did you feel each of the following? Not at all Very much Angry 1 2 3 4 5 6 7 Irritated 1 2 3 4 5 6 7

89

Annoyed 1 2 3 4 5 6 7 Aggravated 1 2 3 4 5 6 7 2. The video was reasonable. SD D N A SA 3. The video was interesting. SD D N A SA 4. The video was pleasant. SD D N A SA 5. The video was not exaggerated. SD D N A SA 6. The video reflected my attitudes. SD D N A SA 7. The content of the video mattered to me. SD D N A SA  Do you think the video you just watched was created more to entertain or more to persuade? To persuade 1 2 3 4 5 6 7 To entertain INSTRUCTIONS. Using the scale below, please respond by clicking on the phrase that best reflects your feeling about each statement. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree 1. While watching the video, I sometimes felt like I wanted to “argue back” to what was going on onscreen. SD D N A SA 2. While watching the video, I sometimes found myself thinking of ways I disagreed with what was being presented. SD D N A SA 3. While watching the video, I couldn’t help thinking about ways that the information being presented was inaccurate or misleading. SD D N A SA 4. I found myself looking for flaws in the way information was presented in the video. SD D N A SA INSTRUCTIONS. Now, we would like to ask you some questions about your perception of risk involving stormwater runoff. 1. In general, how dangerous do you think stormwater runoff would be for you if you do not do anything about it?

Extremely 1 2 3 4 5 6 7 Not at all Dangerous Dangerous 2. In general, what do you think are your chances of suffering from the consequences of stormwater runoff during your lifetime? No Chance 1 2 3 4 5 6 7 Definitely would happen

90

3. What do you think are the chances that you would contract any type of illnesses caused by stormwater runoff during your lifetime? No Chance 1 2 3 4 5 6 7 Definitely would happen 4. If your local area had severe water pollution because of stormwater runoff and no one had done anything to reduce it, what do you think the chances are that it would get worse? No Chance 1 2 3 4 5 6 7 Definitely would happen 5. How likely do you think it is that you would look for more information about stormwater runoff? Very unlikely 1 2 3 4 5 6 7 Very likely 6. How likely do you think it is that you would go to the website to download the stormwater calculator shown in the video clip? Very unlikely 1 2 3 4 5 6 7 Very likely 7. How likely do you think it is that you would go to the website shown in the video to download the stormwater calculator if you experienced a water pollution problem caused by stormwater runoff? Very unlikely 1 2 3 4 5 6 7 Very likely 8. How likely is that you would refer your friends and family members to the website shown in the video to download the stormwater calculator? Very unlikely 1 2 3 4 5 6 7 Very likely 9. How likely is it that you would talk to your family members about adding green features to your home to help prevent pollution by controlling stormwater runoff? Very unlikely 1 2 3 4 5 6 7 Very likely 10. How likely do you think you would encourage your friends and neighbors to add green features to their homes to control stormwater runoff? Very unlikely 1 2 3 4 5 6 7 Very likely POSTTEST QUESTIONNAIRE - PART II INSTRUCTIONS. Using the scale below, please respond by clicking on the phrase that best reflects your feeling about each statement. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree 1. When I was watching this program, I felt like I was really one of the people in the story. SD D N A SA  2. While I was watching this program, activity going on in the room around me was on my mind. SD D N A SA 3. I was mentally involved in the show while watching it. SD D N A SA  4. After the show was over, I found it easy to put it out of my mind. SD D N A SA 5. While I was watching, I was interested to learn how the show would end. SD D N A SA 6. The show affected me emotionally. SD D N A SA

91

 7. I found myself thinking of ways the show could have turned out differently. SD D N A SA  8. I found my mind wandering while watching the show. SD D N A SA 9. I enjoyed the program I watched in the experiment very much. SD D N A SA 10. This is a TV drama that I can enjoy. SD D N A SA  11. Do you think the video you just watched was created more to entertain or more to persuade? To persuade 1 2 3 4 5 6 7 To entertain INSTRUCTIONS. Think about the characters Silver and Dixon from the show you saw. Click on the phrase that best represents your answer. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree 1. Silver… is like me SD D N A SA

is similar to me SD D N A SA

thinks like me SD D N A SA

behaves like me SD D N A SA

2. Dixon… is like me SD D N A SA

is similar to me SD D N A SA

thinks like me SD D N A SA

behaves like me SD D N A SA Now think just about the character Silver, from the show you just watched. For each statement, please use the following scale and click on the phrase that best expresses your own feelings about Silver. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree 1. I think I have a good understanding of Silver. SD D N A SA

