LET ME TELL YOU A STORY: NARRATIVE MESSAGE EFFECTS IN THE
COMMUNICATION OF STIGMA ON MIGRANT WORKERS IN CHINA
By
Xiaodi Yan

A THESIS
Submitted to
Michigan State University
in partial fulfilment of the requirements
for the degree of
Communication—Master of Arts
2017

ABSTRACT
LET ME TELL YOU A STORY: NARRATIVE MESSAGE EFFECTS IN THE
COMMUNICATION OF STIGMA ON MIGRANT WORKERS IN CHINA
By
Xiaodi Yan
Rural-to-urban migrant workers in China are a group who have significant impact on
society. Studies have indicated that there is social stigma towards migrant workers. The goal of
this study is to investigate the role narrative messages play in stigma communication to urban
citizens. Based on a 2 (valence: positive or negative) × 2 (type: narrative or non-narrative)
between-subjects factorial design, 305 participants were randomly assigned to one of four
experimental conditions and one no-message control condition to assess their emotional
reactions and stigma attitudes. The results showed that there are main effects of message valence
on emotional reactions and stigma attitudes. Additionally, the effects of message type on stigma
attitudes differed depending on message valence. There was a main effect of narrative message
on stigma attitudes in positive conditions, while the effect was mediated by message believability
in negative conditions. The results have both theoretical and practical implications.
Keywords: stigma communication, rural-to-urban migrant workers in China, narrative
message, emotions, message believability

TABLE OF CONTENTS

LIST OF TABLES ………………………………………………………………………….……. v
LIST OF FIGURES ………………………………………………………………………….….. vi
INTRODUCTION …………………………………………………………………………….…. 1
STIGMA ON MIGRANT WORKERS IN CHINA ……………………………………………... 2
Labeling ………………………………………………………………………………….. 2
Negative Attributes …………………………………………………………………….… 3
Separating “Us” From “Them” ……………………………………………………….….. 3
Discrimination ………………………………………………………………………....… 4
COMMUNICATION OF STIGMA ………………………………………………………….….. 4
Stigma Cues and Message Valence ………………………………………………….…... 5
Message Valence and Outcomes …………………………………………………….….... 5
Emotional reactions ………………………………………………………….…... 5
Stigma attitudes ………………………………………………………………….. 6
NARRATIVE MESSAGE EFFECTS ………………………………………………………….... 7
Narrative Message …………………………………………………………………….…. 7
Social Influence Power of Narrative Message ………………………………………….... 7
Narrative Message Effects in Stigma Communication …………………………………... 9
The Role of Message Believability ……………………………………………………... 10
METHOD …………………………………………………………………………………….… 11
Participants ……………………………………………………………………………... 11
Design …………………………………………………………………………………... 11
Message Inductions …………………………………………………………………….. 11
Procedure ……………………………………………………………………………….. 12
Measurement …………………………………………………………………………… 13
Emotional reactions ………………………………………………………….…. 13
Stigma attitudes ……………………………………………………………….... 13
Message believability …………………………………………………………... 13
Demographic information ……………………………………………………… 14
RESULTS ………………………………………………………………………………………. 14
Preliminary Analysis …………………………………………………………………… 14
Hypothesis Testing ……………………………………………………………………... 15
Research Question Exploration …………………………………………………….…… 16
DISCUSSION ………………………………………………………………………………….. 17
Theoretical and Practical Implications ……………………………………………….…. 18
Limitations and Future Directions …………………………………………………….... 21
CONCLUSION ………………………………………………………………………………… 23
APPENDICES ………………………………………………………………………………….. 24
APPENDIX A: Message Inductions ……………………………………………………. 25
APPENDIX B: Scale of Stigma Attitudes toward Migrant Workers in China ……….… 27

iii

REFERENCES …………………………………………………………………………………. 31

iv

LIST OF TABLES

Table 1: Descriptive Statistics for All Variables by Condition ………………………………… 29
Table 2: Overall Correlation Matrix ……………………………………………………………. 30

v

LIST OF FIGURES

Figure 1: The Effect of Message Type on Stigma in Negative Conditions ……………………. 30

vi

INTRODUCTION
The label “rural-to-urban migrant workers” in China refers to the population of Chinese
registered in rural areas according to the Hukou System (Chinese Household Registration
System), but who leave their hometown to seek non-agricultural employment in urban areas
(Guan, 2011; Zhang & Luo, 2012). The emergence of this social group has had significant
impact, both positively and negatively, on various aspects of Chinese society (Guan, 2011;
Wong, Chang, & He, 2007). Previous studies have identified that social stigma is experienced by
migrant workers. These studies have been largely qualitative reports based on the experience of
migrant workers (Chen et al., 2011; Guan, 2011; Guan & Liu, 2013; Lin et al., 2011; Wang, Li,
Stanton, & Fang, 2010). The current study is intended to empirically measure urban citizens’
emotional and attitudinal responses to migrant workers.
Communication plays a critical role in the enactment of stigma. Smith (2007) describes
stigma as a communicative process that can induce relevant emotional reactions and stigma
attitudes. Stigma is an affectively charged event (Smith, 2007). One type of message that is
advantageous in eliciting emotions is narratives (Nabi & Green, 2015). Given the long human
history of storytelling, narrative is described as an intuitive and comfortable mode to send and
receive information (Murphy et al., 2015; Kreuter et al., 2007). The social influence potential of
narrative messages has been demonstrated in many studies, while the influence of narrative in
stigma communication is less clear (Bilandzic & Busselle, 2013; Braddock & Dillard, 2016;
Chung & Slater, 2013; Oliver, Dillard, Bae, & Tamul, 2012).
The goal of the current study is to investigate the role that narrative messages play in
stigma communication. Is a narrative message, compared to a non-narrative message, more
effective in bringing intended outcomes predicted by either a positively or negatively valenced

1

message? This study first provides a conceptualization of stigma against migrant workers in
China based on social descriptions of this group. The next section discusses the communication
of stigma and outcomes. Then, a review of studies examining the social influence of narratives is
provided, and the effects in stigma communication are hypothesized. Next, an empirical study
tests the hypotheses, and the results are presented. Finally, implications, limitations, and
directions for future research are discussed.
STIGMA ON MIGRANT WORKERS IN CHINA
Goffman (1963) defined stigma as “an attribute that is deeply discrediting” (p. 3) and
someone who bears such attributes is “disqualified from full social acceptance” (p. 11) and
“reduced in our minds from a whole and usual person to a tainted, discounted one” (p. 12). The
current study conceptualizes stigma by adopting Link & Phelan’s (2001) model that stigma has
four interrelated components: labeling, negative attributes, separating “us” from “them”, and
discrimination.
Labeling
Stigma begins with certain type of human difference being identified and labeled. The
difference can be physical, social, or moral (Meisenbach, 2010). The difference can be visible or
invisible, and can be concealable or disclosed. Previous studies showed that migrant workers
self-report themselves being labeled as “cheap laborers” or “outsiders” by urban citizens (Guan,
2011; Guan & Liu, 2013). They hold the belief that they have less status than urban citizens in
many aspects. They perceive themselves as unofficial to cities, and that they are treated as
temporary employees hired to work in cities (Zhang & Luo, 2012). Migrant workers described
that in urban citizens’ eyes, the type of work they do is laborious, non-technical, hazardous and
indecent (Wong et al., 2007; Zhang & Luo, 2012). Additionally, migrant workers reported that

