Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
A multilevel communication approach to understanding human trafficking prevention behaviors in Indonesia
(USC Thesis Other)
A multilevel communication approach to understanding human trafficking prevention behaviors in Indonesia
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
i
A MULTILEVEL COMMUNICATION APPROACH TO UNDERSTANDING
HUMAN TRAFFICKING PREVENTION BEHAVIORS IN INDONESIA
by
Prawit Thainiyom
________________________________________________________
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(COMMUNICATION)
August 2018
Copyright 2018 Prawit Thainiyom
ii
DEDICATION
This dissertation is dedicated to my mom, Pheeraya Jongjaroenmaneekun, who
encouraged me to pursue the highest level of education because she never had an opportunity to
study beyond an elementary school. This journey could not have been done without you.
iii
ACKNOWLEDGMENTS
First and foremost, I would like to thank my advisor Patricia Riley for guiding me
throughout this dissertation process. This work would also not be possible without the
contributions from Sandra Ball-Rokeach, Lourdes Baezconde-Garbanati, Rhacel Parrenas,
Magaret McLaughlin, Mark Latonero, and Shiela Murphy. It has been a challenging, but highly
rewarding project in which our research output will continue to benefit the international anti-
trafficking community.
I have to thank my friends and colleagues at MTV EXIT, particularly Matt Love, Alex
Heath, Simon Goff, Tara Dermott, and Ruici Tio for collaborating with me to make this project a
reality. In addition, this project could not have been done without the dedication from partners in
Indonesia: Audrey Eclat, Sumadi Wijaya, Yudi, Muhammad Rizal, Nithin Coca, Hersinta, Olivia
Hutagaol, Renold Lim, Nurul Qoiriah, Dewi Ratnawulan, and Daniel Lindgren. Thank you so
much for helping me to navigate this important work in your country.
I am blessed to have a second family of friends and loved ones during this PhD program.
They are Todd Bartoo, Martin Huack, Peter Stark, Frank Gaugler, Marc Caldwell, and Wen-Chia
Chang. I am forever grateful for all of your emotional support. To my USC Annenberg family:
Nahoi Koo, Bei Yan, Wei Wang, Nan Zhao, Diana Lee, Ritesh Mehta and Erin Kamler. You are
the foundation of my intellectual curiosity and sanity!
This work was made possible with support from the American people through the United
States Agency for International Development (USAID) under the Democracy Fellows and
Grants Program (AID-0AA-A-12-0039), the USC Graduate School, and the Annenberg Doctoral
Student Summer Research Fellowship. The contents are the sole responsibility of the author and
iv
do not necessarily reflect the views of USAID; the United States Government; or the Democracy
Fellows and Grants Program implementer, Institute of International Education.
v
TABLE OF CONTENTS
Dedication ii
Acknowledgments iii
List of Tables vi
List of Figures vii
Abstract viii
Chapter 1: Introduction 1
Current Research on Human Trafficking 3
Research on Human Trafficking Prevention Campaigns 5
Gaps in Empirical Research on Human Trafficking Prevention 9
Research Objectives and Organization of the Dissertation
14
Chapter 2: Prevalence and Experiences of Human Trafficking Victims in Indramayu,
Indonesia
18
Methods 30
Results 36
Discussion 43
Limitations and Future Direction 47
Chapter 3: Assessing the Effects of a Human Trafficking Prevention Campaign and Its
Determinants through a Multilevel Communication Model
50
Methods 58
Results 66
Discussion 73
Limitations and Future Direction
76
Chapter 4: The Narrative Effects and the Role of Transportation and Identification in
Changing Human Trafficking Prevention Outcomes
78
Methods 83
Results 91
Discussion 95
Limitations and Future Direction
102
Chapter 5: Conclusions and Recommendations
104
References
111
Appendix I: Survey Instruments 132
Appendix II: Focus Group Questions
147
vi
LIST OF TABLES
Table 2.1. Characteristics of the participants by sex
37
Table 2.2. Calculation of the estimates and confidence intervals for the number of
trafficked persons in Indramayu
38
Table 2.3. Distribution count and percent of the participants who had experienced
human trafficking conditions
39
Table 2.4. Experiences of human trafficking by sex in Indramayu
40
Table 2.5. Experiences of escape or rescue, protection and justice, and health problems
among self-identified human trafficking victims in Indramayu
42
Table 3.1. Characteristics of the participants in Indramayu by group assignment and
data collection period
67
Table 3.2. Documentary effects on knowledge, attitudes, norms, efficacy, perceived
risks, skills & abilities, intention, and behavior at 1-week and 4-month
follow-up
69
Table 3.3. Unstandardized OLS regression coefficients (standard errors) predicting
knowledge, attitudes, norms, efficacy, perceived risks, skills & abilities,
intention, and behavior
72
Table 4.1. Means and standard deviations of change in knowledge, attitudes, norms,
intention and behavior by sex and experimental condition (narrative or
control)
91
Table 4.2. Means and standard deviations of transportation and identification with
characters of the documentary by sex
93
Table 4.3. Effects of transportation and identification with characters on knowledge,
attitudes, norms, intention, and behavior by sex (n=319)
93
vii
LIST OF FIGURES
Figure 1.1. Number of human trafficking citations from Google Scholar search from
2000-2017
4
Figure 1.2. The public health model of a scientific approach to prevention
15
Figure 2.1. Indramayu regency map
27
Figure 2.2. Multi-stage sampling procedure
31
Figure 2.3. Conceptualization and operationalization of human trafficking
35
Figure 3.1. A Multilevel Communication Model
53
Figure 3.2. Flowchart of the participants in the randomized control trial
59
Figure 3.3. Longitudinal mean scores of human trafficking prevention outcomes for the
exposed group
73
Figure 4.1. Flowchart of the participants in the mixed-method explanatory sequential
design
83
viii
ABSTRACT
This dissertation responds to the need to explore knowledge, attitudes, and behaviors
related to human trafficking prevention through a multilevel communication perspective. The
project is divided into three studies. The first study focuses on estimating the prevalence of
human trafficking victims in Indramayu, Indonesia. Results demonstrate that 15.4 percent of
Indramayu residents were estimated to be victims of human trafficking. Female migrants also
experienced higher prevalence of human trafficking, as well as encountered more restriction on
their movement, and exploitative labor practices. The second study evaluates the effects of an
anti-trafficking campaign and explores the determinants of human trafficking prevention through
a Multilevel Communication Model. Results reveal that campaign exposure was associated with
increase in knowledge, perceived risks, skills & abilities, efficacy and intention. Prior experience
with human trafficking and migration, and communication ecological factors were found to be
strong determinants of trafficking prevention-related outcomes. The third study investigates the
role of transportation and identification of the characters in the documentary in changing human
trafficking prevention outcomes. Results demonstrate that transportation increased favorable
attitudes and intention to prevent trafficking as well as decreased norms to migrate. Meanwhile,
identification with characters had differing effects on male and female participants. Specifically,
women who identified more with the female characters increased norms to migrate, and had less
knowledge, attitudes, and preventive behaviors. Recommendations for future research direction
and policy implications on the planning, implementation, and evaluation of human trafficking
prevention campaigns are also discussed.
1
CHAPTER 1: INTRODUCTION
The rise of globalization in the last few decades has triggered tremendous economic
growth at the global level. However, this growth widens disparity as most economic
opportunities are concentrated in urbanized centers in the more developed countries. Increased
inequality propels hundreds of millions of people around the world to migrate and seek jobs for a
better life. In the process of searching for new economic opportunities, some migrants become
victims of human trafficking when they are deceived into work that subjects them to threat,
violence and hazardous working conditions with little or no pay, and no freedom to escape (Anti-
Slavery International, 2006; Human Rights Watch, 2010, 2006, 2004; ILO, 2006a, 2006b; IOM,
2010; Liu, 2010). Other exploitative labor conditions experienced by victims of human
trafficking include excessive working days or hours, total restriction of movement, verbal and
physical abuse, confiscation of identification and traveling documents, and a general violation
against local labor laws such as lack of an employment contract (ILO, 2009; IOM, 2011).
The United Nations Palermo Protocol (2004) defines human trafficking as:
…the recruitment, transportation, transfer, harboring, or receipt of persons, by
means of the threat or use of force or other forms of coercion, of abduction, of
fraud, of deception, of the abuse of power or of a position of vulnerability or of the
giving or receiving of payments or benefits to achieve the consent of a person
having control over another person, for the purpose of exploitation. Exploitation
shall include, at a minimum, the exploitation or the prostitution of others or other
forms of sexual exploitation, forced labour or services, slavery or practices similar
to slavery, servitude, or the removal of organs. (p. 42)
2
The Palermo Protocol also includes individuals under the age of 18 who are recruited,
transported, harbored, or received by traffickers for the purpose of exploitation as victims even in
the instances where there is no use of force, coercion, fraud or deception (UNODC, 2004).
International Labour Organization (2017a) estimates that 40.3 million people around the
world are victims of human trafficking. The economic profits of human trafficking are estimated
to be US$150 billion every year (ILO, 2014) and it is the second most profitable criminal
industry in the world, after drug trafficking (Belser, 2005). Human trafficking can be partially
attributed to irregular migration and deceitful practices by recruitment agents who are looking to
make profits by trading human beings as cheap labor to employers that end up exploiting them
(GAATW, 2011). Apart from being a serious violation of international law and human rights,
trafficked victims, especially women and children, often face negative health outcomes as they
are subjected to physical violence, rape, repetitive stress injuries, back pain, and sexually
transmitted diseases such as HIV/AIDS (Orphant, 2001; Feingold, 2005; Zimmerman, et. al.,
2006;). Survivors also suffer psychological trauma such as intense feelings of shame, anxiety,
panic disorder, post-traumatic stress disorder, depression, drug abuse, and eating disorders (Stark
& Hodgson, 2003; Feeny, et al. 2004).
While human trafficking is a phenomenon that occurs in every part of the world,
Southeast Asia is recognized as one of the biggest hubs (Perry & McEwing, 2013). Indonesia,
the largest country in the region with over 250 million people (United Nations Development
Programme, 2018), is considered to be a major source of human trafficking as people migrate
from rural to urban destinations domestically or to foreign countries (Department of State, 2013).
Nine million Indonesian migrant workers are estimated to be working in 142 countries around
the world with Saudi Arabia, Malaysia, Singapore, and Hong Kong as top destinations (World
3
Bank, 2017). Over 714,000 Indonesians are estimated to be victims of human trafficking (Global
Slavery Index, 2014) and a majority of these victims experience human trafficking conditions
during their time as a migrant (IOM, 2012).
Human trafficking is also a gender issue as it affects women at a disproportionate rate. A
systematic review on social determinants of human trafficking in Southeast Asia found gender to
be the second most cited facilitator of trafficking after poverty (Perry & McEwing, 2013).
Indonesian migrant women are more susceptible to human trafficking as over 80% of all
Indonesia migrants are female (Wulan, 2014). Data from the IOM Counter Trafficking Unit from
2005-2012 further confirms this gender disparity as over 85% of rescued trafficked victims are
women. The local culture of Mluruduit, defined literally as to find and collect money for a better
future, are often pushed by parents onto their daughters and influence them to seek jobs abroad
or in other major cities within Indonesia (Bajari, 2013). As a result, many Indonesian women
migrate abroad to work as domestic maids in countries such as Malaysia, Hong Kong, and Saudi
Arabia while others migrate within Indonesia to engage in commercial sex work (Bajari, 2013).
Other social determinants identified to be important barriers or facilitators of human trafficking
are level of policy and enforcement, age, rates of migration, displacement and conflict, ethnicity,
culture, caste status, formal education, citizenship status, maternal education, and birth order
(Perry & McEwing, 2013).
Current Research on Human Trafficking
The review of recent literature on human trafficking reveals that the issue is under-
researched all around the world (Laczko, 2005; Mattar, 2004; Schauer and Wheaton, 2006). This
is because human trafficking research is a relatively new field where researchers have only
started to pay more attention since 2000 when high-level policy makers from over 80 countries
4
convened in Palermo, Italy for a United Nations Convention and signed on to “the Protocol to
Prevent, Suppress and Punish Trafficking in Persons, Especially Women and Children”
(Raymond, 2002). This measure was officially ratified in almost 90% of the United Nations
member states around the world, expanding the international efforts to address human trafficking
over a relatively short period. In the most recent policy development, provisions to combat
human trafficking was included in 3 of the 17 United Nations 2030 Sustainable Development
Goals (UNODC, 2016). These goals call for the UN member states to eliminate all forms of
violence, exploitation, and trafficking against women, girls, and children, as well as to take
immediate measures against forced labor, modern slavery, and human trafficking.