92

2. I tend to understand the reasons why Silver did what she did. SD D N A SA 3. While viewing the show, I could feel the emotions Silver portrayed SD D N A SA 4. At key moments in the show, I felt I knew exactly what Silver was going through. SD D N A SA 5. When I watched Silver on the program, I felt I understood the way she felt. SD D N A SA 6. When I watched Silver on the program, I imagined myself doing the same thing she was doing. SD D N A SA 7. When I watched Silver on the program, I really felt as if I were one of the people taking part in the action. SD D N A SA 8. When I watched Silver, I felt like I was a part of her social world. SD D N A SA 9. Silver makes me feel comfortable, like I’m with a friend. SD D N A SA 10. I look forward to watching Silver when her show is on. SD D N A SA 11. If Silver appeared on another show, I would want to watch it. SD D N A SA 12. I see Silver as a natural, down-to-earth person. SD D N A SA 13. Silver seems to understand the kinds of things I want to know. SD D N A SA 14. If I saw a story about Silver in a newspaper or magazine, I would want to read it. SD D N A SA 15. I miss seeing Silver when her show isn’t on for some reason. SD D N A SA 16. If I could, I would like to meet Silver in person. SD D N A SA 17. When Silver shows me how she feels about an issue, it helps me make up my own mind about the issue. SD D N A SA 18. I like to compare my ideas with what Silver says. SD D N A SA 19. I feel sorry for Silver when she makes a mistake. SD D N A SA Now, consider the character Dixon, from the show you just watched. For each statement, please click on the phrase that best expresses your own feelings about this Dixon. SD = strongly disagree D = disagree N = neither agree nor disagree/ neutral A = agree SA = strongly agree

93

1. I think I have a good understanding of Dixon. SD D N A SA 2. I tend to understand the reasons why Dixon did what he did. SD D N A SA 3. While viewing the show, I could feel the emotions Dixon portrayed SD D N A SA 4. At key moments in the show, I felt I knew exactly what Dixon was going through. SD D N A SA 5. When I watched Dixon on the program, I felt I understood the way he felt. SD D N A SA 6. When I watched Dixon on the program, I imagined myself doing the same thing he was doing. SD D N A SA 7. When I watched Dixon on the program, I really felt as if I were one of the people taking part in the drama. SD D N A SA 8. When I watched Dixon, I felt like I was a part of his social world. SD D N A SA 9. Dixon makes me feel comfortable, like I’m with a friend. SD D N A SA 10. I look forward to watching Dixon when his show is on. SD D N A SA 11. If Dixon appeared on another show, I would want to watch it. SD D N A SA 12. I see Dixon as a natural, down-to-earth person. SD D N A SA 13. Dixon seems to understand the kinds of things I want to know. SD D N A SA 14. If I saw a story about Dixon in a newspaper or magazine, I would want to read it. SD D N A SA 15. I miss seeing Dixon when his show isn’t on for some reason. SD D N A SA 16. If I could, I would like to meet Dixon in person. SD D N A SA 17. When Dixon shows me how he feels about an issue, it helps me make up my own mind about the issue. SD D N A SA 18. I like to compare my ideas with what Dixon says. SD D N A SA 19. I feel sorry for Dixon when he makes a mistake. SD D N A SA INSTRUCTIONS. Now, we would like to ask you some questions about your perception of risk involving bipolar disorder. 1. In general, how dangerous do you think bipolar disorder would be for you if you did not get treatment? Extremely 1 2 3 4 5 6 7 Not at all