2

some “visible” characteristics make them distinguishable. For instance, they think they appear
untidy to urban citizens (Chen et al., 2011; Guan, 2011; Wang et al., 2010). Additionally, since
migrant workers are not local to the place where they work, they are aware that they have strong
foreign accents (Guan, 2011).
Negative Attributes
Once certain individuals have been identified and labeled, certain beliefs about the
character of the stigmatized are likely to be assigned, such as negative personal characteristics,
discredited moral character, deviant beliefs. Migrant workers believe that they are stereotyped as
uneducated and low in intelligence (Guan, 2011; Wong et al., 2007). They are described as poor,
inferior, and dirty by urban citizens (Guan, 2011; Zhang & Luo, 2012). Some migrant workers
reported that they have been questioned about their health and hygienic condition (Chen et al.,
2011; Wang et al., 2010). Additionally, migrant workers think they are identified with peril and
threat in urban citizens’ opinions. They are thought to be responsible for increased crime rates
and various events threatening social stability and security (Guan, 2011; Wong et al., 2007).
Separating “Us” From “Them”
After attributing negative stereotypes to human difference, the process of separating “us”
from “them” is enacted. Migrant workers think that they cannot mingle in or be accepted by
legitimate urban citizens (Guan, 2011; Zhang & Luo, 2012). One frequently encountered
compelling case is that if migrant workers have a seat on the bus, despite available seats next to
them, urban citizens tend to stand in the aisle and are not willing to sit next to the migrant worker
(Guan & Liu, 2013). Migrant workers explained that urban citizens are different from migrant
workers, so that urban citizens would not sit together with them. (Lin et al., 2011; Wang et al.,
2010).

3

Discrimination
Often, stigmatized individuals are denied access to various social resources, and find
themselves at the lower end of the social hierarchy (Link & Phelan, 2001). The stigmatized are
likely to experience both individual and structural discrimination. Migrant workers perceive
themselves being placed at the bottom of the social hierarchy in cities (Chen et al., 2011). They
are excluded from various social resources, such as education, health care, employment, housing,
adequate nutrition, etc. (Wong et al., 2007; Zhang & Luo, 2012). Additionally, discrimination of
migrant workers occurs at both individual and institutional levels (Guan, 2011; Wong et al.,
2007; Zhang & Luo, 2012). Discrimination can be seen in social interactions as well as in
unfavorable policies and in the general social environment (Guan, 2011; Wong et al., 2007;
Zhang & Luo, 2012).
Based on this description, rural-to-urban migrant workers perceive themselves to be
stigmatized by urban citizens in China. Smith (2007) considered that stigma is enacted through
spreading messages with stigma cues within one’s communities. Communication is critical to
stigmatization. It would be of significance to understand the communicative mechanism
underlying the enactment of stigma towards migrant workers among urban citizens. Thus, the
process of stigma communication and outcomes are discussed below.
COMMUNICATION OF STIGMA
Smith (2007) considered stigma as a communicative event. The enactment of stigma
occurs through communication of “messages spread through communities to teach their
members to recognize the disgraced and to react accordingly” (Smith, 2007, p. 464). Stigma
messages induce certain emotional reactions and encourage the activation of relevant social
attitudes and stereotypes (Smith, 2007).

4

Stigma Cues and Message Valence
According to Smith (2007), stigma messages include four content cues: mark, group
labeling, responsibility, and peril. A mark refers to some undesirable trait, behavior or condition
that defines people in a particular situation. Group labeling is description of the stigmatized
group as a separate group entity. Responsibility refers to a perception of choice and control over
the stigmatized condition and resulting group threat. Peril warns about danger and threat, and
unmarked members must protect themselves through collective efforts to eliminate the
stigmatized.
In this study, a negative message refers to a message with stigmatizing cues. A negative
message emphasizes that migrant workers are marked and labeled negatively, are responsible for
their situation, and are perilous to urban citizens. On the contrary, a positive message refers to a
message with de-stigmatizing cues. A positive message opposes negative marks or labels
attached to migrant workers, attributes their situation to factors beyond their control, and does
not describe migrant workers as perilous to urban citizens.
Message Valence and Outcomes
Stigma messages deal with intergroup relations. When group identities become salient,
people may assess a situation in terms of group benefits and harms. Depending on the message
valence, indicated by content cues in the message, corresponding emotional and attitudinal
reactions may be induced (Smith, 2007; Smith, 2014).
Emotional reactions. Stigma messages can elicit emotional reactions shaped by
evolutionary pressure to push individuals to react automatically (Smith, 2007). A negative
message with stigmatizing cues is likely to evoke various negative emotions. A negative message
indicates that the proper group functioning is threatened. Different types of group threats can

5

evoke different discrete emotions (Cottrell & Neuberg, 2005). “Perceptions of contamination
(physical or moral) are associated with disgust, obstacles evoking anger, and physical threats to
safety suggest fear” (Smith, 2007, p. 472). By contrast, a positive message with de-stigmatizing
cues is less likely to evoke these negative emotional responses (Cottrell & Neuberg, 2005).
Stigma attitudes. Stigma messages can help form and bolster stigma attitudes (Smith,
2007; Smith, 2014). Stigma attitudes consist of negative evaluations about the targets. A negative
message with stigmatizing cues makes negative evaluations and judgments about the targets
accessible to the stigmatizer (Smith, 2007). A negative message indicates that the targets are
undesirable, stereotypes the targets with negative attributes, blames the target for responsibility,
and warns the threats from the targets (Link & Phelan, 2001; Meisenbach, 2010; Smith, 2007).
These negative evaluations about the targets contribute to form stigma attitudes toward the
targets (Smith, 2007). On the contrary, a positive message with de-stigmatizing cues
communicates positive evaluations about the targets. These positive perceptions and evaluations
about the targets can contribute to reduce stigma attitudes toward the targets.
In this study, negative messages with stigmatizing cues and positive messages with destigmatizing cues are employed to investigate the effects of message valence on emotional
reactions and on stigma attitudes. The first hypothesis for investigation is derived from the
rationale presented.
Hypothesis 1 (H1): A negative message is likely to result in (a) more disgust, (b) more
anger, (c) more fear, and (d) higher stigma attitudes, compared to a positive message.
According to studies on message effects, not only does message valence have an impact
on related outcomes, the presentation of information in a message also has outcomes. Given the

6

important role of emotions in stigma, narrative messages should be especially relevant to this
inquiry, since emotions are a fundamental part of the narrative experience (Nabi & Green, 2015).
Narrative messages have drawn increasing research attention in recent years, and have
been shown to have promising social influence potential (Braddock & Dillard, 2016; McQueen,
Kreuter, Kalesan, & Alcaraz, 2011; Moyer-Gusè, Chung, & Jain, 2011; Niederdeppe, Shapiro, &
Porticella, 2011). Whether and how narrative messages influence the communication of stigma
remains to be investigated. Thus, narrative message effects are discussed below.
NARRATIVE MESSAGE EFFECTS
Narrative Message
A narrative is defined as “a representation of connected events and characters that has an
identifiable structure, is bounded in space and time, and contains implicit or explicit messages
about the topic being addressed” (Kreuter et al., 2007, p. 222). What sets narratives apart from
non-narratives is that a narrative is about concrete events occurring to specific characters in
certain settings (de Graaf, Sanders, & Hoeken, 2016). Attitudes or opinions are oftentimes
explicitly expressed in a non-narrative message compared to a narrative message.
Narrative is considered the basic mode of human communication (Hinyard & Kreuter,
2006). “Humans are innate storytellers, and storytelling has played a vital role in transmitting
perspective and normative information in most cultures for thousands of years” (Murphy et al.,
2015, p. 2117). It is the basic venue through which we learn about the world around us.
Narratives are an intuitive way of sending and receiving information (Kreuter et al., 2007).
Social Influence Power of Narrative Message
The social influence potential of narrative messages has been well established. Narratives
have been attested to be effective in bringing about intended outcomes in domains of politics,