Research on human trafficking has been growing steadily since 2000. Figure 1 displays
the number of articles and citations found on Google Scholar Web Search with “human
trafficking” as a search term from 2000 to 2017. A total of 80,750 articles were found during this
period. This is still a relatively small number compared to a more established topic such as AIDS
which generates over 1.5 million articles during the same period.
Figure 1.1 Number of human trafficking citations from Google
Scholar search from 2000-2017.
5
Research on Human Trafficking Prevention Campaigns
Despite the scope of the human trafficking problem, awareness remains low, as
evidenced by a study in Indonesia where only 13% of the respondents were familiar with this
issue (Lindgren, 2010). The lack of awareness and negative attitudes toward marginalized
populations who are susceptible to human trafficking open doors for communication scholars to
collaborate with local and international anti-trafficking organizations, inter-governmental
organizations, government agencies, and community-based organizations to develop appropriate
communication strategies and campaigns to inform at-risk populations to prevent themselves
against human trafficking by seeking information about safe migration, discussing with their
peers and family about the risks and prevention methods, as well as reporting trafficking
incidents to the authorities. In addition, communication efforts should also influence decision
makers at societal and policy level to address gender inequality, sociocultural and structural
factors that continue to put marginalized populations at risk.
Human trafficking prevention campaigns are typically created with the goal of reducing
human trafficking incidents by targeting a sizable number of people within a particular time-
frame through a series of activities via communication and media channels and influencing them
to adopt preventive behaviors. Common anti-trafficking campaign activities include television
programs, public service announcements, radio drama serials, posters, billboards, awareness-
raising events, and peer education training in schools with the aim to educate target audience to
seek information on safe migration, call the migrant and trafficking hotline, register with migrant
organizations before traveling, and discuss how to solve the issue with people in their
community.
6
Research on human trafficking prevention campaigns has focused on critical analysis
through the theoretical lens of feminist and migration perspectives. Nieuwenhuys and Pécoud
(2007) criticize awareness-raising campaigns in Western Europe as a state-sanctioned tool for
migration control to discourage people from Eastern Europe to move into Western Europe.
Campaign messages often use fear to associate undocumented migration with human trafficking,
and indirectly inform potential migrants that the solution to prevent themselves from being
trafficked is to stay in their country of origin. In another study, Andrijasevic (2007) analyzes
visual representation of female victims in an anti-sex trafficking campaign by International
Organization for Migration (IOM) in Europe. She summarizes that female victims are presented
under “the paradigmatic image of a young and naïve innocent lured or deceived by evil
traffickers into a life of sordid horror from which escape is nearly impossible” (Doezema, 2000).
By overstating the dangers of migration, anti-trafficking campaigns suggest that women who
seek to migrate for work in Western Europe will eventually get themselves into forced
prostitution, a situation that will occur beyond their control. Therefore, they should remain in
their ‘safe’ home town and subjugate to the familiar role of heterosexual domesticity (Pollock,
1998).
This problematic depiction of female victims can be attributed to the limitations of
communication materials (e.g., posters, print ads, billboards) that normally allow a single image
and minimal accompanying text. In order to fit the dominant human trafficking narrative, anti-
trafficking campaigns have to succinctly display the ideal female victim image to communicate
that all forms of sex trafficking are prostitution and rape, despite contested research that claims
majority of prostitution are performed voluntarily (Bernstein, 2010). The widespread beliefs that
sex trafficking dominates the phenomenon of human trafficking are further supported by a
7
content analysis study on human trafficking news coverage that shows about half of these news
in the US, Great Britain, and Canada were focused on sex trafficking while 20% featured labor
trafficking, and the remaining covered human smuggling (30%), illegal adoption (3%), and
national security (1%) (Gulati, 2010). The news of sex trafficking of young women and children
often create public outrage, push anti-trafficking actors to champion against commercial sex
industry, and reinforce the stereotypes that human trafficking is caused by unscrupulous people,
brothels should be raided to rescue the victims, and traffickers must be arrested to solve the
problem.
The dominant sex trafficking discourse shifts away from the reality that human
trafficking is a process and outcome of globalization that triggers irregular migration by people
who voluntarily leave their town to seek better employment opportunities in more developed
areas. Many of these migrants initially go through formal labor recruitment agencies with the
intention of migrating safely and legally, but are deceived, get trafficked illegally across borders,
and end up in exploitative conditions in their destinations. They experience abusive practices
such as debt bondage, little or no wages, excessive working hours, and substandard health and
unsafe working conditions. A report by International Labor Organization (2017a) estimated that
a vast majority of victims of forced labor are in the category of labor exploitation (81%) rather
than sexual exploitation (19%). Victims of forced labor engage in economic activities such as
agriculture, manufacturing, construction, fisheries, and domestic work (ILO, 2012). The gravity
of exploitation also varies in each case of human trafficking where some migrants may not even
consider themselves as “victims” because they perceive themselves to be better off than being
unemployed or underemployed back home (Marshall, 2001).
8
Other scholar like Davidson (2010) argues that sex trafficking narrative is part of a neo-
abolitionist ideology that stymies efforts by immigrant groups to form collaboration with
important stakeholders in the society to counter human trafficking from a broader perspective of
social, economic, and political reform. The dominant display of what human trafficking victims
should look like through popular culture, news media, and anti-trafficking campaigns by the state
and non-profit sector has created different categories of who should be constituted as ‘deserving
victims’, leaving behind marginalized population (e.g., undocumented migrants who are
exploited in factories and farms under unsanitary working conditions) as ‘undeserving’ from
rights and freedom.
The consequences of neo-abolitionist campaigns include increased regulation of sexual
activities and enforcement of strict migration control under the disguise of preventing human
trafficking. This leads to initiation of tough immigration laws that bar unskilled workers from
migrating to the host country to work in ‘risky’ industries such as restaurants and entertainment
venues. In addition, undocumented workers who are currently exploited in the labor sector and
do not meet the criteria of sex trafficking victims are often arrested, fined, and deported without
due process (Aradau, 2004). In sharp contrast to the neo-abolitionist ideology, advocates for the
rights for migrant workers propose the adoption of a pro-rights ideology, in which human
trafficking must be addressed by supporting a larger immigrant population who are vulnerable to
exploitation due to limited immigration rights, lax regulation in labor laws, corruption within law
enforcement, and xenophobic attitudes against the migrant population. It is important that the
pro-migrant rights voices are heard in local and international arenas so that alternative ways of
tackling human trafficking, immigration control, migrant labor, and commercial sex work are
translated into policy. Yet, government policies and communication strategies have so far
9
hindered these possibilities (Shalit, Heynen & Meulen, 2014). Given the difficulty in enacting a
pro-rights ideology which requires a comprehensive reform in immigration, national security,
and labor policies, it is not a surprise that this approach has not found broad support among
policy makers and general public (Tomkinson, 2012). It is easier to frame counter human
trafficking argument under neo-abolitionist ideology and grant protection to a few ‘deserving’
victims without looking at the deeper systematic social, political, and environmental structures.
These factors continue to push people from poor, rural regions who are desperate for a better life
to migrate to urban areas with more economic activities, exposing them to exploitation by
employers due to insufficient rights and protection for this population.
Gaps in Empirical Research on Human Trafficking Prevention
So far, this chapter has reviewed critiques of human trafficking prevention campaigns
from the feminist and migration perspectives. This section shifts the focus to the discovery
paradigm of communication research which employs positivist approach in understanding human
trafficking prevention through ‘objective’ observations and systematic methods of data collection
(Merrigan & Huston, 2009). The nature of this positivist approach allows us to design a study
with samples that are representative of the larger target population, develop measurement of
variables (e.g., knowledge, attitudes, and behaviors related to human trafficking prevention) that
can be recorded objectively and reliably through data collection methods (e.g., survey, focus
groups, content analysis), and analyze the data to describe scope and scale of the variables, and
test their relationships and differences with one another.
There is a dearth of empirical research on human trafficking prevention campaigns,
particularly those that generate data to help us understand how these efforts reduce the problem
of human trafficking (Downman, 2014; Tyldum et al., 2005). Most reports on prevention and
10
awareness activities often contain only a description of program activities (UNODC, 2008a;
UNODC, 2008b) instead of reporting on the actual research design, monitoring and evaluation
process, and quantifiable and generalizable outcomes (e.g., level of knowledge, change in
attitudes and prevention behaviors).
There are apparent research gaps when it comes to evaluating the effectiveness of human
trafficking prevention campaigns. In a global review of evaluation in anti-trafficking initiatives
by Hames, Dewar and Napier-Moore (2010), the authors note that:
…evaluation has been little more than an afterthought and at best conceived as self-edited
reporting on project outcomes by governmental and non-governmental actors alike. This
is not enough. What is needed, is independent external objective evaluation; evaluation
that is based on professional methodology and standards, informed by trafficking
expertise. Evaluation is the single most critical addition necessary to strengthen anti-
trafficking work; resources for evaluation must be an integral part of all anti-trafficking
projects. (p. 3)
The lack of evaluation research is troubling as we cannot objectively identify what works, which
activities should be amplified or eliminated based on their effectiveness, and how millions of
dollars in planning and implementing anti-trafficking programs should be further allocated to
ensure desirable impacts.
Nevertheless, there are a few studies that evaluate human trafficking prevention
programs. Van de Laan and his colleagues (2011) found two studies from a systematic review of
144 evaluations of cross-border sex trafficking prevention programs in rural Nepal and Israel that
met a number of basic criteria for scientific research practices (Centre for Research on
Environment Health and Population Activities, 2004; Hashash, 2007). The two studies showed
11
that the programs resulted in an increase in knowledge and awareness of human trafficking and
of the hotline number.
Magenta (2007) published an evaluation report of an educational campaign on human
trafficking in Moldova that promoted the use of hotline through television advertising. This study
used a structured questionnaire with 13 questions and data were collected with a pretest/posttest
cross-sectional design from 400 participants in four different regions of Moldova. Results did not
show a significant increase in knowledge and awareness from the educational campaign. Fewer
than 3% of the participants stated that they had used the trafficking hotline. However, the survey
had many methodological challenges, primarily a weak design in which most variables were
measured at the nominal data level, limiting the ability to determine reliability and higher level
statistical analyses.
Hames, Dewar and Napier-Moore (2010) identified research by MTV EXIT, which
commissioned Rapid Asia, a research consulting firm, to conduct independent impact
evaluations of MTV EXIT’s campaign activities in select markets in Asia. The evaluation used
an index of 15 questions measuring knowledge, attitudes, and practices (a KAP model) for
trafficking prevention, with measures on knowledge of the definitions of trafficking,
discriminatory or empathic attitudes toward trafficked persons, and behaviors such as reporting
abuse or seeking more information about an overseas job before accepting it. Data were collected
in a pretest, posttest I (immediately after campaign exposure), and posttest II (follow-up
interview one month after exposure) design with a control group. Participants in the treatment
groups were shown one of the two documentaries (Sold or Traffic). Results indicated moderate
increase in KAP index scores of 15 to 19 points (Thainiyom, 2011). MTV EXIT was the only
12
organization in the global review to provide their research methodology but their reports were
not publicly available for peer review or replication.
Although MTV EXIT’s evaluation research was more rigorous methodologically than
earlier trafficking studies, there is still room for improvement, particularly when the research is
compared to other applied fields such as public health communication. On closer examination of
the MTV EXIT evaluation reports, there were inconsistent sampling methods and data collection
points that might have decreased validity of the findings (i.e., using cross sectional samples with
random sampling for control groups, in contrast to convenience panel samples for posttest I and
posttest II exposure groups). It also appeared that the KAP question items had not been tested for
reliability and validity, as some attitudinal items were incorrectly worded as measures of social
norms, which increases third person effects—a situation in which respondents inflate the
intensity of their answer on the assumption that other people have more polarizing attitudes than
they themselves have (Davidson, 1983). There is also a weak correlation between attitudes and
behavior items (i.e., the attitude items measured general attitudes toward the victims rather than
attitudes toward the desired behaviors like reporting suspicious cases or seeking safe migration
advice). In addition, important contextual factors, such as respondents’ media and
communication behaviors and environmental constraints were not measured, which limits the
opportunity to identify risk and protective factors at interpersonal and community levels. The
reports also did not discuss how the results could be used in the design and planning of
subsequent campaigns.
Based on the review of previous research efforts, communication scholars should address
the lack of empirical studies on evaluation of human trafficking prevention campaigns by
13
designing a study with theory-driven framework, rigorous methodology, and translational
research practice.