94

Dangerous Dangerous  2. In general, what do you think are your chances of developing bipolar disorder during your lifetime? No Chance 1 2 3 4 5 6 7 Definitely would happen  3. If you had bipolar disorder and did not seek treatment, what do you think the chances are that you would get worse? No Chance 1 2 3 4 5 6 7 Definitely would happen  4. If you had bipolar disorder and did not receive treatment, what do you think the chances are that you would end up doing something self-destructive? No Chance 1 2 3 4 5 6 7 Definitely would happen 5. How likely is it that you would search for information about bipolar disorder if you thought you had bipolar disorder? Very unlikely 1 2 3 4 5 6 7 Very likely 6. How likely do you think it is that you would seek help from mental health professionals if you experienced bipolar disorder symptom (s)? Very unlikely 1 2 3 4 5 6 7 Very likely Now think about people you care about and their risk involving bipolar disorder. 1. In general, how dangerous do you think bipolar disorder would be for him/her/them if he/she/they did not get treatment? Extremely 1 2 3 4 5 6 7 Not at all Dangerous Dangerous  2. In general, what do you think are his/her/their chances of developing bipolar disorder during his/her/their lifetime? No Chance 1 2 3 4 5 6 7 Definitely would happen  3. If he/she/they had bipolar disorder and did not seek treatment, what do you think the chances are that he/she/they would get worse? No Chance 1 2 3 4 5 6 7 Definitely would happen  4. If he/she/they had bipolar disorder and did not receive treatment, what do you think the chances are that he/she/they would end up doing something self-destructive? No Chance 1 2 3 4 5 6 7 Definitely would happen 5. How likely do you think you would help him/her/them search for information about bipolar disorder if you thought he/she/they had bipolar disorder? Very unlikely 1 2 3 4 5 6 7 Very likely 6. How likely do you think you would be to encourage him/her/them to seek help by contacting mental health professionals if you thought she/he/they had bipolar disorder? Very unlikely 1 2 3 4 5 6 7 Very likely INSTRUCTIONS. Now, please click on the phrase that best represents how you feel. 1. Many people like me have bipolar disorder. SD D N A SA 2. Many college students have bipolar disorder. SD D N A SA 3. Many young people like me have bipolar disorder. SD D N A SA 4. Many Americans have bipolar disorder. SD D N A SA 5. Most people with bipolar disorder seek treatment.

95

SD D N A SA 6. Most people with bipolar disorder have friends who support them. SD D N A SA 7. Most people with bipolar disorder have family members who support them. SD D N A SA 8. Bipolar disorder is a common disease. SD D N A SA 9. Bipolar disorder affects many people. SD D N A SA Now, please respond to the following questions about the television drama – 90210. 1. Can you recall ever seeing this episode before you saw it today? YES NO 2. How often do you watch 90210? Never Some weeks Most weeks Every week 3. Have you seen an episode of 90210 in the last month? YES NO 4. Have you seen the character, Silver in any episode of 90210 in the last month? YES NO 5. About how old is the character Silver in the episode you just watched today? ______6. About how old is the character Dixon in the episode you just watched today? ______INSTRUCTIONS. Now we would like to ask you some questions about your experience involving mental illness. 1. Have you or anyone you care about ever been diagnosed with bipolar disorder? Yes No 2. Approximately how many people do you know who suffer from bipolar disorder? ______3. Have you ever…(check all that apply) □ looked for information about treatment of Bipolar Disorder for yourself or someone else □ made an appointment with a mental health professional to discuss Bipolar Disorder for yourself or someone else □ visited some Bipolar Disorder organizations’ websites for yourself or someone else □ thought about preventing Bipolar Disorder for yourself or someone else □ None of the above 4. About how many TIMES have you searched for information about bipolar disorder for yourself or someone else? Number of times______5. About how many PEOPLE have you discussed bipolar disorder with?______6. Have you ever talked to any mental health professionals? ______Yes ______No Please answer the following questions about yourself. Age in years: ______Do you consider your ethnicity to be primarily: □ African-American □ Asian-American □ Caucasian □ Hispanic

96

□ Native American □ Other (Please specify) ______Gender □ Male □ Female

Thank you! You have completed Step 1 of this two-part study. To complete your participation, please complete a very brief online questionnaire. You will receive an email with this link in about 2 weeks.