7

policies, education, health, intergroup relations, etc. (Bilandzic & Busselle, 2013; Iguarta, 2010;
Lemal & Van den Bulck, 2010; McQueen et al., 2011; Murphy et al., 2015; Oliver et al., 2012).
A meta-analysis conducted by Braddock & Dillard (2016) provided integrative evidence of
positive relationships between exposure to a narrative and narrative-consistent beliefs, attitudes,
intentions, and behaviors.
Studies have also provided evidence for the advantages of narrative messages over other
types of messages in terms of social influence. McQueen et al. (2011) found that narratives
produced stronger cognitive and affective responses than traditional informational messages.
Lemal & Van den Bulck (2010) also found that participants exposed to a narrative message were
more likely to perform the behaviors recommended in the messages compared to non-narrative
conditions. Murphy et al. (2013, 2015) have yielded similar findings that compared to nonnarrative messages, narrative messages were more effective in increasing related knowledge and
attitudes. Results of Oliver et al. (2012) also showed that narrative-formatted stories produced
more emotional reactions, intended attitudes and behavioral intentions.
Multiple processes are involved in the mechanism through which narrative messages
have intended effects. The most cited process in narrative message effects is transportation
(Appel & Richter, 2010; Banerjee & Greene, 2012; de Graaf et al, 2009; Hinyard & Kreuter,
2006; Murphy et al., 2013). Transportation is defined as an integrative melding of affective and
cognitive resources with the story (Green, 2004). Transported individuals are so absorbed and
immersed in the narrative world that they perceive the narrative more like an actual experience,
so they tend to accept the message propositions (Appel & Richter, 2010; Green, 2004; Hinyard
& Kreuter, 2006). Highly transported individuals are also more likely to experience higher
emotional arousals, which could further contribute to corresponding evaluation and motivation

8

tendencies (Nabi & Green, 2015). Transportation is associated with changes in knowledge,
attitudes, and behaviors (Murphy et al., 2013).
Narrative Message Effects in Stigma Communication
The narrative message type is especially relevant to the issue of stigma. As message
recipients experience transportation, they are affectively engaged and absorbed in the story
world. Narrative is very effective in evoking emotions. Emotions are considered a fundamental
part of narrative experience (Nabi & Green, 2015).
Emotions play an important role in the enactment of stigma (Smith, 2007). Emotional
reactions can encourage cognitive reactions that make certain evaluations more accessible to
individuals, and thus have influences on relevant attitudes. Studies show that discrete negative
emotions, namely disgust, fear, and anger, are responsible for the development of stigma
attitudes (Smith, 2007; Smith, 2014).
Narrative messages make certain emotional experiences more salient, and thus will result
in more valence-predicted outcomes. In this sense, this study applies narrative message effects to
the communicative processes of stigma. Narrative messages are compared with non-narrative
messages to look at the message effects on emotional reactions and stigma attitudes. The study
proposes that a narrative message is effective in bringing about intended outcomes, more so than
a non-narrative message.
Hypothesis 2 (H2): A narrative message is likely to result in more valence-predicted
outcomes compared to a non-narrative message, such that:
H2 (a): A negative narrative message is likely to result in (i) more disgust, (ii) more
anger, (iii) more fear, and (iv) higher stigma attitudes, compared to a negative non-narrative
message.

9

H2 (b): A positive narrative message is likely to result in (i) less disgust, (ii) less anger,
(iii) less fear, and (iv) lower stigma attitudes, compared to a positive non-narrative message.
The Role of Message Believability
Additionally, an important issue in social influence communication is to understand how
message recipients perceive the information (Bettman, Payne, & Staelin, 1986). Fishbein and
Ajzen (1975) stated that “…with persuasive communication, the major problem is to ensure that
the receiver accepts (i.e., believes) the communication which attempts to link the object and the
attribute.” (p. 389). Furthermore, Petty and Cacioppo (1986) also emphasized that it is necessary
for message arguments to be perceived as believable in order to influence attitudes.
The concept of believability was originally proposed by Beltramini (1982) in advertising
context. Message believability focuses on the message content, but not the source. It suggests the
message recipients’ confidence in the truthfulness and perception of plausibility of the
information (Andrews, Netemeyer, & Durvasula, 1990).
The relationship between message believability and attitude and behavior change has
been explored in various contexts including advertising, political campaign, and social norm
message studies (Beltramini, 1988; Glazer, Smith, Atkin, & Hamel, 2010; O’Cass & Griffin,
2006). However, the role of message believability in stigma communication is not clear. It is
worthwhile to ask the following research question:
Research Question 1: Does message believability play a role in the effects of message
valence and message type on emotional reactions and stigma attitudes?

10

METHOD
Participants
The study was conducted with 305 college students in Tianjin and Beijing, China. Most
participants were of Han ethnicity with an average age of 20.15 years old (SD = 2.00). 34.43% of
the participants were male. The cities of Tianjin and Beijing are major destinations of rural-tourban migrant workers (Guan, 2011; Guan & Liu, 2013), so that the issue of stigma on migrant
workers is likely to occur.
Design
This study is a 2 (message valence: positive or negative) × 2 (message type: narrative or
non-narrative) between-subjects factorial design plus one no-message control group. Participants
self-selected to respond to an online survey powered by Qualtrics, with the survey link
distributed through an information board in universities. The survey was originally written in
English, and was translated and back translated by two research assistants who are fluent in both
English and Chinese (Brislin, 1970). Disagreements in translation were resolved.
Message Inductions
Before composing the messages, informal interviews with college students in Tianjin,
China were conducted by the author, with an intention to collect college students’ opinions and
experience concerning migrant workers. Immersion and crystallization procedures (Borkan,
1999) were employed to examine the information obtained through the interviews. The most
frequent themes categorized by the four stigma cues: mark, label, responsibility, and peril
(Smith, 2007), were used to compose the message inductions.
For narrative and non-narrative messages respectively, the statements in a positive and a
negative message are parallel, differing only in whether the content cues are de-stigmatizing or

11

stigmatizing. For positive and negative messages respectively, the arguments in a narrative and a
non-narrative message are equivalent in terms of the thesis and the amount, differing only in
whether the arguments are presented in a narrative or a non-narrative format.
The messages were originally written in English, and were translated and back translated
by two research assistants fluent in both English and Chinese. Disagreements were resolved. The
length of the four messages was the same both in English (142 words) and in Chinese (225
words). The messages were included in Appendix A.
A pre-test of the messages was conducted to evaluate the merit of the four message
inductions prior to collecting data on the main study. Forty college students in Tianjin, China,
were randomly assigned to read one of four messages, and were asked to rate the message on 7point Likert scales (from 1 = strongly disagree to 7 = strongly agree) with items asking message
valence, message type, and reading difficulty. Manipulation checks showed that positive
messages were perceived as more positive than negative messages, that the distinction between
narrative and non-narrative messages was clear, and that the four messages were comparably
clear and easy to understand.
Procedure
Participants were randomly assigned to one of four experimental conditions or a nomessage control condition. At the beginning of the survey, informed consent was obtained from
participants. Except for the control group, participants were then exposed to one message
randomly selected from the four messages. The message could be either positive or negative in
valence, and could be either narrative or non-narrative in type.