Theory-driven framework. While organizations like MTV EXIT has utilized the
Knowledge Attitude and Practice (KAP) Model to guide their evaluation research, the KAP
model has been criticized for its limited usefulness in informing practitioners about program
planning (Cleland, 1973; Nichter, 1993; Pelto & Pelto, 1997; Yoder, 1997). The main critique is
that the model focuses largely on individual variables, ignoring other important factors that may
be salient in preventing the adoption of desired behaviors at the interpersonal and community
levels. Adding measurements on interpersonal variables such as interpersonal discussion, and
community factors like communication hot spots and comfort zones to discuss human trafficking
could help us in understanding what facilitates or inhibits the risks of human trafficking, and why
some counter-trafficking messages are unsuccessful in shifting the desired attitudes and
behavior.
Rigorous methodology: Most of the previous human trafficking campaign research
designs did not meet the rigorous scientific standard of including at least a pretest/posttest design
with control groups, using random sampling to increase the representativeness of the findings,
and employing panel samples in longitudinal studies to increase the validity of the results. The
questionnaires often had weak survey construction without scale reliability and validity, and they
often reported only descriptive statistics. Hence, researchers could not understand or investigate
the complex underlying mechanisms that might hinder the adoption of the desired trafficking
prevention behaviors.
Translational research practice: There has been very little collaboration between
academic institutions and anti-trafficking organizations to conduct research to explore public
14
knowledge, attitudes, and behavior on human trafficking prevention that would inform
practitioners and policy makers in the design, implementation, and evaluation of the
effectiveness of their programs. More collaborations between the academic and the anti-
trafficking sector are needed to build empirical evidence of knowledge, attitudes, and
experiences of target populations who are at risk of being trafficked for labor exploitation.
Research Objectives and Organization of the Dissertation
In order to address the aforementioned research gaps and advance the field of
communication and human trafficking prevention, this dissertation was formed in collaboration
with MTV EXIT, International Organization for Migration (IOM), and local anti-
trafficking/migrant organizations in Indonesia. The research objectives of this project are to…
• generate empirical data on the prevalence and experiences of human trafficking
victims in Indonesia,
• identify determinants of human trafficking prevention, and evaluate the
effectiveness of a human trafficking prevention campaign through a Multilevel
Communication Model,
• explore the role of narratives as persuasion strategies to influence knowledge,
attitudes and behaviors related to human trafficking prevention,
• and inform practitioners and policy makers of the best practices in designing and
implementing effective human trafficking prevention programs
This dissertation utilizes the public health model of a scientific approach to prevention to
organize its chapters. This public health model consists of four steps, demonstrated in Figure 1.2,
as (1) defining the problem, (2) identifying risk and protective factors, (3) developing and testing
prevention strategies, and (4) assuring widespread adoption by scaling up effective programs
15
(Mercy, et al., 1993). Public health scholars and practitioners have called for human trafficking
research agenda to follow this model (Alpert & Chin, 2017; Rothman et al., 2017).
Figure 1.2 The public health model of a scientific approach to prevention
(Mercy, et al., 1993)
This dissertation is organized as followed:
Chapter Two addresses the first step of human trafficking prevention: defining the
problem by estimating the prevalence and describing the experiences of human trafficking
victims in Indramayu, Indonesia. This chapter describes the problems in defining human
trafficking, and the methodological challenges in estimating the prevalence of human trafficking.
A mixed-method study through survey and focus groups research was conducted to estimate the
number of human trafficking victims in Indramayu, Indonesia as well as describing different
conditions of trafficking, responses to seek help, escape outcomes, and health problems among
the victims.
Chapter Three focuses on the second and third steps of the public health model:
identifying risk and protective factors, and designing and testing prevention strategies. This
chapter introduces a Multilevel Communication Model as a theory-driven conceptual framework
to identify risk and protective factors of human trafficking prevention, and test the effects of a
16
human trafficking prevention campaign. Survey research with longitudinal randomized control
trial and a panel design (N=527) was conducted in Indramayu, Indonesia. Eight human
trafficking prevention outcomes (e.g., knowledge, attitudes, perceived norms, intention, and
behavior) were measured at the baseline, one-week and four-month follow-ups. The study also
measured 13 background influences (e.g., sex, age, level of education, household income, and
prior experience with human trafficking) and communication ecological factors (integrated
connection to storytelling network, communication hotspots, interpersonal discussion about
human trafficking, migrant network size, and media exposure to trafficking information) as
potential determinants of human trafficking prevention outcomes.
Chapter Four further explores the third step of the public health model in designing and
testing prevention strategies by investigating the role of transportation and identification of
characters in a documentary film in persuading target audience to change their knowledge,
attitudes, norms, intention, and behavior related to human trafficking prevention. 527 male and
female participants from Indonesia took part in the survey before and after viewing a
documentary that aimed to promote human trafficking prevention. Regression analyses
demonstrated that transportation increased favorable attitudes and intention to prevent trafficking
as well as decreased norms to migrate. Meanwhile, identification with characters in the
documentary had different effects on male and female participants. Specifically, women who
identified more with the female character increased norms to migrate, and had less knowledge,
attitudes, and preventive behavior. Men who identified more with the male character had less
knowledge about trafficking. This indicated that these participants might counter-argue and
develop reactance to the recommended messages in the documentary based on their migration
situation and prior experience with the prevention services.
17
Chapter Five concludes this dissertation by discussing how a multilevel communication
approach could be used to guide researchers and practitioners in designing and testing human
trafficking prevention strategies. The chapter also offers policy recommendations to assure
widespread adoption and scaling of effective human trafficking prevention campaign practices.
18
CHAPTER 2: PREVALENCE AND EXPERIENCES OF HUMAN TRAFFICKING
VICTIMS IN INDRAMAYU, INDONESIA
Introduction
This chapter focuses on the first step in addressing human trafficking prevention:
defining what it is and describing the scale of the problem through prevalence research (Alpert &
Chin, 2017).
Determining the prevalence of human trafficking has become one of the most important
research agendas within the anti-trafficking community around the world. After all, accurate
estimates of the trafficked population are essential for government agencies and related
organizations to create effective policies, and allocate appropriate resources to provide victim
assistance, increase investigation and prosecution efforts, and develop initiatives to prevent
further incidents of trafficking (Picarelli, 2015; Zhang, 2012a). Nevertheless, researchers have
noted the challenges in estimating the prevalence data. Journal of Human Trafficking and Anti-
Trafficking Review, two prominent journals in the field, recently dedicated a special issue on the
problems and contentious debates of gathering and using prevalence data to combat human
trafficking (Raphael, 2017; Yea, 2017). Scholars agree that there have been inconsistent global
estimates of human trafficking prevalence over the last 12 years; from International Labour
Organization (2005)’s 2.45 million to Kevin Bales (2012)’s 27 million, and the 2016 Global
Slavery Index which claimed 45.8 million people were victims of modern slavery and human
trafficking (Walk Free Foundation, 2016). The wide range of global prevalence estimates over
the years can be attributed to the two things: (1) the problems of defining human trafficking; and
(2) inconsistent methodologies that were utilized to count the number of victims.
19
Challenges in Defining Human Trafficking
Scholars have commented on the challenges in estimating the number of human
trafficking victims due to the lack of consensus on a legal definition at the local, regional, and
country levels, mistaken inclusion of prostitution and smuggling as human trafficking, and
interchangeable use of related terms such as ‘trafficking in persons’, ‘forced labor’, ‘modern-day
slavery’, and ‘modern slavery’ across different prevalence studies (Fedina & DeForge, 2017).
For example, International Labour Organization (ILO), one of the most reputable organizations
that publishes global prevalence data, has conflated the terms ‘forced labor’, ‘human trafficking’,
and ‘modern slavery’ in its reports over the last 12 years. In its groundbreaking report that
offered the first global estimates of human trafficking, ILO (2005) defined human trafficking as
a distinct, albeit related, concept from forced labor. Forced labor is defined under the 1930 ILO
Forced Labor Convention (No. 29) as “all work or service which is exacted from any person
under the menace of any penalty and for which the said person has not offered himself
voluntarily”. On the other hand, human trafficking, under the article 3(a) of the UN Trafficking
Protocol of 2000, is referred as:
…the recruitment, transport, transfer, harbouring or receipt of a person by such means as
threat or use of force or other forms of coercion, of abduction, of fraud or deception for
the purpose of exploitation. Exploitation includes, as a minimum, the exploitation of the
prostitution of others or other forms of sexual exploitation, forced labour or services,
slavery or practices similar to slavery, servitude or the removal of organs. (pp. 7)
Out of the global estimates of 12.3 million victims of forced labor, 2.45 million were counted as
victims of human trafficking (ILO, 2005).
20
In the second global estimates report, ILO (2012a) did not identify human trafficking as a
different concept from forced labor and estimated there were 20.9 million victims of forced labor
and/or human trafficking worldwide. For the most recent prevalence report, ILO (2017a)
collaborated with the Walk Free Foundation, and International Organization for Migration
(IOM), and opted to use ‘modern slavery’, defined as “situations of exploitation that a person
cannot refuse or leave because of threats, violence, coercion, deception, and/or abuse of power”
(pp. 16) as a new umbrella term for related concepts such as human trafficking, forced labor,
forced marriage, slavery and slavery-like practices. The report estimated 40.3 million victims of
modern slavery: 24.9 million of whom were in forced labor and 15.4 million in forced marriage.
The inclusion of forced marriage, defined as someone who did not give consent and was forced
into a marriage, increased the global estimates of modern slavery drastically. These examples
from ILO reports demonstrated the evolving concept of human trafficking and conflation with
other terms, resulting in inconsistent prevalence numbers from 2.45 million in 2005, to 20.9
million in 2012, and to 40.3 million in 2017.
The UN Protocol (2000) defines human trafficking as a complex process that includes (1)
the act of moving victims from one location to the place of employment; (2) the means to control
the victims through force, coercion, deception or fraud; (3) for the purpose of sexual exploitation
or forced labor. (4) Children under the age of 18 who have been exploited in prostitution or
forced labor are also considered to be victims of human trafficking regardless of the presence or
absence of victim movement, and the use of force, coercion, fraud, or deception. Despite the UN
definition, previous empirical research studies did not point to common conceptualization and
operationalization of human trafficking that met all of these four conditions. Researchers have
selectively picked different ways to measure the number of trafficked victims. In a study to
21
estimate the number of human trafficking victims in Nepal, respondents were asked in two
interview questions if they thought they have ever been trafficked, and if they have any friends
and/or family who have been trafficked (Archer, Boittin & Mo, 2016). The authors admitted that
the two items were not valid measures as many respondents did not even know what human
trafficking was. In another study to determine the number of sex trafficking victims in
Cambodia, Steinfatt and Baker (2011) defined victims as people under the age of 18 who were
working in prostitution and/or those who could not leave the venue. Pennington and colleagues
(2009) chose to define human trafficking victims in five Eastern European countries with three
survey items as people who have travelled abroad for work and, upon arrival, were locked up and
forced to work for little or no pay in the industries such as domestic/nursing,
enterprise/agriculture/construction, or prostitution. While these three items covered more
elements in the UN definition, they glossed over concepts such as the means to control victims
through coercion, deception, and fraud, as well as trafficked children under the age of 18.
Among recent empirical studies, Zhang (2012b) offered conceptual definition by
describing trafficking violation as “any infringement (direct or threatened) of physical integrity
or freedom of movement/communication” (pp. 153). He went on to operationalize the concept
more comprehensively by developing 21 survey items to measure trafficking violation such as
confiscating identification documents, being held hostage by traffickers and demanded ransom
from family members, physical abuse, threats of physical abuse, and restriction of freedom and
communication with loved ones. In addition, he created another 18 survey items to measure a
related concept, abusive labor practices, that include deception and lies, as well as unfair and
exploitative labor practices as sub-category variables. It is unclear why Zhang (2012b) separated
22
abusive labor practices concept from trafficking violation since items under abusive labor
practices still fall under the UN human trafficking definition.