FOLLOW-UP QUESTIONNAIRE (Online) Before you begin, we need you to provide us the confidential code you created in the laboratory session of this study so that we can match this questionnaire with the one you filled out previously. PLEASE NOTE: IT IS IMPORTANT THAT YOU USE THE CODE YOU CREATED LAST TIME DURING THE LABORATORY SESSION OF THIS STUDY. In order for you to remember it, please use: a. The LAST TWO digits of your current cell phone number e.g.: (850) 537-2821, the last two digits would be 21 b. The FIRST letter of your FIRST name (NOT nickname) e.g.: Stacey, the first letter would be s c. The two-digit MONTH of your birthday e.g.: 09/06/1992, the two-digit month would be 09 d. The last TWO letters of your LAST name e.g.: Young, the last two letters would be ng As an example, if your phone number was (850) 537-2821, your first name was Stacey, you were born in September, and your last name was Young, your confidential code would be 21s09ng. Please type your code below (PLEASE ASK IF YOU HAVE ANY QUESTIONS). Your Code: ______The following section is for reporting extra credit only. Your information will be separated from the questionnaire to make sure you are anonymous as a research participant. Please provide your first and last name: ______Please provide your official FSU e-mail address (e.g.: [email protected]) below for the online questionnaire two weeks later (PLEASE NOTE: Your email address will be separated from the current questionnaire to make sure you are anonymous as a research participant). ______Please indicate the course you want to get credit for: Listofcourses 1. During your participation in the lab session of this study, you were shown video with excerpts from the TV show 90210. In general, how favorable were your impressions of the show? ○Very unfavorable ○ 2 ○ 3 ○ 4 ○ 5 ○ 6 ○ Very favorable

97

2. How likely is it that you would search for information about bipolar disorder if you thought you had bipolar disorder? Very unlikely 1 2 3 4 5 6 7 Very likely 3. How likely do you think it is that you would seek help from mental health professionals if you experienced bipolar disorder symptom (s)? Very unlikely 1 2 3 4 5 6 7 Very likely 4. How likely do you think you would be to help people you care about search for information about bipolar disorder if you thought he/she/they had bipolar disorder? Very unlikely 1 2 3 4 5 6 7 Very likely 5. How likely do you think you would be to encourage people you care about to seek help by contacting mental health professionals if you thought he/she/they had bipolar disorder? Very unlikely 1 2 3 4 5 6 7 Very likely

6. During the last two weeks have you …(check all that apply) □ looked for information about bipolar disorder for yourself or someone else □ contacted a mental health professional for information about bipolar disorder for yourself or someone else □ made an appointment for yourself or someone else with a mental health professional to discuss bipolar disorder for yourself or someone else □ visited some bipolar disorder organizations’ websites for yourself or someone else □ thought about preventing Bipolar Disorder for yourself or someone else □ none of the above 7. Since you filled out the last questionnaire (about two weeks ago), about how many TIMES have you searched for information about bipolar disorder for yourself or someone else? Number of times______8. Altogether during this two week time period, about how many PEOPLE have you discussed bipolar disorder with?______9. In the last two weeks, have you talked to any mental health professionals for yourself or someone else? ______Yes ______No 10. Did you mention the show you saw to anyone (e.g., friends, roommates, etc. ) after the study? Yes No 11. Not including anything you may have seen for this study, did you recently see an episode of 90210 where the character “Silver” reveals that she has bipolar disorder? Yes No You have now completed the final part of this study. Thank you!

98

REFERENCES

Altamura, A. C., Dell’Osso, B., Berlin, H. A., Buoli, M., Bassetti, R., & Mundo, E. (2010). Duration of untreated illness and suicide in bipolar disorder: a naturalistic study. European archives of psychiatry and clinical neuroscience, 260(5), 385-391.

Appel, M., & Richter, T. (2010). Transportation and need for affect in narrative persuasion: A mediated moderation model. Media Psychology, 13(2), 101-135.

Arnett, J. J. (2000). Optimistic bias in adolescent and adult smokers and nonsmokers. Addictive Behaviors, 25(4), 625-632.

Bae, H. S., & Kang, S. (2008). The Influence of viewing an entertainment–education program on cornea donation intention: a test of the theory of planned behavior. Health Communication, 23(1), 87-95.

Bandura, A. (1995). Exercise of personal and collective efficacy in changing societies. Self-efficacy in changing societies, 15, 334.

Bandura, A. (2001). Social cognitive theory of mass communication. Media psychology, 3(3), 265- 299.

Bandura, A. (2002). Social cognitive theory of mass communication. Media effects: Advances in theory and research, 2, 121-153.

Bandura, A. (2004). Social cognitive theory for personal and social change by enabling media. Entertainment-education and social change: History, research, and practice, 75-96.

Beck, V. (2004). Working with daytime and prime-time television shows in the United States to promote health. In A. Singhal, M.J. Cody, E.M. Rogers & M. Sabido (Eds.), Entertainment- Education and Social Change: History, Research, and Practice (pp. 207-224). Mahwah, NJ: Erlbaum Associates.

Brehm, S. S., & Brehm, J. W. (1981). Psychological reactance: A theory of freedom and control. San Diego, CA: Academic Press.