12

After the message induction and manipulation check questions, participants were asked to
indicate their emotional reactions, perception of message believability, and stigma attitudes on 7point Likert scales. Demographic information was collected at the end of the survey.
Measurement
Emotional reactions. Participants were asked to rate on 7-point emotion scales (from 1 =
not at all to 7 = very much). Emotions measured are disgust, anger, and fear. Each emotion was
measured with three items obtained from Dillard & Shen (2007): disgust (sickened, disgusted,
and revolted), anger (irritated, angry, and aggravated), and fear (fearful, afraid, and scared).
Reliability of the disgust, anger, and fear scales are Cronbach’s alpha = 0.87, 0.93, 0.91,
respectively.
Stigma attitudes. The scale measuring stigma attitudes toward migrant workers was
adapted from the Link & Phelan (2001) article, in which stigma was conceptualized to have four
dimensions: label, negative attributes, separation, and discrimination. Response to this scale was
marked on 7-point Likert Scales (from 1 = strongly disagree to 7 = strongly agree), with higher
scores indicating higher stigma attitudes toward migrant workers (Appendix B).
Confirmatory factor analysis showed that the only valid factor was separation, measured
with six items. Fit indices for this subscale of stigma attitudes are as follow: NFI = 0.97, CFI =
0.99, GFI = 0.98, AGFI = 0.95, RMSEA = 0.06. Reliability is Cronbach’s alpha = 0.78. This
sub-dimension, separation, was used in analyses to reflect stigma attitudes.
Message believability. Four 7-point semantic differential scale items adapted from
Beltramini (1988) were included: unrealistic---realistic, unlikely to happen---likely to happen,
convincing---unconvincing (recode), reasonable---unreasonable (recode). The last item was

13

excluded subsequently due to invalidity. Reliability of the message believability scale is
Cronbach’s alpha = 0.89.
Demographic information. Participants were asked about their age, gender, ethnicity,
and major in college.
RESULTS
Preliminary Analysis
Among the total 305 participants who entered the online study, 37 of them dropped out
after answering only a few questions, and 7 of them reported an age under 18. These responses
were excluded from further consideration. The numbers of cases in the five conditions are 50, 51,
54, 52, and 54, respectively. These 261 usable responses were analyzed by SPSS.
One-way ANOVAs were conducted to ensure randomization by entering demographic
variables as dependent variables, condition as factor. The results showed that there is no
significant difference for participants’ demographics among conditions. This showed that
participants in the five conditions were comparable.
Descriptive statistics of all variables by condition, and the overall correlation matrix were
presented in Table 1 and Table 2. Stigma attitudes toward migrant workers in the control
condition was a little above the midpoint of the scale (M = 4.17, SD = 0.59), which means that
without manipulation, respondents did not express strong stigma attitudes toward migrant
workers in this study. Mean scores for disgust, anger, and fear responses in control condition
were 1.50 (SD = 0.79), 1.45 (SD = 0.74), and 1.53 (SD = 0.89), respectively, meaning that
without manipulation, respondents showed little negative emotional reactions to migrant
workers.

14

Hypothesis Testing
A two-way ANOVA was conducted to test H1(a), H2(a)(i) and H2(b)(i), which are
concerned with the main effect of message valence and the interaction effect of message valence
by message type on disgust response. The data showed that message valence had a significant
main effect on disgust response, F (1, 203) = 70.66, p < 0.001, 𝜂 2 = 0.26. Participants who
received the positive message (M = 1.88, SD = 1.31) had significantly less disgust response
compared to those who received the negative message (M = 3.55, SD = 1.53), consistent with
H1(a). However, there was no main effect of message type or interaction effect of message
valence by message type on disgust response. Therefore, the data are inconsistent with H2(a)(i)
and H2(b)(i).
To test H1(b), H2(a)(ii) and H2(b)(ii), which are concerned with the main effect of
message valence and the interaction effect of message valence by message type on anger
response, a two-way ANOVA was conducted. The data showed that message valence had
significant main effect on anger response, F (1, 203) = 51.72, p < 0.001, 𝜂 2 = 0.20. Participants
who received the positive message (M = 2.01, SD = 1.39) had significantly less anger response
compared to those who received the negative message (M = 3.47, SD = 1.52), consistent with
H1(b). However, no main effect of message type or interaction effect of message valence by
message type on anger response was found, inconsistent with H2(a)(ii) and H2(b)(ii).
Similarly, a two-way ANOVA was conducted to test H1(c), H2(a)(iii) and H2(b)(iii),
which are concerned with the main effect of message valence and the interaction effect of
message valence by message type on fear response. The data showed that message valence had a
significant main effect on fear response, F (1, 203) = 20.17, p < 0.005, 𝜂 2 = 0.05. Participants
who received the positive message (M = 1.65, SD = 1.07) had significantly less fear response

15

compared to those who received the negative message (M = 2.28, SD = 1.51), consistent with
H1(c). However, there was no main effect of message type or interaction effect of message
valence by message type on fear response. Thus, the data are inconsistent with H2(a)(iii) and
H2(b)(iii).
To test H1(d), H2(a)(iv) and H2(b)(iv) about the main effect of message valence and the
interaction effect of message valence by message type on stigma attitudes, a two-way ANOVA
was employed. The results showed that there was a significant main effect of message valence on
stigma attitudes, F (1, 203) = 47.43, p < 0.001, 𝜂 2 = 0.39, consistent with H1(d). Additionally,
there was a significant interaction effect of message valence by message type on stigma attitudes,
F (1, 196) = 2.97, p < 0.005, 𝜂 2 = 0.02. To further specify the interaction effect, independent
sample t-tests were conducted separately for positive and negative conditions. The results
showed that in positive conditions, participants who read the narrative message reported
significantly lower stigma attitudes (M = 2.99, SD = 0.66) compared to those who read the nonnarrative message (M = 3.32, SD = 0.48). However, no significant difference between the effects
of narrative and non-narrative messages on stigma attitudes was found in negative conditions.
Therefore, H2(b)(iv) was supported, while H2(a)(iv) was not.
Research Question Exploration
Message believability was found to significantly differ depending on message valence
and message type. To specify, message believability was perceived to be significantly higher in
positive conditions (M = 4.95, SD = 1.29) than negative conditions (M = 3.51, SD = 1.28).
Additionally, narrative messages (M = 4.11, SD = 0.94) were perceived to have significantly
higher message believability than non-narrative messages (M = 2.89, SD = 1.29) in negative
conditions, but not in positive conditions.