These empirical studies demonstrated the complexities of operationalizing human
trafficking, and no consensus has been reached on how to measure the concept. For practical
reason, Zhang (2012a) argued that the simpler ILO definition of forced labor which constitutes
the extraction of labor involuntarily through the menace of penalty could be used to measure
human trafficking. This recommendation was adopted in the most recent global estimates by ILO
(2017b) in which the involuntary nature of labor was operationalized by six survey items that
asked if the respondents and/or someone in their immediate family network were (1) “forced to
work by an employer or recruiter,” (2) “forced to work to repay a debt with an employer or
recruiter and not allowed to leave,” (3) “offered one kind of work, but forced to do something
else and not allowed to leave,” (4) “forced to work for a master as a slave,” (5) “had to work in
order to help another family member who was forced to work by an employer,” and (6) “forced
to work for an employer so that another person would receive a job, land, money, or other
resources.” Participants who answered yes to any of these six items were considered to be a
victim of human trafficking, and would move on to answer 14 follow-up items created to
measure menace of penalty that included experiences of physical and sexual violence, threats of
violence, threats against family, threats of legal action, locked in workplace or living quarters,
punished through depravation of food and sleep, punished through fines, forced to consume
alcohol or drugs, withheld passport or identification documents, withheld wages, forced to repay
debt, and restricted communication with loved ones.
23
Methodological Challenges in Estimating Human Trafficking Victims
Despite efforts to determine the prevalence of human trafficking with better precision
over the years, researchers like Weitzer (2010, 2012, 2014a, 2014b) and Gallagher (2014, 2017)
have criticized the methodologies that underlie existing prevalence studies as spurious and
unreliable. The earlier human trafficking prevalence studies often rely on secondary data sources
such as reported cases from law enforcement, social service organizations, and media reports, as
well as expert opinions to calculate the estimates. This is problematic since the number of
victims in a particular location may be non-systematically recorded by authorities and could not
be reliably extrapolated to represent a wider region (Tyldum & Brunovskis, 2005). Moreover,
official records may represent only a fraction of the victim population since majority of them
have not been identified or rescued (Pennington, et al., 2009).
Challenges in counting human trafficking victims remain, especially for those currently
living in exploitative conditions and are hard to reach due to the hidden nature of their situation.
Many trafficked victims do not possess legal immigration status, and work in isolated
environments (e.g., rural farms, fishing boats offshore, and sex workers in secret establishments).
To count these hidden populations, non-probability sampling methods such as respondent-driven
sampling may be the most appropriate approach (Fedina & DeForge, 2017). Respondent-driven
sampling provides participants with incentives to answer the survey, as well as to recruit other
people in their network to join the study. Nevertheless, respondent-driven sampling has
limitations and is not an appropriate method to produce national estimates. It should be used to
provide only local or regional estimates from an area where participants are recruited (Fedina &
DeForge, 2017).
24
Recent prevalence studies rely more use of random sampling household surveys from the
‘source’ country to obtain macrolevel data that is representative of the general population
(David, 2017). This is a welcome shift since random sampling surveys use probability methods
to select samples and allow us to make more accurate inferences from a population of interest
(Gordis, 2009). Random sampling surveys have high external validity and the results can be
generalized with higher certainty about the population. However, there is also a tendency for
researchers to extrapolate random sampling survey data from a few sites and generalize findings
to cover a much wider region. Gallagher (2017) criticized that the global estimates of 45.8
million victims from the 2016 Global Slavery Index Report were derived mostly from random
sampling surveys by Gallup Poll in just 25 countries (Walk Free Foundation, 2016). The
prevalence data in these 25 countries were then extrapolated for the remaining non-surveyed
countries based on secondary sources (e.g., news reports) and other factors such as vulnerability
to human trafficking, geography, and country context. The 2016 Global Slavery Index report
outlined this extrapolation methodology, by first dividing 167 countries into 12 clusters based on
similar risk profiles. The average prevalence data of the surveyed countries in each cluster was
then applied to calculate other non-surveyed countries in the same group, and additional
refinements were made to 40 countries based on factors such as level of conflict, state-imposed
forced labor, geopolitical issues and remoteness of the location (small island states). Critics have
pointed out the lack of transparency and logic in how different countries were grouped into a
cluster. Gallagher (2014) concluded:
At some points, application of the extrapolation “protocol” verges on the
ludicrous…After noting that almost no reliable information exists on slavery in China,
the Index’s authors cheerfully proceed to declare that they are comfortable with China
25
being considered pretty much the same as other East Asian nations like South Korea,
Taiwan and Japan. (p. 3)
It should be noted that the staggering 45.8 million figures were calculated mostly from
the self-reported survey data of 470 out of 29,000 respondents who answered that they or their
family member had experienced forced labor or forced marriage. Therefore, global estimates
should be taken with extreme caution due to the limited availability of data sources, and the
flawed methodologies that were used to derive the number.
Nevertheless, continual efforts are being made by leading anti-trafficking organizations
to improve the validity and reliability of estimating global prevalence data. In the most recent
global estimates of 40.3 million victims, ILO (2017a) collaborated with the Walk Free
Foundation and IOM to derive this number from nationally representative, random sampling
survey data of 71,758 individuals in 48 countries. This is an improvement from the sample size
of 29,000 respondents in 25 countries in the 2016 Global Slavery Index (Walk Free Foundation,
2016). While random sampling household surveys are considered to be one of the better ways in
estimating prevalence, it may not be an appropriate method to calculate victims who are
currently exploited and hidden in establishments in ‘destination’ cities and countries. Random
sampling surveys may be more suitable to gather data of former trafficked victims who are now
living safely in their homes and no longer experience exploitative labor conditions.
Consideration should be made to determine the locations where majority of former trafficked
victims are living during research design. For example, it could be a ‘source’ area with outgoing
migration, and former victims have returned to live in their homes.
26
Microlevel vs Macrolevel Prevalence Data
Most of the challenges discussed thus far concern the estimation of human trafficking
prevalence at the macrolevel nationally and globally. Given the pressure from international
policy makers who demand macrolevel prevalence data to allocate resources to combat human
trafficking, some scholars are questioning the usefulness of prioritizing national and global
prevalence data over microlevel data from a town, a city, or a small region of a country
(Feingold, 2017; Dottridge, 2017; Wietzer, 2014b). It would be ill-advised to create anti-
trafficking policies based solely on the macrolevel data due to its inaccuracies. For example, the
national prevalence data of 736,100 people in Indonesia was calculated based on the proportion
of 7 out of 1,000 survey respondents who answered yes to having been a victim of forced labor
or forced marriage (Walk Free Foundation, 2016). After weighing the survey data, the
prevalence rate was inferred to be 0.286% of the national population. The raw survey data from 7
self-identified victims make it impossible to inform policy makers with meaningful and reliable
analysis such as pin-pointing which geographic locations in the country have higher incidence of
human trafficking or identifying specific industries where labor exploitation is more pervasive
than others.
For this reason, microlevel prevalence research that utilizes quantitative and qualitative
methods should still be valued since it can offer more advantages than macrolevel studies. First,
human trafficking estimates are more reliable because of the smaller study parameters and better
measurable contexts. Second, complexities and insights of victim experiences can be captured
and described to reflect unmet needs and opportunities for more effective responses. Lastly,
more sensitive data collection instruments could be developed with cultural considerations and
tailored to the local population to detect patterns of trafficking hotspots and to explore associated
27
risk factors that will be helpful in the planning and implementation of anti-trafficking
interventions (Weitzer, 2014b; Robinson, Branchini & Thame, 2017).
Study Site and Research Questions
In order to address the current challenges of human trafficking prevalence research, this
chapter focuses on a microlevel study that aims to estimate the number of human trafficking
victims in Indramayu, Indonesia, and describe their experiences during and after exploitation.
This study is part of a larger research project formed in partnership with US Agency for
International Development, MTV EXIT Foundation, IOM, the London School of Public
Relations–Jakarta, and local migrant organizations to evaluate the effectiveness of an anti-
trafficking campaign, and to generate data on knowledge, attitudes, and prevention behaviors in
Indonesia.
Figure 2.1 Indramayu regency map
Figure 2.1 displays the location of Indramayu, a regency in West Java province, with a
population of approximately 1.71 million people (Central Statistics Agency of Indramayu
Regency, 2016). Indramayu was selected as the research site based on findings from the IOM
28
Counter Trafficking Unit (2014) which indicated that West Java was the province with the
highest number of human trafficking victims. Over 26% of all trafficked victims who received
psychosocial services by IOM in Indonesia came from West Java. Within West Java, Indramayu
is the regency with the largest number of victims (N=345), over 70% more than Cirebon
(N=199), the regency with the next highest number of victims.
Residents in Indramayu experienced one of the highest rates of poverty and
unemployment in the province (Pusdalisbang Office, 2014). Because of the lack of job
opportunities, many young men and women migrate to other parts of Indonesia such as Jakarta,
Riau Islands, North Sumatra, East Java, and South Sulawesi. Male domestic migrant workers end
up working in the industries with hazardous conditions such as logging, fishing, manufacturing,
and construction. Women in Indramayu got married at a very young age: over 70% of them
entered into a marriage before the age of 19 (Pusdalisbang Office, 2014). Other sociocultural
factors such as premature marriage that leads to early divorces, tolerant attitude toward divorce
as a way out of an unsatisfactory marriage, increasing number of divorced mothers that have to
financially provide for their children, expectation for women to migrate, seek work and remit
money back home, and general live and let live attitude have been cited as reasons that
encourage women from Indramayu to migrate and seek work in bars, restaurants and related
entertainment venues and eventually become sex workers (Lim, 1998).
For international migration, over 74% of the outgoing migrants from Indramayu are
women (Listinai, 2017). These female migrants travel to seek work as domestic helpers in
countries such as Saudi Arabia, Taiwan, Malaysia, Singapore and Hong Kong (Rizkytha &
Abraham, 2015). They are one of the most vulnerable groups to experience human trafficking as
they are prone to exploitative conditions such as bonded labor, unpaid salary and insurance
29
claim, physical abuse, sexual harassment, mistreatment and oppression, prison sentence and
disappearance, HIV infection, social isolation, and even death (Emmers, 2003; Kistyarini, 2013;
Shehadeh & McCoy, 2014; Utami, 2012). Despite these risks, Indonesian women continue to
search for work abroad. Their motivation to migrate can be explained by a theory of migration
by Lee (1966) which explains that the decision to migrate is a consequence of the evaluation of
differences in benefits between staying in the place of origin and moving to the place of
destination. Conditions that are found to make staying in the place of origin less attractive
become the “push factors”, and cause stress and intense pressure to migrate. The destination
abroad becomes the “pull factor” because it is considered to be a promising land with more
opportunities. Therefore, Indonesian women are willing to take risks to migrate if they feel the
benefits outweigh staying where they are (Weber et al., 2002). A qualitative study by Rizkytha
and Abraham (2015) found that Indonesian migrant women perceived working as a domestic
helper abroad was a prestigious career because they had to compete and meet extensive
requirements (e.g., learning related skills and acquiring language proficiencies) for the jobs in
other countries. In addition, successful stories from returned migrants who improved their
economic status by flaunting their wealth (e.g., renovating their houses, buying new cars, and
wearing expensive jewelry) created a social norm where women were expected to make a
sacrifice and leave their family behind to migrate abroad to work and attain a similar status
(Mashud, 2010; Silvey, 2007; Supriyoko, 1990).
Given the high level of vulnerability of migrants from Indramayu, this study will attempt
to answer the following research questions:
RQ1: What is the prevalence of human trafficking victims in Indramayu, Indonesia?
30
RQ2: What are their experiences of (a) the human trafficking conditions, (b) finding
ways to escape or getting rescued, (c) receiving protection services and seeking
justice, and (d) health problems associated with trafficking?
Given the disproportionate rate of migration that gears toward women, this study will
also explore:
RQ3: Do men and women significantly vary on the experiences of human trafficking
conditions, escape or rescue actions, protection and justice, and health problems?
Methods
This study employed a mixed-method design with explanatory sequential by using
quantitative surveys to capture human trafficking prevalence and count frequencies of various
exploitative conditions and experiences between men and women. Focus groups were conducted
as a qualitative follow-up study to help interpret the survey results (Creswell & Plano Clark,
2011). This mixed-method design is appropriate to explain complex social phenomenon like
human trafficking, which may not be fully analyzed and contextualized by only quantitative or
qualitative data. The quantitative survey is a principal method while focus groups data is used as
a supportive element.
Quantitative Survey Study
To obtain representative data from the population in Indramayu, the survey was
implemented using a multi-stage sampling method moving from the kabupaten (regency) level,
to kecamatan (districts), then Rukun Tetangga (neighborhoods), and finally households. Out of
31 districts in Indramayu, 18 districts with the largest populations were selected for the study.