Brodie, M., Foehr, U., Rideout, V., Baer, N., Miller, C., Flournoy, R., & Altman, D. (2001). Communicating health information through the entertainment media. Health affairs, 20(1), 192-199.

Burgoon, M., Alvaro, E., Grandpre, J., & Voulodakis, M. (2002). Revisiting the theory of psychological reactance. The persuasion handbook, 213-232.

Busselle, R., & Bilandzic, H. (2008). Fictionality and perceived realism in experiencing stories: A model of narrative comprehension and engagement. Communication Theory, 18(2), 255-280.

99

Busselle, R., & Bilandzic, H. (2009). Measuring narrative engagement. Media Psychology, 12(4), 321-347.

Busselle, R., & Bilandzic, H. (2013). Narrative Persuasion. In Dillard, J. P., & Shen, L. (Eds.). The SAGE Handbook of Persuasion: Developments in Theory and Practice. SAGE, 200-215.

Chen, S., & Chaiken, S. (1999). The heuristic-systematic model in its broader context. Dual-process theories in social psychology, 73-96.

Cohen, J. (2001). Defining identification: A theoretical look at the identification of audiences with media characters. Mass Communication & Society, 4(3), 245-264.

Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience. New York: Harper & Row.

Csikszentmihalyi, M. (1997). Finding flow: The psychology of engagement with everyday life. New York: BasicBooks.

Diefenbach, D. L., & West, M. D. (2007). Television and attitudes toward mental health issues: Cultivation analysis and the third-person effect. Journal of Community Psychology, 35(2), 181-195.

Dillard, J. P., & Shen, L. (2005). On the nature of reactance and its role in persuasive health communication. Communication Monographs, 72(2), 144-168.

Escalas, J. E. (2007). Self-Referencing and Persuasion: Narrative Transportation versus Analytical Elaboration. Journal of Consumer Research, 33(4), 421-429.

Eyal, K., & Rubin, A. M. (2003). Viewer Aggression and Homophily, Identification, and Parasocial Relationships With Television Characters. Journal Of Broadcasting & Electronic Media, 47(1), 77.

Gerrig, R. J. (1993). Experiencing narrative worlds. New Haven, CT: Yale University Press.

Gibbons, F. X., Gerrard, M., Blanton, H., & Russell, D. W. (1998). Reasoned action and social reaction: Willingness and intention as independent predictors of health risk. Journal of personality and social psychology, 74, 1164-1180.

Giles, D. C. (2002). Parasocial interaction: A review of the literature and a model for future research. Media psychology, 4(3), 279-305.

Gonzalez, J. M., Alegria, M., & Prihoda, T. J. (2005). How do attitudes toward mental health treatment vary by age, gender, and ethnicity/race in young adults? Journal of Community Psychology, 33, 611–629.

100

Goossens, L., Beyers, W., Emmen, M., & Van Aken, M. A. (2002). The imaginary audience and personal fable: Factor analyses and concurrent validity of the “new look” measures. Journal of Research on Adolescence, 12(2), 193-215.

Graesser, A. C. (1981). Prose comprehension beyond the word. New York: Springer-Verlag.

Green, M. C., & Brock, T. C. (2000). The role of transportation in the persuasiveness of public narratives. Journal of personality and social psychology, 79(5), 701-721.

Green, M. C., & Brock, T. C. (2002). In the Mind’s Eye: Transportation-Imagery Model of Narrative Persuasion. Mahwah, NJ, US: Lawrence Erlbaum Associates Publishers.

Green, M. C., Brock, T. C., & Kaufman, G. F. (2004). Understanding media enjoyment: The role of transportation into narrative worlds. Communication Theory, 14(4), 311-327.

Green, M. C., & Clark, J. L. (2012). Transportation into narrative worlds: implications for entertainment media influences on tobacco use. Addiction.

Greenberg, B.S., Salmon, C.T., Patel, D., Beck, V. & Cole, G. (2004). Evolution of an E-E research agenda. In A. Singhal, M.J. Cody, E.M. Rogers & M. Sabido (Eds.), Entertainment- Education and Social Change: History, Research, and Practice (pp. 191-206). Mahwah, NJ: Erlbaum Associates.

Hether, H. J., Huang, G. C., Beck, V., Murphy, S. T., & Valente, T. W. (2008). Entertainment- education in a media-saturated environment: examining the impact of single and multiple exposures to breast cancer storylines on two popular medical dramas. Journal of health communication, 13(8), 808-823.