16

There was also a significant correlation between message believability and stigma
attitudes in negative conditions (r = 0.30, p < 0.005), but not in positive conditions. In negative
conditions, narrative messages were associated significantly with higher message believability (r
= 0.48, p < 0.001), which further was associated significantly with higher stigma attitudes.
However, in negative conditions, narrative messages did not result in significantly higher stigma
attitudes directly. Therefore, it is plausible to conjecture that in negative conditions, the effect of
narrative messages on stigma attitudes was mediated by message believability. To test this
mediation model (Figure 1), a path analysis was conducted with PATHE. The obtained
correlation (r = 0.14) between message type and stigma attitudes equals the predicted correlation
(r = 0.14) between them, which produces an error of zero. Therefore, the data are consistent with
this mediation model.
DISCUSSION
Since the implementation of Reform and Opening-Up policy in 1980s, Chinese society
has been experiencing huge social transitions in various domains, where all kinds of social
changes, conflicts, and problems are intensified. Social transitions are still taking place today in
China. The social group of rural-to-urban migrant workers has been a double-edged sword to the
development of China (Guan, 2011; Lin et al., 2011; Wang et al., 2010). On the one hand,
migrant workers serve as an indispensable labor force contributing to rapid economic growth and
urbanization progress. On the other hand, their marginalized, disadvantaged, and embarrassing
social situations in urban cities have posed great challenges not only to their wellbeing, but also
to the operation and management of the whole society (Wang et al., 2010; Wong et al., 2007;
Zhang et al., 2009). Additionally, large-scale population migration caused by this social group
have had significant impact on social structure, mobility, and stability.

17

Migrant workers in China have received a lot of attention. Studies were conducted to
investigate various sociological issues related to migrant workers (Chen et al., 2011; Guan, 2011;
Lin et al., 2011; Wang et al., 2010; Wong et al., 2007; Zhang et al., 2009; Zhang & Luo, 2012).
Previous studies were predominately qualitative inquiries of perceived stigma from the
perspective of migrant workers themselves (Guan, 2011; Guan & Liu, 2013; Wang et al., 2010).
The current study adds to the knowledge about social stigma towards migrant workers from
perspective of the stigmatizer, providing quantitative empirical evidence for ways in which
migrant workers are stigmatized.
Theoretical and Practical Implications
The results of the current study showed that in the control condition, participants did not
express strong stigma attitudes. This result contradicts claims made in existing qualitative studies
(Guan, 2011; Guan & Liu, 2013; Wang et al., 2010). One possible explanation is that there is a
discrepancy between enacted stigma by urban citizens and perceived stigma by migrant workers.
Discrepant results between stigmatizer’s enacted stigma and recipients’ experience of stigma are
often reported in stigma studies (Lekas, Siegel, & Schrimshaw, 2006). Another plausible
explanation is that participants in this study suppressed their stigma attitudes from explicit
expression, in consideration of social desirability. It is also possible that college student sample
may not share a broader stigma held by other segments of the Chinese urban population.
The effects of message valence on emotional reactions and stigma attitudes found in this
study were consistent with previous research (Smith, 2007, 2010). Participants who received
positive messages with de-stigmatizing cues reported significantly lower disgust, anger, and fear
responses, as well as lower stigma attitudes, compared to those who received negative messages
with stigmatizing cues. The effect sizes found in this study are very considerable. Additionally,

18

comparisons with the control condition indicate that while negative messages are more effective
in inducing negative emotional responses, positive messages are more effective in reducing
stigma attitudes.
Another finding worth noticing is that narrative messages were not able to significantly
elicit more valence-consistent emotional responses than non-narrative messages. This result is
inconsistent with previous literature on the important role of emotions in narrative message
effects (McQueen et al., 2011; Nabi & Green, 2015; Oliver et al., 2012). One plausible
explanation is concerned with the participants in this study being a Chinese population. In
Chinese culture, it is not appropriate to display or verbalize intense emotions, especially for
negative emotions. Studies on display rules have demonstrated different cultural norms for
emotional expression (Matsumoto, 1990; Matsumoto et al., 2008). It is possible that participants
who read narrative messages did experience higher level of emotional responses, but they did not
explicitly express or verbalize them in self-report measures, in accordance to cultural norms of
emotional display. However, this explanation requires additional investigation.
The effects of message type on stigma attitudes were very interesting. On the one hand,
narrative messages have significant direct effects in reducing stigma attitudes in positive
conditions, which provided some support for the social influence power of narrative messages in
attitude change (Bilandzic & Busselle, 2013; Braddock & Dillard, 2016; McQueen et al., 2011;
Moyer-Gusè et al., 2011; Murphy et al., 2015; Niederdeppe et al., 2011; Oliver et al., 2012). To
reduce stigma attitudes through communicating positive messages, narrative is more effective
than a non-narrative type of message. In other words, it is justified in this study to employ
narrative messages in developing relevant interventions to combat stigma attitudes toward ruralto-urban migrant workers in China.

19

On the other hand, narrative messages demonstrated an indirect effect mediated by
message believability in inducing stigma attitudes in negative conditions. This result provided a
new understanding of the mechanism under which stigma is communicatively enacted. Zhu &
Smith (2016) discussed rumor as a mechanism of stigma communication, as interpersonal
sharing of stigma messages contributes to the diffusion and adoption of stigma through
communities. The current study adds to this knowledge. In order to elicit stigma attitudes,
stigmatizing messages need to be perceived as believable, and a narrative type of message can
increase the perception of message believability. This explains why many rumors take the form
of storytelling to appear more believable, and further influence people’s perceptions and
attitudes.
Why are there different paths through which narrative messages influence stigma
attitudes in positive and negative conditions? To paraphrase, why does message believability
matter in negative conditions? One possible explanation is to balance the consequences of
message processing. It is assumed that reading a message can potentially lead to changes in the
recipients, including attitudes, beliefs, and affects. Processing of messages can have either
positive or negative hedonic consequences for the recipients, and that people tend to achieve or
maintain positive states (Wegener & Petty, 1994; Wegener, Petty, & Smith, 1995). In this study,
participants in the control condition had almost no negative emotional responses and low stigma
attitudes. When participants received negative messages, which signals potential negative
changes in affects and attitudes, participants are likely to employ coping strategies to maintain
their relatively positive states. There are two possible strategies: one is to disengage oneself from
careful processing of the message; the other is to critically evaluate the message with an
intention to counter-argue (Nabi, 2002). However, stigmatizing messages signal serious threats

20

to group functioning. From the evolutionary perspective taken by Smith (2007), negative
messages with stigmatizing cues are more important for one’s survival so that people should pay
attention. In this case, the second strategy, to process negative messages in more critical ways, is
employed. One way to do so is to evaluate the believability of the message. If the message
believability is high, the signaled potential group threats should not be neglected, and the
intention to maintain positive states becomes secondary. If the message believability is low, the
signaled potential group threats are ignored, and one’s positive states are maintained. On the
contrary, when participants received positive messages, since their states can change in positive
directions, they would not engage in critical evaluation of message believability.
Limitations and Future Directions
There are several limitations existing in the current study. Firstly, the stigma measure in
the current study is a self-report explicit measure. However, participants may be unlikely to selfreport they have stigma because of social desirability. Future studies are encouraged to include
implicit measures of stigma as well, such as Implicit Association Test (Stier & Hinshaw, 2007;
Wang, Brownell, & Wadden, 2004). Similarly, for emotional response measures, self-report has
similar limitations.
Secondly, there is a question of whether narrative messages are equivalent to nonnarrative messages, since the character in the former type of message is an individual, while the
character in the latter is a group. Although it is a potential threat to internal validity of the study
design, it serves the ecological validity in that it represents the character included in each type of
message in the real world. In this sense, the choice between internal and ecological validity in the
current study is understandable, but future studies could explore to address this issue.