Six districts (Bongas, Kandanghaur, Krangkeng, Lelea, Losarang, and Sukagumiwang) were
then chosen at random to be included in the survey sites. Fifteen researchers visited 90
31
neighborhoods in these six districts (number of neighborhoods in each district were allocated
proportionally to the population size of each district) and conducted face-to-face interviews with
5-6 households in each neighborhood. To make sure there was no systematic selection bias
among respondents, a Kish Grid was used for each household in which more than one person
was eligible to participate. To be eligible to participate, respondents had to be 18-39 years old,
live in Indramayu, had completed at least SD education (elementary school), be able to read and
write in Bahasa Indonesian. Respondent age was limited from 18 to 39 years old based on the
recommendation by our local partners that this population was most at risk of being trafficked as
they were likely to migrate. In addition, this project was part of a larger study to evaluate the
effects of an anti-trafficking campaign by MTV Networks with youths as the target audience.
Therefore, adults older than 39 years old were not included in this study. Institutional review
board at the investigators’ university approved the recruitment procedures, the survey and focus
group instruments.
Figure 2.2 Multi-stage sampling procedure
32
Overall, 577 individuals were recruited but 50 declined to participate, giving us a final
sample of 527 people. This sample size is well above 384, the number needed to estimate the
prevalence at 95% confidence interval. Local research firm BOI Research Services was
contracted to administer the survey in Indonesia. Data collection took place for about four weeks
in April 2014. Figure 2.2 describes the multi-stage sampling procedure for this study.
Qualitative Study
Focus group, a carefully planned, interactive interview among a group of people
(normally 4-8) who participate in a guided discussion of a topic of interest, is one of the most
widely used qualitative tools in the field of social sciences (Stewart & Shamdasani, 2014). Focus
group generates shared, magnified discussion on a topic that triggers immediate evidence from a
variety of similar or contrasting opinions and experiences (Morgan, 1997). Direct evidence from
the interactive group discussion can provide more accurate, direct conclusions on the
participants’ lived experiences.
Focus group discussions (FGDs) were held with 2 groups of women and 2 groups of men
(a total of 4 groups with 8 participants in each FGD) in Indramayu in May 2014. Recruitment
criteria included participants who were 18-39 years old, literate with at least grade 6 education
(able to read and write in Bahasa Indonesian), had migrated for work or knew a family member
who had migrated. FGD moderator asked the participants for their understanding of what human
trafficking was, who was at risk, what motivated and pushed them to migrate, and their
experiences of human trafficking. The results of these four FGDs were used to provide context
and complement the understanding of human trafficking prevalence and exploitative conditions,
as well as different experiences between men and women.
33
Survey Development
The survey instrument went through a rigorous development process. Preliminary
interviews were conducted with 17 anti-trafficking experts in Indonesia to review existing tools
and discuss any ethical concerns that might put research participants at risk. A stakeholder
workshop was subsequently organized in Jakarta in November 2013 with 24 participants from
local anti-trafficking organizations, government agencies, and universities. The workshop was
partly designed to elicit survey items to measure human trafficking prevalence and related
concepts. After the workshop, the first draft of the survey was created and pretested with 18 local
participants to verify accurate interpretation of the items, appropriate use of scale measurements,
and identify problems such as question order and length. The survey was revised and piloted
with 100 participants in Indramayu to test the validity and reliability of its scales. The survey
took 30-45 minutes to complete. Revisions were made, and the third version of the survey was
circulated to our partners for another review. The survey was finalized and translated into Bahasa
Indonesian for data collection in April 2014.
Measures
Demographic and socioeconomic variables. Sex, age, level of education, annual
household income, living under national poverty line, ethnic language spoken, marital status,
length of residence in Indramayu, household size, number of dependents, employment status, and
religion were collected.
Experience with migration. Participants were asked if they have ever moved away from
home for work. Yes was coded as 1 and No was coded as 0.
Human trafficking. This study followed Zhang’s (2012a) recommendation to narrowly
interpret human trafficking under the ILO definition of forced labor which consists of two
34
dimensions: someone who has experienced involuntary labor and/or were under the menace of
penalty. Figure 2.3 summarizes the conceptualization and operationalization of human
trafficking and the 16 survey items under the concept. To be identified as a victim of human
trafficking, participants must have migrated for work, and responded Yes (coded as 1) to any of
the 16 items adapted partly from the trafficking violation and abusive practice measures by
Zhang (2012b). Involuntary labor is operationalized by 2 items “Do you think you were a victim
of human trafficking or exploitation?” and “I was working with reduced pay or without pay to
repay my debt to the employer and/or recruitment agency”. Menace of penalty consists of 3 sub-
categorizes: threats and physical harm against the victim while traveling to the workplace or at
the workplace (operationalized by 4 items such as “I was physically abused at my workplace”),
restriction and depravation (measured by 6 items that include restriction of movement and
communication such as “I was not allowed to communicate freely with my family and friends”
and depravation of living conditions, e.g., “I was denied medical treatment when I was sick or
had an injury”), and exploitative labor practices (including 4 items, e.g., “I was forced to work
excessively long hours”).
Respondents who self-identified as a victim of human trafficking moved on to answer
additional questions related to their experience in trying to escape their situation, seek protection
and restorative justice, and if they had any health problems due to being trafficked.
Escape or rescue. moved on to answer Yes (coded as 1) or No (coded as 0) to 13
statements that described what they did to escape the exploitative situation (e.g., I called the anti-
human trafficking/NGO hotline for help,” “I went to the police/local authority to file a report,”
and “I was rescued from the workplace by the police, NGO, and/or embassy employees.”
35
Figure 2.3 Conceptualization and operationalization of human trafficking
Protection and justice. Participants answered yes (1) or no (0) to 7 additional items under
“Did you do any of the following after you were rescued and/or left your employment?”. These
items were concerned with protection and justice for trafficked victims. Examples of the items
are “I cooperated with the police to testify against the employer and/or sponsor in a criminal
lawsuit,” “I filed a worker insurance claim to receive my compensation/lost wages,” “I was
placed in a rehabilitation shelter,” and “I had a medical check-up and treatment.”
Health problems. Participants answered if they have experienced any of the following 9
health problems after being trafficked (e.g., physical injury, chronic pain, depression, anxiety,
and sexually transmitted diseases).
Data Analysis
The statistical analysis for this study included computing descriptive statistics to obtain a
better understanding of the characteristics of the samples in Indramayu, including their
socioeconomic and demographic variables, and experiences with migration and human
36
trafficking. Crosstab analysis or independent t-tests between men and women were subsequently
run to explore the significant differences for each variable. For FGDs, conversations were audio
recorded, transcribed, and translated from Javanese/Bahasa Indonesian to English. Discussion
around experiences of human trafficking with appropriate quotes are used to explain survey
results in the discussion section.
Results
Characteristics of the Participants
Table 2.1 displays descriptive statistics of the participant characteristics by sex. 47% of
the participants were men, and 34% were 18-24 years old, followed by 23% who were 25-29
years old, 19% age 30-34, and 25% from the age range of 35-39. For their level of education,
40% reported that they had completed SD (elementary school), 30% completed SMP (Grade 9),
and 27% graduated from SMA (Grade 12). Almost 80% of the participants had an annual
household income less than 30 million Rupiah (2,595 USD). About half were living below
national poverty level at 312 USD per person per year (Priasto, 2015). Based on the recruitment
criteria, all of our participants could speak Bahasa Indonesian. In addition, majority spoke
Javanese (80%), followed by Sundanese (11%) and Cirebonan (10%). About 65% were married
and 32% were single. Over 44% were employed (either full-time, part-time or self-employed)
while 43% were experiencing unemployment. All but one individual practiced Islam. About 28%
of the participants had migrated out of Indramayu for employment. On average, the participants
had lived in Indramayu for 26 years (SD = 8.1), had 5.2 family members (SD = 1.6) living in the
same household, and had 1.2 dependents (SD = 1.3) to look after.
Crosstab analysis or independent t-test was run for each variable. Findings suggest there
were significant differences between men and women on the level of education, language
37
Table 2.1 Characteristics of the participants by sex
Categorical Variables
Male
n (%)
Female
n (%)
Total
n (%)
p
Sex 250 (47.4) 277 (52.6) 527 (100.0) -
Age in years
18-19 29 (11.6) 25 (9.0) 54 (10.2) 0.330
20-24 67 (26.8) 56 (20.2) 123 (23.3) 0.074
25-29 54 (21.6) 67 (24.2) 121 (23.0) 0.481
30-34 43 (17.2) 55 (19.9) 98 (18.6) 0.434
35-39 57 (22.8) 74 (26.7) 131 (24.9) 0.299
Education
Graduated from SD (Grade 6)** 85 (34.0) 127 (45.8) 212 (40.2) 0.006
Graduated from SMP (Grade 9) 69 (27.6) 88 (31.8) 157 (29.8) 0.296
Graduated from SMA (Grade 12)** 84 (33.6) 58 (20.9) 142 (26.9) 0.001
Graduated from a University* 12 (4.8) 4 (1.4) 16 (3.0) 0.025
Annual household income ℾ
15 million Rp or less (1,297 USD) 104 (41.6) 107 (38.6) 211 (40.0) 0.487
15 million - 30 million Rp (1,297 - 2,595 USD) 98 (39.2) 111 (40.1) 209 (39.7) 0.838
30 million - 45 million Rp (2,595 - 3,892 USD) 35 (14.0) 49 (17.7) 84 (15.9) 0.248
over 45 million Rp (3,892 USD) 13 (5.2) 10 (3.6) 23 (4.4) 0.372
Living below national poverty line 133 (53.2) 138 (49.8) 271 (51.4) 0.438
Ethic language spoken
Sundanese** 18 (7.2) 40 (14.4) 58 (11.0) 0.008
Javanese* 209 (83.6) 212 (76.5) 421 (79.9) 0.043
Cirebonan 26 (10.4) 25 (9.0) 51 (9.7) 0.594
Other 12 (4.8) 11 (4.0) 23 (4.4) 0.642
Marital status
Married*** 120 (48.0) 221 (79.8) 341 (64.7) 0.000
Single*** 122 (48.8) 45 (16.2) 167 (31.7) 0.000
Divorced/Widowed 8 (3.2) 11 (4.0) 19 (3.6) 0.635
Employment status
Employed*** 162 (64.8) 72 (26.0) 234 (44.4) 0.000
Unemployed*** 47 (18.8) 181 (65.3) 228 (43.3) 0.000
Student** 41 (16.4) 24 (8.7) 65 (12.3) 0.007
Practice Islam 250 (100.0) 276 (99.6) 526 (99.8) -
Returned Migrant 66 (26.4) 83 (30.0) 149 (28.3) 0.364
Continuous Variables M (SD) M (SD) M (SD) p
Years of residence 25.8 (7.8) 26.2 (8.5) 26 (8.1) 0.500
Household size 5.3 (1.6) 5.2 (1.5) 5.2 (1.6) 0.302
No. of dependents 1.3 (1.4) 1.1 (1.1) 1.2 (1.3) 0.157
*p < .05, ** p < .01, ***p < .001 There's a significant difference between male and female in the same
variable.
ℾ Exchange rate at 1 USD = 11,563 Rp in April 2014 from https://www.exchange-
rates.org/Rate/USD/IDR/4-30-2014
38
spoken, marital status, and employment status. Men had more opportunities to complete higher
education as more of them graduated from SMA (34%) and a university (5%) than women (21%
and 1% respectively). More women were married (80% compared to 48%). Women also
experienced unemployment at a much higher rate than men (65% compared to 19%).
Prevalence of Human Trafficking in Indramayu
The results of human trafficking prevalence in Indramayu are displayed in Table 2.2. The
prevalence of human trafficking is 15.4%. It should be noted that this prevalence rate is only
representative of the Indramayu population who were 18 to 39 years old, and not for the entire
population. 98,654 people were estimated to be victims of human trafficking in Indramayu in
2014. The 95% upper and lower bounds of the estimate were 78,897 and 118,411 respectively.
There is a higher prevalence among women (17.3%) than men (13.2%).
Table 2.2 Calculation of the estimates and confidence intervals for the number of trafficked persons in
Indramayu
Total Population in
Indramayu*
(18-39 years old)
Prevalence of
Human
Trafficking (%)
Estimated Number of
Trafficked Person
in Indramayu
95% Confidence Interval
Bounds
Lower Upper
Total 641,861 15.37 98,654 78,897 118,411
Male 329,788 13.20 43,532 29,699 57,365
Female 312,073 17.33 54,078 40,178 67,987
*Calculated from Central Statistics Agency of Indramayu Regency (2016) and a report by Pusdalisbang
Office (2014)
Prevalence of Human Trafficking Conditions
The prevalence of human trafficking conditions was categorized as followed: trafficked
victims, counted if respondents had experienced at least one of the sixteen human trafficking
conditions; involuntary labor, consists of any of the two conditions of self-report as a victim and
forced to work with reduced or no pay to repay debt to the employer; menace of penalty, which
includes any of the 14 violations under threats and physical harm, restriction and depravation,
39
and exploitative labor practices. Table 2.3 displays the distribution count and percent of the
participants who had experienced human trafficking conditions under each category while Table
2.4 further demonstrates the breakdown prevalence of each of the 16 human trafficking
conditions by sex.