Hoffner, C., & Buchanan, M. (2005). Young adults' wishful identification with television characters: The role of perceived similarity and character attributes. Media Psychology, 7(4), 325-351.

Hoffner, C. A., & Cohen, E. L. (2012). Responses to obsessive compulsive disorder on monk among series fans: Parasocial relations, presumed media influence, and behavioral outcomes. Journal of Broadcasting & Electronic Media, 56(4), 650-668.

Horton, D., & Wohl, R. R. (1956). Mass communication and parasocial interaction: Observations on intimacy at a distance. Psychiatry, 19, 215–229.

Jeong, H., & Park, H. S. (2013). The Effect of Parasocial Interaction on Intention to Register as Organ Donors Through Entertainment-Education Programs in Korea. Asia-Pacific Journal of Public Health.

Jorm, A. F. (2012). Mental health literacy: Empowering the community to take action for better mental health. American Psychologist, 67(3), 231.

101

Kaiser Family Foundation. (2004). Entertainment education and health in the United States. A report to the Kaiser Family Foundation. Menlo Park, CA: Henry J. Kaiser Family Foundation.

Kallgren, C. A., Reno, R. R., & Cialdini, R. B. (2000). A focus theory of normative conduct: When norms do and do not affect behavior. Personality and Social Psychology Bulletin, 26, 1002– 1012.

Kessler, R. C., Chiu, W. T., Demler, O., & Walters, E. E. (2005). Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Archives of general psychiatry, 62(6), 617.

Klein, C. T., & Helweg-Larsen, M. (2002). Perceived control and the optimistic bias: A meta- analytic review. Psychology and Health, 17(4), 437-446.

Knowles, E. S., & Linn, J. A. (2004). The importance of resistance to persuasion. In E. S. Knowles & J. A. Linn (Eds.), Resistance and persuasion (pp. 3–11). Mahwah, NJ: Lawrence Erlbaum.

Kupfer, D. J. (2005). The increasing medical burden in bipolar disorder. JAMA: the journal of the American Medical Association, 293(20), 2528-2530.

Lang, A. (2000). The limited capacity model of mediated message processing. Journal of Communication, 50(1), 46-70.

Lapsley, D. K., & Hill, P. L. (2010). Subjective invulnerability, optimism bias and adjustment in emerging adulthood. Journal of youth and adolescence, 39(8), 847-857.

Leboyer, M., Henry, C., Paillere‐ Martinot, M. L., & Bellivier, F. (2005). Age at onset in bipolar affective disorders: a review. Bipolar Disorders, 7(2), 111-118.

Lee, B. K., Park, H. S., Choi, M. I., & Kim, C. S. (2010). Promoting Organ Donation Through an Entertainment—Education TV Program in Korea Open Your Eyes. Asia-Pacific Journal of Public Health, 22(1), 89-97.

Levy, M. R. (1979). Watching TV news as para-social interaction. Journal of Broadcasting, 23, 69– 80.

Mazzocco, P. J., Green, M. C., Sasota, J. A., & Jones, N. W. (2010). This story is not for everyone: Transportability and narrative persuasion. Social Psychological and Personality Science, 1(4), 361-368.

Merikangas, K. R., Akiskal, H. S., Angst, J., Greenberg, P. E., Hirschfeld, R., Petukhova, M., & Kessler, R. C. (2007). Lifetime and 12-month prevalence of bipolar spectrum disorder in the National Comorbidity Survey replication. Archives of General Psychiatry, 64(5), 543.

102

Miller, C. H., Lane, L. T., Deatrick, L. M., Young, A. M., & Potts, K. A. (2007). Psychological reactance and promotional health messages: The effects of controlling language, lexical concreteness, and the restoration of freedom. Human Communication Research, 33(2), 219- 240.

Moyer-Gusé, E. (2007). Entertainment television and safe sex: Understanding effects and overcoming resistance (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses database. (UMI No. 3274446)

Moyer-Gusé, E. (2008). Toward a Theory of Entertainment Persuasion: Explaining the Persuasive Effects of Entertainment-Education Messages. Communication Theory, 18(3), 407-425.

Moyer-Gusé, E., Chung, A. H., & Jain, P. (2011). Identification with characters and discussion of taboo topics after exposure to an entertainment narrative about sexual health. Journal of Communication, 61(3), 387-406.