21

Thirdly, three of the four dimensions in the stigma attitudes scale are invalid in this study.
On the one hand, it could be that the items in the three invalid dimensions need improvement. On
the other hand, it is possible that the construct of stigma is not culturally equivalent. In other
words, the conceptualization by Link & Phelan (2001) is based on a western perspective, which
may not be the same in Chinese contexts. The cultural equivalence of stigma as well as
developing a culturally appropriate measure of stigma are worth exploring in future studies.
There is also a limitation existing in current experimental design. Participants in the
negative conditions are elicited with negative emotional responses by the experimental
inductions. It is suggested in future studies that participants be debriefed after the experiment. A
positive message can be presented at the end to those participants in the negative conditions in
order to reduce negative impacts of the experimental inductions on the participants.
Additionally, the current study was not able to include possible covariates such as
familiarity with migrant workers, prior experience with migrant workers, prior message
exposure, existing stigma attitudes, etc. Although they are not the major interests of current
study, and they are assumed to be randomly distributed among conditions, it would be very
interesting to include these variables in future inquiries.
Finally, the use of a college student sample limits the external validity of the research
findings. In terms of the issue of stigma against rural-to-urban migrant workers in China, the
stigmatizer population are urban citizens. College students are not representative of the
population in terms of age, education, occupation, social relations with migrant workers, etc.
This population likely affected the magnitude of the scores for dependent variables as well.
Future studies could use a more representative sample to see if the research findings in current
study are replicable and generalizable.

22

CONCLUSION
While participants in the control condition did not report strong stigma attitudes and high
negative emotional reactions (i.e. disgust, anger, and fear) toward migrant workers in this study,
experimental inductions did show interesting effects. First, message valence had significant main
effects on emotional reactions and stigma attitude. Positive messages with de-stigmatizing cues
led to significantly lower stigma attitudes and less negative emotional reactions, compared to
negative messages with stigmatizing cues. Second, there was an interaction effect of message
valence by message type on stigma attitudes. Narrative messages led to significantly lower
stigma attitudes, compared to non-narrative messages, in positive message conditions. Third, in
negative message conditions, the effects of narrative messages in predicting stigma attitudes
were mediated by the perception of message believability.

23

APPENDICES

24

APPENDIX A: Message Inductions
I. Non-narrative Message
A. positive condition
Migrant workers are usually hard-working people. The reason why their clothes might be dirty is
that they are usually employed in laborious jobs with a dirty working environment. Migrant
workers do a great favor to urban citizens by doing these jobs. These jobs are undesirable for
urban citizens, but they are necessary for the proper functioning of the society. The
disadvantageous situation of migrant workers in urban cities is resulted by factors beyond their
control. Since migrant workers come from rural areas, they are very modest and tend to consider
other people’s convenience before their own. Migrant workers are also very innocent, honest and
kind-hearted people. They are a group of people very easy to get along with. Urban citizens are
encouraged to have interactions with migrant workers. Urban citizens could even learn
something from the good qualities that migrant workers possess.
B. negative condition
Migrant workers are usually cheap laborers. The reason why their clothes might be dirty is that
they are dirty and inferior people. Urban citizens do great favors to migrant workers by providing
them with these job opportunities. These jobs do not require any skill and do not contribute to the
proper functioning of the society. The disadvantageous situation of migrant workers in urban
cities is resulted by their own choices. Since migrant workers come from rural areas, they are just
uneducated and rude and tend to behave in very selfish and inconsiderate ways. Migrant workers
are also greedy, dishonest, and dangerous people. They are a group of people that urban citizens
should keep distance from. Urban citizens are warned to avoid any contact with migrant workers.

25

Urban citizens should be wary of migrant workers to protect themselves from being victims of
crimes.
II. Narrative Message
A. positive condition
Once I was riding on the subway, a young man in a uniform with City Construction logo got on.
I noticed that his uniform was covered with mud. I was quite sure that he was a migrant worker
building the overpass bridge near campus, initiated for traffic safety. His job must have kept him
from dressing tidy. There were many people on the subway, and the only available seat was right
next to me. As soon as the migrant worker sat down, he shrank his body to avoid contaminating
my clothes with the mud on his uniform. A while later, I left my seat to get off. Then, I heard
someone calling me, “madam, I guess this is your cellphone.” I realized the migrant worker was
returning my cellphone that I left on my seat. I smiled and tapped on his shoulder.
B. negative condition
Once I was riding on the subway, a young man in a uniform with City Construction logo got on.
I noticed that his uniform was covered with mud. I was quite sure that he was a migrant worker
building the overpass bridge near campus, which had resulted in many inconvenient detours. He
could have dressed himself tidy. There were many people on the subway, and the only available
seat was right next to me. As soon as the migrant worker sat down, I smelled terrible body odor,
and his dirty clothes were contaminating my clothes. A while later, I felt a tug on my pocket and
realized the migrant worker was trying to steal my cellphone. I gave him a stare, and he said
rudely, “What? Is there a problem with you?” I knitted my brows and ran away from him
quickly.

26

APPENDIX B:
Scale of Stigma Attitudes toward Migrant Workers in China
(Rate from 1=Strongly Disagree to 7=Strongly Agree)
Label
1. People think migrant workers are foreigners.
2. People think migrant workers are cheap labor.
3. People think being migrant workers are criminals.
4. People think being migrant workers are a burden on society.
Negative Attributes
5. People think migrant workers are uneducated.
6. People think migrant workers are inferior.
7. People think migrant workers are dirty.
8. People think migrant workers are dangerous.
9. People think migrant workers are rude.
10. People think migrant workers are unhealthy.
Separation
11. People like me are not willing to have a meal with a migrant worker.
12. People like me are very different from migrant workers.
13. People like me are not willing to have a romantic partner who is a migrant worker.
14. Migrant worker should keep distance from people like me.
15. People like me feel uncomfortable being around a migrant worker.
16. Migrant workers do not belong to the same group with people like me.
Discrimination

27

17. Migrant workers are financially disadvantaged.
18. Migrant workers live in the bottom of the social hierarchy.
19. Migrant workers are excluded from access to various social resources.
20. Migrant workers experience discrimination.
21. The overall social environment is unfavorable for migrant workers.
22. Many policies are disadvantageous for migrant workers.

28

Table 1: Descriptive Statistics for All Variables by Condition
Message

Positive

Positive

Negative

Negative

Non-Narrative

Narrative

Non-Narrative

Narrative

Control

Overall

Disgust

Anger

Fear

Stigma

Believability

1.96

2.14

1.73

3.32

4.84

(1.42)

(1.51)

(1.19)

(0.48)

(1.40)

1.81

1.90

1.58

3.00

5.06

(1.21)

(1.28)

(0.94)

(0.66)

(1.19)

3.71

3.72

2.04

4.04

2.89

(1.65)

(1.56)

(1.37)

(0.50)

(1.29)

3.40

3.22

2.52

4.19

4.11

(1.40)

(1.45)

(1.61)

(0.54)

(0.94)

1.50

1.45

1.53

4.17

----

(0.79)

(0.74)

(0.89)

(0.59)

----

2.46

2.47

1.88

3.74

4.23

(1.59)

(1.57)

(1.27)

(0.77)

(1.47)

Note. Standard deviations are in the parentheses. N = 261.