Table 2.3 Distribution count and percent of the participants who had experienced human trafficking
conditions
Human
Trafficking
Conditions
n
(%)
Total Number
of Persons
(n=527)
0 1 2 3 4 5 6 7 8
Trafficked victims
446
(84.6)
28
(5.3)
38
(7.2)
3
(0.6)
7
(1.3)
2
(0.4) -
2
(0.4)
1
(0.2)
81
(15.4)
Involuntary labor
498
(94.5)
23
(4.4)
6
(1.1) - - - - - -
29
(5.5)
Menace of penalty
457
(86.7)
24
(4.6)
34
(6.5)
6
(1.1)
3
(0.6)
2
(0.4)
1
(0.2) - -
70
(13.3)
Threats and
physical harm
521
(98.9)
4
(0.8)
2
(0.4) - - - - - -
6
(1.1)
Restriction and
depravation
474
(89.9)
35
(6.6)
15
(2.8)
3
(0.6) - - - - -
53
(10.1)
Exploitative
labor practices
483
(91.7)
34
(6.5)
8
(1.5)
2
(0.4) - - - - -
44
(8.3)
Note: This table shows distribution count and percent of the participants and the number of human
trafficking conditions they had experienced. For example, under trafficked victims, there were 446
individuals (84.6%) in the sample who had not experienced any human trafficking conditions; 28
individuals (5.3%) had experienced one condition of human trafficking, 38 individuals (7.2%) had
experienced two conditions of human trafficking, etc., for a total of 81 individuals (15.4%) who had
experienced at least one condition of human trafficking.
Over 15.4% of the participants were trafficked victims as they had experienced at least
one human trafficking condition. Most trafficked victims experienced two conditions of human
trafficking (7.2%), and 2.9% encountered three or more conditions. 5.5% also experienced at
least one form of involuntary labor: 2.8% self-reported to be a trafficked victim while 3.8%
mentioned they were under debt bondage, forced to work with reduced or no pay to repay money
40
back to their employer or recruitment agent. 13.3% of the participants had encountered at least
one type of menace of penalty, with restriction and depravation being the most common
(10.1%), followed by exploitative labor practices (8.3%), and threats and physical harm (1.1%).
Table 2.4 Experiences of human trafficking by sex in Indramayu
Categorical Variables
Male
n (%)
Female
n (%)
Total
n (%)
p
Sex 250 (47.4) 277 (52.6) 527 (100.0) -
Trafficked victim 33 (13.2) 48 (17.3) 81 (15.4) 0.189
Involuntary labor 17 (6.8) 12 (4.3) 29 (5.5) 0.215
Work with reduced or without pay to repay debt to
employers or recruitment agencies
11 (4.4) 9 (3.2) 20 (3.8) 0.490
Self-identified as a victim of trafficking & exploitation 8 (3.2) 7 (2.5) 15 (2.8) 0.643
Menace of penalty** 23 (9.2) 47 (17.0) 70 (13.3) 0.009
Threats and physical harm 4 (1.6) 2 (.7) 6 (1.1) 0.343
Threat of violence against oneself and/or loved ones 2 (0.8) 2 (0.7) 4 (0.8) 0.918
Physical abuse while traveling to the workplace 2 (0.8) 0 (0.0) 2 (0.4) 0.136
Physically abuse at the workplace 1 (0.4) 1 (0.4) 2 (0.4) 0.942
Sexual abuse at the workplace 0 (0.0) 0 (0.0) 0 (0.0) -
Restriction and depravation 19 (7.6) 34 (12.3) 53 (10.1) 0.075
Employer confiscated passport, ID or other legal
documents*
4 (1.6) 16 (5.8) 20 (3.8) 0.012
Not allowed to communicate freely with family and friends 9 (3.6) 16 (5.8) 25 (4.7) 0.241
Not allowed to quit the job* 5 (2.0) 15 (5.4) 20 (3.8) 0.040
Denied medical treatment when sick or after an injury 4 (1.6) 1 (0.4) 5 (0.9) 0.143
Imprisoned at the workplace 1 (0.4) 1 (0.4) 2 (0.4) 0.942
Denied proper food and/or water 1 (0.4) 1 (0.4) 2 (0.4) 0.942
Exploitative labor practices* 14 (5.6) 30 (10.8) 44 (8.3) 0.030
Not allowed to keep the earned money** 3 (1.2) 19 (6.9) 22 (4.2) 0.001
Not given a day off 6 (2.4) 11 (4.0) 17 (3.2) 0.308
Forced to work excessively long hours 6 (2.4) 11 (4.0) 17 (3.2) 0.308
Forced to consume alcohol by my employers 0 (0.0) 0 (0.0) 0 (0.0) -
*p < .05, **p < .01 There's a significant difference between male and female in the same variable
Within restriction and depravation, not allowed to communicate with loved ones (4.7%),
not allowed to quit the job (3.8%), and confiscation of passport, identification card or other legal
documents (3.8%) were reported as the three most common conditions. For exploitative labor
practice, participants cited not allowed to keep the money earned (4.2%) as the most frequent
41
situation. Threats or physical harm from recruitment agents or employers were uncommon as
only 1.1% of the participants had experienced them.
A crosstab analysis was run to explore whether there were significant differences
between male and female participants for each human trafficking condition. We found that
women were significantly more likely than men to experience menace of penalty conditions,
2
(1, N=527) = 6.88, p = .009, such as confiscation of passports, identification and other legal
documents,
2
(1, N=527) = 6.28, p = .012, and not allowed to quit the job,
2
(1, N=527) = 4.20,
p =.04. Women also experienced more exploitative labor practices,
2
(1, N=527) = 4.70, p = .03,
in particular, not allowed to keep their wages,
2
(1, N=527) = 10.52, p = .001.
Experiences of Escape or Rescue, Protection and Justice, and Health Problems
15 participants
1
who self-reported as a victim of human trafficking further answered a
series of questions related to (1) escape or rescue actions taken or received to get out of the
trafficking situation; (2) protection and justice services and activities they have received or
pursued; and (3) health problems that persisted because of trafficking.
For escape or rescue, the most common response was to call their family members for
help (53%), followed by 47% who escaped the workplace on their own to contact police or a
non-governmental organization (NGO). In terms of protection and justice experiences, 47% of
these participants had medical check-ups and treatment while 33.3% had filed and successfully
received worker insurance payments to cover the lost wages withheld from their former
employers. For health problems, most of the self-identified victims reported that they had to
1
Note: The remaining 66 participants who were identified as trafficked victims did not answer questions related to
these 3 concepts because they did not self-report as a victim, and therefore were not requested to answer these
additional questions.
42
struggle with mental health problems such as anxiety (67%) and depression (47%). Only one
individual mentioned physical injury as a health problem.
Men were more likely to escape the trafficking situation independently and sought help
from police, a NGO, or an embassy,
2
(1, N=15) = 11.25, p = .001. These findings are consistent
with other human trafficking research in Indonesia that women are more susceptible to becoming
victims of human trafficking and less likely to seek help (Bajari, 2013). Table 2.4 summarizes
the experiences of self-identified trafficking victims during and after exploitation.
Table 2.5 Experiences of escape or rescue, protection and justice, and health problems among self-
identified human trafficking victims in Indramayu
Categorical Variables
Among Self-Identified Victims
Male
n (%)
Female
n (%)
Total
n (%)
p
Sex 5 (33.3) 10 (66.7) 15 (100.0) -
Escape or rescue 5 (100.0) 10 (100.0) 15 (100.0) -
Called family members 2 (40.0) 6 (60.0) 8 (53.3) 0.464
Escaped on their own and went to the
police, NGO or embassy for help**
5 (100.0) 2 (20.0) 7 (46.7) 0.001
Stopped showing up for work 0 (0.0) 5 (50.0) 5 (33.3) 0.053
Eventually dismissed by the employer 0 (0.0) 5 (50.0) 5 (33.3) 0.053
Called the hotline for help 0 (0.0) 5 (50.0) 5 (33.3) 0.053
Called the embassy/consulate 0 (0.0) 4 (40.0) 4 (26.7) 0.099
Called passersby near the workplace 2 (40.0) 2 (20.0) 4 (26.7) 0.409
Rescued by the police, NGO, and/or
embassy employees
0 (0.0) 4 (40.0) 4 (26.7) 0.099
Approached fellow workers for help 0 (0.0) 1 (10.0) 1 (6.7) 0.464
Protection and justice 5 (100.0) 9 (90.0) 14 (93.3) 0.464
Had medical check-up and treatment 2 (40.0) 5 (50.0) 7 (46.7) 0.714
Filed a worker insurance claim to receive
my lost wages
1 (20.0) 4 (40.0) 5 (33.3) 0.439
Successfully received their worker
insurance payment
0 (0.0) 5 (50.0) 5 (33.3) 0.053
Placed in a rehabilitation shelter 1 (20.0) 1 (10.0) 2 (13.3) 0.591
Received financial assistance from the
government
1 (20.0) 0 (0.0) 1 (6.7) 0.143
Health problems 5 (100.0) 10 (100.0) 15 (100.0) -
Anxiety 4 (80.0) 6 (60.0) 10 (66.7) 0.439
Depression 2 (40.0) 5 (50.0) 7 (46.7) 0.714
Physical injury 0 (0.0) 1 (10.0) 1 (6.7) 0.464
*p < .05, **p < .01 There's a significant difference between male and female in the same variable
43
Discussion
Survey data from this study confirms previous research that described Indramayu as a
region with extreme poverty where half of the households lived under national poverty line.
There was also high unemployment rate, especially for women who experienced unemployment
at 3.5 times higher rate than men. The lack of job opportunities, early marriages, and lower
education attainment put pressure on Indramayu women to migrate and seek work elsewhere to
earn money to take care of their family. In addition, cultural factors of filial piety to help parents
financially as well as making a sacrifice to provide better opportunities for younger siblings and
children were reflected from the focus groups:
“I chose to migrate abroad because what other choices do I have? There is nothing for us
to do here. I need to make money to help my parents and pay for my younger siblings to
attend school so they don’t end up working in farms or cleaning homes and building like
me for the rest of their lives. I can make 2.5 million Rp (216 USD) a month working as a
helper for a family in Saudi Arabia. There is no way for me to get the same amount of
pay even if I can find work around here. Some of us struggle to even make 2.5 million Rp
for the entire year.” (Returned female migrant, 27 years old, former domestic helper in
Saudi Arabia).
While women disproportionately faced more socioeconomic challenges, both men and
women migrated out of Indramayu at about the same rate of 26% and 30% respectively. Findings
from another survey study by Listinai (2017) found that most male migrants (70%) travelled
domestically to cities such as Jakarta, the urban capital city of Indonesia, to look for low-skilled
labor jobs while 86% of female migrants opted for international destinations such as Taiwan,
Oman, Saudi Arabia, UAE, and Singapore. International labor migration was preferred among
44
women due to the demand for Indonesian female domestic helpers, promise of higher wages, and
social prestige associated with working abroad. The international nature of the employment often
placed women at a much higher risk of human trafficking, indicated from the data that human
trafficking prevalence was higher among women, and they faced more restrictive and
exploitative situations such as having their passports confiscated, trapped in a job and could not
leave, as well as not being paid by their employers.
“I did not realize that my first three months of work was considered as probational period
and I was not paid anything for it. I was told to work every day even though our contract
stated that I was entitled to one day off per week. I could not leave because my Kafeel
(sponsor) took away my passport and gave it to my boss. He said they were afraid we
would run away so they had to hold on to it.” (Returned female migrant, age 22 years old,
former domestic helper in Saudi Arabia).
While prevalence of human trafficking among the general population was at 15.4%, the
rate was much higher among returned migrants, especially female migrant workers, 60% of
whom experienced at least one condition of human trafficking. Nevertheless, only 10% of
migrant workers identified themselves as a victim of human trafficking and exploitation due to
the lack of awareness of the term, perdagangan manusia, translated literally to English as
‘trading of human beings’. Upon being asked to explain what perdagangan manusia was, one
female participant in FGD responded: “I agree with perdagangan manusia, because of it, I can
go overseas as long as it causes me no harm. Or maybe it is like this? For example, there is a
sponsor here, there is a sponsor there. Between sponsors, they cooperate and switch people. That
is perdagangan manusia?”