Moyer-Gusé, E., Jain, P. and Chung, A. H. (2012). Reinforcement or Reactance? Examining the Effect of an Explicit Persuasive Appeal Following an Entertainment‐ Education Narrative. Journal of communication, 62(6), 1010-1027.

Moyer-Gusé, E., Mahood, C., & Brookes, S. (2011). Entertainment-Education in the Context of Humor: Effects on Safer Sex Intentions and Risk Perceptions. Health Communication, 26(8), 765-774.

Moyer-Gusé, E., & Nabi, R. L. (2010). Explaining the effects of narrative in an entertainment television program: Overcoming resistance to persuasion. Human Communication Research, 36(1), 26-52.

Movius, L., Cody, M., Huang, G., Berkowitz, M., & Morgan, S. (2007). Motivating television viewers to become organ donors. Cases in Public Health Communication & Marketing, 1, 1- 21.

Nahm, S., Le, K., de Castro Buffington, S., Raider, S., & Susan Resko, M. M. (2010). Engaging youth through partnerships in entertainment education. Cases in Public Health Communication & Marketing, 4, 57-78.

National Institute of Mental Health. (2008). Bipolar Disorder. Retrieved from http://www.nimh.nih.gov/health/publications/bipolar-disorder/index.shtml

Nock, M. K., Hwang, I., Sampson, N., Kessler, R. C., Angermeyer, M., Beautrais, A., & Levinson, D. (2009). Cross-National Analysis of the Associations among Mental Disorders and Suicidal Behavior: Findings from the WHO World Mental Health Surveys. Plos Medicine, 6(8), 1-17.

Osby, U., Brandt, L., Correia, N., Ekbom, A., & Sparen, P. (2001). Excess mortality in bipolar and unipolar disorder in Sweden. Archives of General Psychiatry, 58(9), 844.

103

Palmer, A. (2002). Reaching youth worldwide: Johns Hopkins Center for Communication Programs, 1995–2000. Working Paper No. 6. Baltimore, MD: Johns Hopkins University, Center for Communication Programs.

Rhodes, N., & Ewoldsen, D. R. (2009). Attitude and norm accessibility and cigarette smoking. Journal of Applied Social Psychology, 39(10), 2355-2372.

Rubin, A. M., & Perse, E. M. (1987). Audience activity and soap opera involvement: A uses and effects investigation. Human Communication Research, 14, 246-268.

Perse, E. M., & Rubin, R. R. (1989). Attribution in social and parasocial relationships. Communication Research, 16, 59–77.

Petty, R. E., & Cacioppo, J. T. (1981). Attitudes and persuasion: classic and contemporary approaches. Dubuque, IA: Brown.

Petty, R. E., & Wegener, D. T. (1999). The elaboration likelihood model: Current status and controversies.

Pirkis, J., Blood, R., Francis, C., & McCallum, K. (2006). On-Screen Portrayals of Mental Illness: Extent, Nature, and Impacts. Journal Of Health Communication, 11(5), 523-541.

Piotrow, P. T., & de Fossard, E. (2004). Entertainment-education as a public health intervention. In A. Singhal, M. J. Cody, E.M. Rogers, & M. Sabido (Eds), Entertainment-education and social change (pp. 39-60). Mahwah, NJ: Erlbaum.

Poindexter, D. O. (2004). A history of entertainment-education, 1958–2000. Entertainment– Education and Social Change: History, Research, and Practice. Mahwah, NJ: Lawrence Erlbaum Associates, 21-37.

Quick, B. L., & Stephenson, M. T. (2007). Further evidence that psychological reactance can be modeled as a combination of anger and negative cognitions. Communication Research, 34(3), 255-276.

Quick, B, Shen, & Dillard (2013). Reactance Theory and Persuasion. In Dillard, J. P., & Shen, L. (Eds.). The SAGE Handbook of Persuasion: Developments in Theory and Practice. SAGE, 167-179.

Ritterfeld, U., & Jin, S. A. (2006). Addressing media stigma for people experiencing mental illness using an entertainment-education strategy. Journal of health psychology, 11(2), 247-267.

Satcher, D. (2000). Mental health: A report of the Surgeon General--Executive summary. Professional Psychology: Research and Practice, 31(1), 5.

104

Singhal, A., & Rogers, E.M. (2004). The status of entertainment-education worldwide. In A. Singhal, M. J. Cody, E.M. Rogers, & M. Sabido (Eds), Entertainment-education and social change (pp. 3-20). Mahwah, NJ: Erlbaum.