29

Table 2: Overall Correlation Matrix
1
(1) Valence

2

3

4

5

7

1.00

(2) Type

.01

(3) Disgust

-.51**

-.08

.87

(4) Anger

-.45**

-.12

.87**

.93

(5) Fear

-.24**

.06

.60**

.63**

.91

(6) Stigma

-.63**

-.07

.18**

.11

.11

.25**

-.47**

-.43**

(7) Message
Believability

6

.49**

1.00

-.12

0.78
-.24**

.89

Note. For valence, 1 = negative, 2 = positive. For type, 1 = non-narrative, 2 = narrative. For
others, higher scores mean higher level of respective variables. Reliabilities of scales are in the
diagonal. ** p < 0.005 (2-tailed).

Figure 1: The Effect of Message Type on Stigma in Negative Conditions

Message
Type

0.48

Message
Believability

30

0.30

Stigma
Attitude

REFERENCES

31

REFERENCES

Andrews, J. C., Netemeyer, R. G., & Durvasula, S. (1990). Believability and attitudes toward
alcohol warning label information: the role of persuasive communications
theory. Journal of Public Policy & Marketing, 9, 1-15. Retrieved from
http://www.jstor.org.proxy1.cl.msu.edu/stable/pdf/30000193.pdf
Appel, M., & Richter, T. (2010). Transportation and need for affect in narrative persuasion: A
mediated moderation model. Media Psychology, 13, 101-135.
doi:10.1080/15213261003799847
Banerjee, S. C., & Greene, K. (2012). Role of transportation in the persuasion process: Cognitive
and affective responses to antidrug narratives. Journal of Health Communication, 17,
564-581. doi:10.1080/10810730.2011.635779
Beltramini, R. F. (1982). Advertising perceived believability scale. In Corrigan, D. R., Kraft, F.
B., & Ross, R. H. (Eds.) Proceedings of the Southwestern Marketing Association.
Wichita, KS: Wichita State University.
Beltramini, R. F. (1988). Perceived believability of warning label information presented in
cigarette advertising. Journal of Advertising, 17, 26-32. Retrieved from
http://www.tobaccolabels.ca/wp/wp-content/uploads/2013/12/USA-1988-PerceivedBelievability-of-Warning-Label-Information-Presented-in-Cigarette-Advertising-Jour-ofAdvertising-Beltramini.pdf
Bettman, J. R., Payne, J. W., & Staelin, R. (1986). Cognitive considerations in designing
effective labels for presenting risk information. Journal of Public Policy & Marketing, 5,
1-28. Retrieved from https://faculty.fuqua.duke.edu/~jrb12/bio/Jim/26.pdf
Bilandzic, H., & Busselle, R. W. (2013). Narrative persuasion. In Dillard, J.P. & Shen, L. (Eds.),
The Sage handbook of persuasion: Developments in theory and practice (pp. 200–219).
Thousand Oaks, CA: Sage.
Borkan, J. (1999). Immersion/Crystallization. In Crabtree, B.F. & Miller, W. L. (Eds.) Doing
Qualitative Research (2nd Edition). pp. 179-194. Thousand Oaks, CA: Sage Publications.
Braddock, K. H., & Dillard, J. P. (2016). Meta-analytic evidence for the persuasive effect of
narratives on beliefs, attitudes, intentions, and behaviors. Communication Monographs,
83, 1-24. doi:10.1080/03637751.2015.1128555
Brislin, R. W. (1970). Back-translation for cross-cultural research. Journal of cross-cultural
psychology, 1, 185-216. Retrieved from
http://journals.sagepub.com.proxy1.cl.msu.edu.proxy2.cl.msu.edu/doi/pdf/10.1177/13591
0457000100301

32

Chen, X., Stanton, B., Kaljee, L. M., Fang, X., Xiong, Q., Lin, D., ... Li, X. (2011). Social
stigma, social capital reconstruction, and rural migrants in urban China: A population
health perspective. Human Organization, 1, 22-32.
doi:10.17730/humo.70.1.k76047734m703500
Chung, A. H., & Slater, M. D. (2013). Reducing stigma and out‐group distinctions through
perspective‐taking in narratives. Journal of Communication, 63, 894-911.
doi:10.1111/jcom.12050
Cottrell, C. A., & Neuberg, S. L. (2005). Different emotional reactions to different groups: A
sociofunctional threat-based approach to “prejudice”. Journal of Personality and Social
Psychology, 88, 770–789. doi:10.1037/0022-3514.88.5.770
De Graaf, A., Hoeken, H., Sanders, J., & Beentjes, H. (2009). The role of dimensions of
narrative engagement in narrative persuasion. Communications, 34, 385-405.
doi:10.1515/comm.2009.024
De Graaf, A., Sanders, J., & Hoeken, H. (2016). Characteristics of narrative interventions and
health effects: A review of the content, form, and context of narratives in health-related
narrative persuasion research. Review of Communication Research, 4, 88-131.
http://dx.doi.org/10.12840/issn.2255-4165.2016.04.01.011
Fishbein, M. & Ajzen, I. (1975). Belief, Attitude, Intention, and Behavior: An Introduction to
Theory and Research. Reading, MA: Addison-Wesley.
Glazer, E., Smith, S. W., Atkin, C., & Hamel, L. M. (2010). The effects of sensation seeking,
misperceptions of peer consumption, and believability of social norms messages on
alcohol consumption. Journal of health communication, 15, 825-839.
doi:10.1080/10810730.2010.522222
Goffman, E. (1963). Stigma: Notes on the management of spoiled identity. Englewood Cliffs, NJ:
Prentice-Hall.
Green, M. C. (2004). Transportation into narrative worlds: The role of prior knowledge and
perceived realism. Discourse Processes, 38, 247-266. doi:10.1207/s15326950dp3802_5
Guan, J. (2011). Stigma toward the rural-to-urban migrants in China: A qualitative study.
Interpersona, 1, 21-37.
http://search.proquest.com/openview/21eee80a99aabd82372ed30409227a57/1?pqorigsite=gscholar
Guan, J., & Liu, L. (2013). Recasting stigma as a dialogical concept: A case study of rural-tourban migrants in china: Recasting stigma as a dialogical concept. Journal of Community
& Applied Social Psychology, 24, 75-85. doi:10.1002/casp.2152