45
Upon further examination of the survey data, respondents who self-identified as a victim
was aware of perdagangan manusia term because they had experienced an average of 4
conditions of human trafficking (compared to the remaining victims who experienced an average
of 2 human trafficking conditions). Most of these self-identified victims also received protective
services after being trafficked (e.g., medical check-up or treatment) which gave them more
clarification to understand what they went through was associated with human trafficking.
This leads to the discussion if it is appropriate to count victims of human trafficking
based on the criteria that they have experienced any one condition of human trafficking.
Indonesia law loosely defines human trafficking victim as “a person suffering from
psychological, mental, physical, sexual, economic, and/or social trauma caused by the criminal
act of trafficking in persons” (Department of Justice and Human Rights Republic of Indonesia,
2007). It can be argued that any of the 16 human trafficking conditions in this study leads to
trauma. However, this determination is subjective. Most prevalent conditions in this study are
related to restriction of movement and exploitative labor practices (e.g., confiscation of
passports, not allowed to communicate with loved ones, not allowed to quit the job, debt
bondage, forced to work excessive hours) rather than threats and physical harm that are more
severe in nature. During the presentation of these findings to the local stakeholders in Indonesia,
one audience member suggested that victims should only be counted if they experience threats
and physical harm, the most severe form of human trafficking. This suggestion was met with
resistance since it excluded common exploitative labor conditions that still fell under human
trafficking definition and would reduce the prevalence of human trafficking in this study from
15.4% to 1.1%. A representative from ILO pointed out that there were over 70 indicators of
forced labor (used interchangeably as human trafficking) and even this study only explored a
46
portion of these indicators (ILO, 2012b, pp 21-25). To reduce human trafficking as only those
who experience threats and physical harm would be erroneous to previous protocols and
recommendations that have been published and followed by multilateral organizations such as
ILO and IOM. The prevalence of 15.4% in this study should still stand.
Furthermore, detailed prevalence of each human trafficking condition in Table 2.4
provides useful evidence for migrant organizations, labor departments, law enforcement
agencies, and policy makers to determine relevant resources that should be allocated to their
organizations to collectively address the problem. For example, since 1.1% of the 18-39 years
old population (over 7,000 people) have experienced threats and physical harm, appropriate
resources should be allocated to hire sufficient healthcare professionals to reach these 7,000
people and provide them with counseling and medical services. The high prevalence of
restrictive movement, debt bondage, and labor exploitation at the place of employment means
that Indonesian government should continue to work with foreign governments to ensure these
practices are outlawed. In addition, more services should be set up in countries with the high
number of Indonesian migrant workers and provide channels for migrant workers to report
violations and seek justice. More accessible channels to seek financial retribution are also needed
in ‘source’ areas like Indramayu. A male FGD participant summed up his frustration to seek
unpaid wages after his wife was not paid fairly by her employer in UAE.
“Our government doesn’t care. They did nothing for us. My wife was not paid in full
amount from the contract and we could not find a way to get our money back. Before she
left Indonesia, we had to pay worker insurance claim to the government which was
supposed to help us get some money back if she got into trouble with her employer. But
the process to get any money back is too complicated. We have to go to the government
47
office in Jakarta in person and it is too far from where we live in Indramayu. Where are
we supposed to stay in Jakarta? It takes days to file papers and wait for any responses.
We just gave up and didn’t bother filing this worker insurance claim.”
It should be noted that to address human trafficking and exploitation against migrant
workers, Indonesia government has since implemented a ban against Indonesian women from
traveling to the Gulf countries to work as domestic helpers (Park, 2017). This ban did not stop
women from trying to migrate abroad, and further forced them to seek unregulated brokers and
place them to at greater risks of human trafficking. Areas such as Indramayu will continue to
face high level of out migration due to poverty and lack of job opportunities. Restricting
migration will not solve the problem.
Limitations and Future Direction
Limitations from survey studies such as memory failure and social desirability, especially
in a face to face interview format, could underestimate the experiences of human trafficking.
Respondents might feel ashamed of their past experience and did not wish to reveal the incident
to the interviewer. This study also restricted the age of respondents from 18 to 39, and the
estimated 98,654 victims is only among the population in this age group. The parameter of this
study is limited to the general experiences of human trafficking among returned migrants. Hence,
prevalence of human trafficking in Indramayu industries such as rice farming, fishing, and oil
production is not explored. Probability-based household survey is not an appropriate method to
reach trafficked victims who are currently living in exploitative situation. The prevalence of
human trafficking in this study should be inferred only to those who have migrated and
experienced trafficking violation in their employment destination. More precise estimate of
human trafficking that includes current victims in the local area in Indramayu would need to
48
incorporate other approaches such as respondent-driven sampling surveys among agricultural
and oil production workers.
For future direction, similar multi-stage sampling surveys could be conducted to
determine prevalence of human trafficking at other regencies in West Java. Other provinces with
high level of human trafficking victims such as West Kalimantan, East Java, Central Java, and
North Sumatra could be selected as subsequent sites for prevalence studies. To obtain more
reliable national estimates of human trafficking in Indonesia, mixed strategies of using multi-
stage sampling surveys in ‘source’ areas, and respondent-driven surveys in ‘destination’ cities
(e.g., Jakarta, and Riau Islands) could be conducted. To save data collection costs, formal
partnerships with Badan Pusat Statistik (Central Bureau of Statistics in Indonesia) and related
government agencies are crucial.
This study contributes to a small but growing field of empirical research on the
prevalence of human trafficking at the local level (Curtis et al., 2008; Williamson et al., 2012;
Zhang, 2012a; Zhang et al., 2014; Pitts et al., 2015). To the author’s knowledge, this is the first
study that employed probability-based sampling survey method to estimate the magnitude of
human trafficking in West Java, Indonesia. The prevalence of 15.4% or 98,654 victims in
Indramayu in this study is a steep contrast to the national prevalence of 0.3% or 736,100 victims
by the 2016 Global Slavery Index report (Walk Free Foundation, 2016). Since West Java is a
province with over 46 million people and has a high number of assisted victims across other
regencies, it is possible that the number of victims in West Java alone may exceed the national
estimates from the 2016 Global Slavery Index report. In addition, this study provides more
details of human trafficking conditions that returned migrants have experienced, and the
disparity between males and females. Female migrants experience higher rate of human
49
trafficking, as well as encountered more restriction on their movement, and exploitative labor
practices such as unpaid wages from their employers. Migrant women who work abroad also
have less options to escape their situation. Rather than banning Indonesian women from
migrating abroad to seek work, more resources should be allocated to monitor domestic workers,
as well as to provide more accessible channels to receive protective services, and retributive
justice.
50
CHAPTER 3: ASSESSING THE EFFECTS OF A HUMAN TRAFFICKING
PREVENTION CAMPAIGN AND ITS DETERMINANTS
THROUGH A MULTILEVEL COMMUNICATION MODEL
Introduction
This chapter focuses on the second and third steps of the public health approach in
addressing the problem of human trafficking: (2) identifying determinants of human trafficking,
and (3) designing and testing prevention strategies (Alpert & Chin, 2017).
Indonesia is a major source country for human trafficking where young women and men
migrate for work abroad and end up in exploitative working conditions (Department of State,
2013). The World Bank (2017) estimates that up to 24 percent of Indonesia's 9 million
international migrant workers experience abusive working conditions that can be categorized as
human trafficking. These conditions include withheld salary, excessive working hours, total
restriction of movement, verbal and physical abuse, and confiscation of travel documents.
Despite the scope of the problem of human trafficking, awareness remains low, as evidenced by
a research study in Indonesia where only 13% of the respondents were familiar with this crime
(Lindgren, 2010). Communication efforts are therefore needed to inform at-risk populations to
take preventive measures against human trafficking, discuss with their peers and family of the
risks, and report potential incidents within their community to the authorities.
Given the widespread problem of human trafficking in Indonesia, there is an urgent need
to create large-scale campaigning efforts that raise awareness and increase prevention of human
trafficking. One example is the MTV EXIT campaign which was initiated by MTV Networks
and USAID in 2006 to deliver television programs such as documentaries, public service
announcements, music videos, and live concert events. So far, over $108 million have been
51
invested for the MTV EXIT campaign to reach millions of households in Asia and the Pacific
(USAID, 2012).
Media campaigns are created to produce specific goals, outcomes, and impacts by
targeting a sizable number of people within a particular timeframe through a series of activities
via communication and media channels (Rogers & Storey, 1987). Some studies, such as
Snyder’s (2001) analysis of 48 mediated health promotion campaigns, discover that they can
trigger a modest 7-10% effect on behavioral change. An earlier study that evaluated the effects
of the MTV EXIT campaign in Thailand, China, India, and Japan confirmed favorable outcomes:
exposure to the TV program shifted the Knowledge Attitude Practice (KAP) scores by 7-12%
(Thainiyom, 2011). Media campaigns are becoming important means of informing people about
issues such as human trafficking.
There is a lack of empirical studies on the effects of anti-trafficking interventions on
knowledge, attitudes, and prevention behavior (Magenta, 2007; GAATW, 2010; Van de Laan et
al., 2011; Thainiyom, 2011). These scholars have raised concerns over the low quality of study
designs that measured only a few variables (e.g., awareness, knowledge and behaviors), and did
not meet basic standards of scientific research methods (e.g., lack of randomized control trials to
minimize effects of confounds). Moreover, studies investigating anti-trafficking campaign
effects tend to gather data at only one point, or over a short period of time using cross-sectional
rather than longitudinal panel samples. These practices make it difficult for researchers to claim
the effectiveness of anti-trafficking programs and limit the use of more advanced data analysis in
identifying determinant factors that influence counter-trafficking behaviors over time.
Because of methodological limitations of earlier research, this study will include a more
extended period of data collection, using randomized control trial in a panel design in order to
52
gain a better understanding of human trafficking prevention behavior and its predictors. This
design, in combination with a more advanced statistical methodology, leads to a better
understanding of anti-trafficking campaign effects across a time span. This brings us to the
primary aim of this study, which is to test the efficacy of the MTV EXIT documentary in
changing human trafficking prevention outcomes and identify significant determinants of these
prevention outcomes.
A Multilevel Communication Model (MCM)
Figure 3.1 displays key components of the Multilevel Communication Model which
prioritizes a communication ecological approach into the model construction. Background
influences from the Integrative Model of Behavioral Prediction (IMBP; Fishbein & Ajzen, 2010)
and communication ecological concepts from the Communication Infrastructure Theory (CIT;
Ball-Rokeach et al., 2001; Kim & Ball-Rokeach, 2006a) are placed under determinants of
prevention outcomes. The MCM has the assumption that behavior, intention, skills and abilities,
and their psychosocial variables (knowledge, attitudes, perceived risks, self-efficacy, perceived
norms) are influenced by background factors (demographic, socioeconomic, prior experience
with migration and trafficking, and environmental constraints) and communication ecological
factors such as integrated connection to the storytelling network, communication action context,
interpersonal discussion about trafficking, size of migrant network, and media exposure to
trafficking information.
53
Exploring Campaign Effects through the MCM
A successful awareness-raising and prevention campaign requires a strategic
communication plan that creates and places the right messages according to the principles of
effective behavioral change theories. Salmon and Atkin (2003) propose that research should
include analyzing the behavioral aspects of the problem to determine what actions are to be taken
and specifying key segments of the population whose behaviors are to be changed. Researchers
should then trace backwards from the focal behavior to identify psychosocial factors that
influence behavior (e.g., knowledge, attitudes, social norms, and intention). Many successful
health campaigns have utilized behavioral change theories to guide research design and to obtain
data about the target population (Webb, et. al., 2010). Behavioral change theories help
researchers and practitioners decide which beliefs, attitudes, and other related factors should be
addressed in the message design to best influence the target population to change their behaviors.
Figure 3.1 A Multilevel Communication Model
54
The Integrative Model of Behavioral Prediction (IMBP) is one of the latest behavioral
change communication theories that integrate precursor factors such as knowledge, attitudes,
perceived norms, self-efficacy, perceived risks, intention to perform the behavior, skills and
abilities, and environmental constraints, that have been demonstrated to influence health
behaviors (Fishbein, 2000; Fishbein & Ajzen, 2010). IMBP assumes that behavior is mainly
predicted by intention, and individuals may or may not be able to perform their intention because
of environmental constraints and/or lack of skills and abilities to carry out the behavior. Intention
to perform a specific behavior is also directly influenced by attitudes (positive or negative toward
performing the behavior), perceived norms (perceptions about what others think and do
regarding the behavior), and self-efficacy (perceptions of one’s ability to perform the behavior).