Slater, M. D. (1997). Persuasion processes across receiver goals and message genres. Communication Theory, 7(2), 125-148.

Slater, M. D., & Rouner, D. (2002). Entertainment—education and elaboration likelihood: Understanding the processing of narrative persuasion. Communication Theory, 12(2), 173- 191.

Slater, M. D., Rouner, D., & Long, M. (2006). Television Dramas and Support for Controversial Public Policies: Effects and Mechanisms. Journal Of Communication, 56(2), 235-252.

Stephenson, M. T. (2003). Examining adolescents' responses to antimarijuana PSAs. Human Communication Research, 29(3), 343-369.

Strange, J. J., & Leung, C. C. (1999). How anecdotal accounts in news and in fiction can influence judgments of a social problem’s urgency, causes, and cures. Personality and Social Psychology Bulletin, 25(4), 436-449.

Sood, S., Menard, T., & Witte, K. (2004). The theory behind entertainment-education. In A. Singhal, M. J. Cody, E.M. Rogers, & M. Sabido (Eds), Entertainment-education and social change (pp. 117-145). Mahwah, NJ: Erlbaum.

Steinke, J., Applegate, B., Lapinski, M., Ryan, L., & Long, M. (2012). Gender Differences in Adolescents’ Wishful Identification With Scientist Characters on Television. Science Communication, 34(2), 163-199.

Tal-Or, N., & Cohen, J. (2010). Understanding audience involvement: Conceptualizing and manipulating identification and transportation. Poetics, 38(4), 402-418.

Tondo, L., Isacsson, G., & Baldessarini, R. J. (2003). Suicidal behaviour in bipolar disorder. CNS drugs, 17(7), 491-511.

Valente, T. W., Murphy, S., Huang, G., Gusek, J., Greene, J., & Beck, V. (2007). Evaluating a minor storyline on ER about teen obesity, hypertension, and 5 A Day. Journal of Health Communication, 12(6), 551-566.

Vaughan, P. W., Rogers, E. M., Singhal, A., & Swalehe, R. M. (2000). Entertainment-education and HIV/AIDS prevention: a field experiment in Tanzania. Journal of Health Communication, 5(S1), 81-100.

Wahl, O., Hanrahan, E., Karl, K., Lasher, E., & Swaye, J. (2007). The depiction of mental illness in children's television programs. Journal of Community Psychology, 35(1), 121-133.

105

Wang, P. S., Angermeyer, M., Borges, G., Bruffaerts, R., Chiu, W. T., deGirolamo, G., . . . Üstün, T. B. (2007). Delay and failure in treatment seeking after first onset of mental disorders in the World Health Organization’s World Mental Health Survey Initiative. World Psychiatry,6, 177–185.

Weinstein, N.D. (1980). Unrealistic optimism about future life events. Journal of Personality and Social Psychology, 39, 306–320.

Weinstein, N. D. (1989). Optimistic biases about personal risks. Science, 246(4935), 1232-1233.

Whittier, D. K., Kennedy, M. G., Lawrence, J. S. S., Seeley, S., & Beck, V. (2005). Embedding health messages into entertainment television: Effect on gay men's response to a syphilis outbreak. Journal of health communication, 10(3), 251-259.

Winsten, J. A., & DeJong, W. (2001). The designated driver campaign. Public communication campaigns, 3, 290-294.

Wilkin, H. A., Valente, T. W., Murphy, S., Cody, M. J., Huang, G., & Beck, V. (2007). Does entertainment-education work with Latinos in the United States? Identification and the effects of a telenovela breast cancer storyline. Journal of Health Communication, 12(5), 455- 469.

Zillmann, D., 1991. The logic of suspense and mystery. In: Bryant, J., Zillmann, D. (Eds.), Responding to the Screen: Reception and Reaction Processes (pp. 281-304). Lawrence Erlbaum Associates, Hillsdale, NJ.

106

BIOGRAPHICAL SKETCH

Danyang Zhao earned her Bachelor of Arts degree in English and French from

Heilongjiang University in People’s Republic of China in 2006. She is to receive her Master of

Arts degree in Media and Communication Studies from the School of Communication at The

Florida State University in the spring of 2014. Danyang Zhao’s research interests include narrative persuasion, entertainment-education, health communication, media effects, and mental- health-related attitude and behaviors.

107