33

Hinyard, L.J., & Kreuter, M.W. (2006). Using narrative communication as a tool for health
behavior change: a conceptual, theoretical, and empirical overview. Health Education
and Behavior, 34, 777–792. doi:10.1177/1090198106291963
Hoeken, H., & Sinkeldam, J. (2014). The role of identification and perception of just outcome in
evoking emotions in narrative persuasion. Journal of Communication, 64, 935-955.
doi:10.1111/jcom.12114
Iguarta, J. J. (2010). Identification with characters and narrative persuasion through fictional
feature films. Communications, 35, 347–373. doi:10.1515/comm.2010.019
Kreuter, M. W., Green, M. C., Cappella, J. N., Slater, M. D., Wise, M. E., Storey, D., ... &
Woolley, S. (2007). Narrative communication in cancer prevention and control: A
framework to guide research and application. Annals of Behavioral Medicine, 33, 221235. doi:10.1007/BF02879904
Lekas, H. M., Siegel, K., & Schrimshaw, E. W. (2006). Continuities and discontinuities in the
experiences of felt and enacted stigma among women with HIV/AIDS. Qualitative health
research, 16, 1165-1190. doi:10.1177/1049732306292284
Lemal, M., & Van den Bulck, J. (2010). Testing the effectiveness of a skin cancer narrative in
promoting positive health behavior: A pilot study. Preventive Medicine, 51, 178-181.
doi:10.1016/j.ypmed.2010.04.019
Lin, D., Li, X., Wang, B., Hong, Y., Fang, X., Qin, X., & Stanton, B. (2011). Discrimination,
perceived social inequality, and mental health among rural-to-urban migrants in China.
Community Mental Health Journal, 47, 171-180. doi:10.1007/s10597-009-9278-4
Link, B. G., & Phelan, J. C. (2001). Conceptualizing stigma. Annual Review of Sociology, 27,
363-382. doi:10.1146/annurev.soc.27.1.363
Matsumoto, D., Takeuchi, S., Andayani, S., Kouznetsova, N., & Krupp, D. (1998). The
contribution of individualism vs. collectivism to cross‐national differences in display
rules. Asian Journal of Social Psychology, 1, 147-165. doi:10.1111/1467-839X.00010
Matsumoto, D., Yoo, S. H., & Fontaine, J. (2008). Mapping expressive differences around the
world the relationship between emotional display rules and individualism versus
collectivism. Journal of cross-cultural psychology, 39, 55-74.
doi:10.1177/0022022107311854
McQueen, A., Kreuter, M. W., Kalesan, B., & Alcaraz, K. I. (2011). Understanding narrative
effects: The impact of breast cancer survivor stories on message processing, attitudes, and
beliefs among African American women. Health Psychology, 30, 674-682.
doi:10.1037/a0025395

34

Meisenbach, R. J. (2010). Stigma management communication: A theory and agenda for applied
research on how individuals manage moments of stigmatized identity. Journal of Applied
Communication Research, 38, 268-292. doi: 10.1080/00909882.2010.490841
Moyer-Gusè, 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, 387–406. doi:10.1111/j.1460-2466.2011.01551.x
Murphy, S. T., Frank, L. B., Chatterjee, J. S., Moran, M. B., Zhao, N., de Herrera, P. A., &
Baezconde-Garbanati, L. A. (2015). Comparing the relative efficacy of narrative versus
nonnarrative health messages in reducing health disparities using a randomized trial.
American Journal of Public Health, 105, 2117-2123. Retrieved from
http://search.proquest.com.proxy1.cl.msu.edu/docview/1713658247/fulltextPDF/B49DD
B0512D34335PQ/1?accountid=12598
Murphy, S., Frank, L., Chatterjee, J., & Baezconde-Garbanati, L. (2013). Narrative versus
nonnarrative: The role of identification, transportation, and emotion in reducing health
disparities. Journal of Communication, 63, 116–137. doi:10.1111/jcom.12007
Nabi, R. L. (2002). Anger, fear, uncertainty, and attitudes: A test of the cognitive-functional
model. Communication Monographs, 69, 204-216. doi:10.1080/03637750216541
Nabi, R. L., & Green, M. C. (2015). The role of a narrative's emotional flow in promoting
persuasive outcomes. Media Psychology, 18, 137-162.
doi:10.1080/15213269.2014.912585
Niederdeppe, J., Shapiro, M. A., & Porticella, N. (2011). Attributions of responsibility for
obesity: Narrative communication reduces reactive counterarguing among
liberals. Human Communication Research, 37, 295-323. doi:10.1111/j.14682958.2011.01409.x
O'Cass, A., & Griffin, D. (2006). Antecedents and consequences of social issue advertising
believability. Journal of Nonprofit & Public Sector Marketing, 15, 87-104.
doi:10.1300/J054v15n01_05
Oliver, M. B., Dillard, J. P., Bae, K., & Tamul, D. (2012). The effect of narrative news format on
empathy for stigmatized groups. Journalism & Mass Communication Quarterly, 89, 205–
224. doi:10.1177/ 1077699012439020
Petty, R. E. & Cacioppo, J. T. (1986). Issue involvement as a moderator of the effects on attitude
of advertising and content. In Monroe, K. (Ed.), Advances in Consumer Research (pp.
20-24). Association for Consumer Research.
Smith, R. A. (2007). Language of the lost: An explication of stigma communication.
Communication Theory, 17, 462-485. doi:10.1111/j.1468-2885.2007.00307.x

35

Smith, R. A. (2014). Testing the model of stigma communication with a factorial experiment in
an interpersonal context. Communication Studies, 65, 154-173.
doi:10.1080/10510974.2013.851095
Stier, A., & Hinshaw, S. P. (2007). Explicit and implicit stigma against individuals with mental
illness. Australian Psychologist, 42, 106-117. Retrieved from
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.504.4474&rep=rep1&type=pdf
Wang, S. S., Brownell, K. D., & Wadden, T. A. (2004). The influence of the stigma of obesity on
overweight individuals. International journal of obesity, 28, 1333-1337.
doi:10.1038/sj.ijo.0802730
Wang, B., Li, X., Stanton, B., & Fang, X. (2010). The influence of social stigma and
discriminatory experience on psychological distress and quality of life among rural-tourban migrants in china. Social Science & Medicine, 71, 84-92.
doi:10.1016/j.socscimed.2010.03.021
Wegener, D. T., & Petty, R. E. (1994). Mood management across affective states: The hedonic
contingency hypothesis. Journal of Personality and Social Psychology, 66,1034-1048.
Retrieved from https://search-proquestcom.proxy1.cl.msu.edu/docview/209797432?accountid=12598
Wegener, D. T., Petty, R. E., & Smith, S. M. (1995). Positive mood can increase or decrease
message scrutiny: The hedonic contingency view of mood on message processing.
Journal of Personality and Social Psychology, 69, 5- 15. Retrieved from https://searchproquest-com.proxy1.cl.msu.edu/docview/1295928845?accountid=12598
Wong, D. F. K., Chang, Y. L., & He, X.S. (2007). Rural migrant workers in urban China: living
a marginalized life. International Journal of Social Welfare, 16, 32-40.
doi:10.1111/j.1468-2397.2007.00475.x
Zhang, J., Li, X., Fang, X., & Xiong, Q. (2009). Discrimination experience and quality of life
among rural-to-urban migrants in china: The mediation effect of expectation-reality
discrepancy. Quality of Life Research, 18, 291-300. doi:10.1007/s11136-009-9454-6
Zhang, S., & Luo, G. (2012). China’s migrant workers: How far from being citizens? Asian
Social Science, 9, 171-178. doi:10.5539/ass.v9n1p171
Zhu, X., & Smith, R. (2016). Advancing Research on the Spread of Stigmatizing Beliefs with
Insights from Rumor Transmission. American Behavioral Scientist, 60, 1342-1361.
doi:10.1177/0002764216657382

36