Therefore, our first research question would explore the efficacy of the MTV EXIT documentary
on these behavioral and psychosocial outcomes from IMBP.
RQ1: What are the effects of the MTV EXIT documentary in shaping the human
trafficking prevention outcomes of knowledge, attitudes, perceived norms, self-efficacy,
perceived risks, intention, skills and abilities, and human trafficking prevention behavior?
Background Influences as Determinants of Prevention Outcomes
Despite the inclusion of background variables (e.g., demographics, socioeconomic, and
media exposure) and environmental constraints in the IMBP, existing empirical studies still have
not fully explored the relationships of background influences and environmental constraints with
knowledge, attitudes, perceived norm, self-efficacy, intention and behavior (Bleakey et al., 2011;
Rhodes et al., 2007). The IMBP continues to demonstrate individualistic bias by prioritizing
cognitive variables and assuming background variables would only have indirect effects on
behavior. Yzer (2012) justifies the peripheral placement of background variables in the model by
55
stating “Whereas there may be empirical associations between these (background) variables and
behavior, there are no theoretical reasons to expect that these variables always and in the same
manner shape beliefs.”
An examination of the literature in the broader field of health communication suggests
that individual factors such as sex, age, education, socioeconomic status, and direct personal
experience with the health problem correlates with health-related behaviors and proximal
psychosocial factors (Dobransky & Hargittai, 2012; Galarce et al., 2011; Macphail et al., 2009;
Peltzer et al., 2009; Pinquart & Duberstein, 2004; Taylor-Clark et al., 2010; Tenkorang &
Owusu, 2010). This leads to the second research question for this chapter:
RQ2: What background factors (age, sex, education, socioeconomic status, and prior
experience with migration and trafficking, and environmental constraints) are significant
determinants of human trafficking prevention outcomes?
The Role of Communication as a Determinant of Human Trafficking Prevention
There is a growing scholarship on the role of communication as a determinant of health
and Communication Infrastructure Theory (CIT) has been utilized as a theoretical perspective to
guide this field of research (Matsaganis & Wilkin, 2015; Wilkin, 2013; Wilkin & Ball-Rokeach,
2006, 2011; Wilkin et al., 2010, 2011). CIT proposes that each neighborhood has a distinctive
storytelling network which consists of residents (micro-level), local media, and community-
based organizations (meso/organizational-level; Ball-Rokeach et al., 2001; Kim & Ball-Rokeach,
2006b). CIT claims that individuals’ problem-solving capacities (e.g., human trafficking
prevention) are associated with their connection to a viable storytelling network rooted in a
communication action context (e.g., social and physical environment) that facilitates or hinders
residents to communicate with each other (Ball-Rokeach et al., 2001). In this study, the
56
communication action context consists of communication hotspots (places where residents come
to socialize) and comfort zones (places where residents feel safe and comfortable to discuss a
particular issue; Wilkin et al., 2011). The lack of communication hotspots and comfort zones in
the neighborhood could make it difficult for community organizations to organize a gathering
event for residents to discuss and promote prevention of human trafficking.
Integrated Connection to a Storytelling Network (ICSN)
Strong connection to a storytelling network allows residents to have more opportunities
to interact with one another and solve human trafficking problem within their community,
provided the residential areas have sufficient spaces to facilitate this communication. A resident
with strong ICSN is well-integrated into their neighborhood, knows their neighbors, is connected
to community-based organizations, and stays up-to-date via local media. A resident with weak
ICSN is more isolated, not well-connected to neighbors, community organizations and local
media. ICSN is found to correlate with health outcomes such as increased knowledge about
breast cancer prevention, increased physical activities, and better access to healthcare (Wilkin,
2013). Consistent with previous research that investigates the link between ICSN and health
outcomes, this study will explore:
RQ3: How are integrated connection to a storytelling network and communication action
context (defined as communication hotspots and comfort zones) related to human
trafficking prevention outcomes?
Communication Ecology of Human Trafficking Prevention
While it is helpful to investigate the relationships between ICSN and communication
action context with human trafficking prevention, it should be noted that individual’s
communication ecology consists of a wider range of national and global media channels, social
57
media, Internet, and social networks. To investigate communication as a determinant of human
trafficking prevention, it is important to look at the wider communication resources related to the
human trafficking issue. The communication ecology approach assumes that people actively seek
a combination of available communication resources (e.g., interpersonal, social networks, mass
media, Internet) to satisfy a particular goal in their everyday lives (Wilkin, 2013). For example, a
person whose goal is to migrate safely may find information through classified ad websites and
face-to-face interaction with former migrants in their network who have experience in dealing
with seeking work abroad. The communication ecology approach goes in line with a growing
body of health communication research that explores the relationships of interpersonal
discussion, the size of social network, and media exposure with various health outcomes
(Ackerson & Viswanath, 2009; Morton & Duck, 2001; Seo & Matsaganis, 2013; Valente &
Saba, 1998). In the context of this study, individuals with a larger migrant network should have
more social support to provide them with information, advice, and other resources to make
informed decision when they choose to migrate to seek work in other cities. People who have
interpersonal discussion about human trafficking, and expose themselves to trafficking
information through mass media, social media, and Internet should also have more knowledge,
favorable attitudes, and take actions to protect themselves against human trafficking. This leads
to the last research question:
RQ4: How are interpersonal discussion about human trafficking, size of migrant
network, and media exposure to human trafficking information related to human
trafficking prevention outcomes?
58
Methods
The data from this study is drawn from the survey administered in Indramayu in West
Java, a district with over 1.71 million people in Indonesia (Central Statistics Agency of
Indramayu Regency, 2016). This research study is a longitudinal randomized control trial with a
panel design. Surveys were implemented using multi-stage sampling from the regency level, to
districts, neighborhoods, and households. A random sample of 6 districts in Indramayu were
chosen as the study sites. Fifteen field researchers visited 90 Rukun Tetangga (RT -
neighborhoods) in the selected districts and conducted face-to-face interviews with 5-6
households in each RT. A Kish Grid was used for each household where more than one person
was eligible to participate to make sure there was no systematic selection bias of the respondents.
To be eligible to participate, respondents had to be 18-39 years old, live in Indramayu, completed
at least SD education (elementary school), be able to read and write Bahasa Indonesian, did not
know anything about the MTV EXIT campaign, and, for the exposed group, had the means to
view the MTV EXIT documentary. Overall, 577 participants were recruited into the study but 50
declined to participate. The remaining 527 people who agreed to participate were subsequently
randomly assigned to exposed (N=319) or control (N=208) conditions. Institutional review board
at the investigator’s university approved the study.
Data were collected in three waves from April 2014 through August 2014. In the first
wave, baseline data were collected through face-to-face interviews (N=527). In the second wave,
Posttest 1 data were collected one week after the baseline interviews, also using face-to-face
interviews (exposed, N=319, control, N=208). The third wave of data collection, Posttest 2, took
place following a four-month interval which used a telephone survey (exposed, N = 196, control,
N=142). At the baseline, participants in the exposed group were given a video compact disc
59
(VCD) with the documentary Enslaved: an MTV EXIT Special after the survey interview was
completed. Each participant was instructed to watch the documentary in its entirety for a follow-
up interview about one week later. There were no drop outs at posttest 1 and all participants were
given an MTV EXIT-branded pen to thank them for their time. In the second follow-up (4
months after posttest 1) interviewers called the respondents up to twelve times to complete the
interviews. The drop-out rate was 38.6% for the exposed group and 31.7% for the control group.
The drop-outs included phone numbers that became permanently inactive, and some participants
who declined to participate. Figure 3.2 summarizes the flowchart of the participant recruitment
and group assignment across time.
MTV EXIT Documentary
The intervention material in this study was a 24-minute documentary film titled
Enslaved: An MTV EXIT Special, hosted by Dian Sastrowardoyo, a popular actress in Indonesia.
The Bahasa Indonesian film was produced in 2012 by MTV EXIT and featured stories of three
human trafficking victims; Siti, a woman who migrated to Malaysia and ended up as a domestic
worker in exploitative working conditions; Ismail, a man who was trapped in bonded labor in the
Figure 3.2 Flowchart of the participants in the randomized control trial
60
logging industry in Northern Sumatra; and Ika, a teenage girl who was deceived into forced
prostitution in Batam. The documentary started out with the three victims narrating their poor
living conditions, and aspirations to seek a better life. The story moved on to document the
process by which they were trafficked into exploitative work, and how they eventually were able
to escape from their situations. The film concluded with prevention messages that advised the
viewer to be skeptical of recruitment agents, to seek information and verify job opportunities
through someone they can trust, such as government agencies and migrant worker organizations,
before accepting the offer. They also suggested that anyone considering to migrate should have
complete possession of legal documents and identification before traveling abroad, to call the
hotline number if they need help or report suspicious activities, and to raise awareness of the
issue by sharing information with others. For manipulation check, 68% of the respondents in the
exposed condition answered that they watched the documentary from the beginning to the end
while the remaining 32% completed at least 18 minutes of the film. 84% of the participants also
indicated they saw the film one time while the remaining participants watched the documentary
at least two times.
Measures
All the variables of interest are described below. Outcome variables were collected at the
baseline, posttest 1 and posttest 2, unless otherwise stated. All background influences and
communication ecological variables were collected at the baseline.
Outcome Variables
Human Trafficking Prevention Behavior. For this outcome variable, participants were
asked if they have ever tried to seek information to migrate safely for work for either: (1)
themselves (2) family members, and/or (3) their friends. A follow-up question of how they find
61
safe migration information such as through “family member,” “a sponsor,” “newspaper,” and
“government agency (BP3TKI or Disnakertrans).” Each mentioned answer was accorded 1 point.
Respondents also answered whether or not they had called a counter-trafficking hotline number.
Yes was coded as 1 and No coded as 0. The scores were then aggregated to form a composite
index for this variable.
Intention to practice human trafficking prevention behavior. Participants rated 6
possibilities to answer the question, “If you were to migrate for work in the next few years, how
likely would you be to do any of the following?” using a 5-point Likert-type scale anchored at 1
= Very unlikely and 5 = Very likely. Two examples of the possibilities are: “become a member of
an NGO or migrants organization to receive advice and assistance on migration,” and “ask
friends and family for advice.” The items were aggregated to form a composite index. Intention
is transformed using square function (x to x
2
) to meet normality distribution. The Cronbach’s α
ranges from .83 to .90.
Skills and abilities. Participants responded to a question that asked if they remember a
hotline number, and to report the number. Eight follow-up items on the purpose of calling the
hotline such as, “to seek information about a recruitment agency,” “to verify the legitimacy of an
employer,” and “to report suspicious activity,” were also asked. “Yes” answers and the correct
answer on the hotline number were coded as 1 and “No” as 0. The 9 items were added to form an
aggregate index which ranges from 0-9.
Perceived norms. Participants were asked to rate their level of agreement (scale from 1 =
Strongly Disagree to 5 = Strongly Agree) with two statements: if they had an 18-year old
daughter, they would ask her to move away from home for work, and if they had an 18-year old
62
son, they would ask him to move away from home for work. The two items were averaged to
form an index and its Cronbach’s α is from .72 to .89.
Efficacy. Participants rated 6 items on how effective they believed each method would be
in preventing a migrant worker from getting in trouble when working away from home—using a
5-point Likert-type scale from 1 = Not at all helpful to 5 = Extremely helpful. Examples of the
statements were “seeking information from the local and/or national government (BP3TKI
and/or Disnakertrans),” and “calling a hotline to seek advice on safe migration for work.” These
six items were averaged to form an efficacy composite score and the Cronbach’s alpha of the
scale is .73 to .92.
Perceived risks. Three items about the perceived risks of human trafficking anchored at 1
= Strongly disagree and 5 = Strongly agree. The three items are: “I believe that this is a serious
problem in Indonesia,” “I believe that a worker in trouble suffers serious negative consequences
in his/her life,” and “I believe that we should not be concerned about this issue in this country.”
The last item was reverse coded and all the three items were averaged to form a composite index
of perceived risks. This variable is transformed using square function (x to x
2
) to meet normality
distribution. The Cronbach’s α ranges from .81 to .83.
Attitudes regarding human trafficking prevention behavior. Participants were asked to
rate their level of agreement with five statements (on a scale from 1 = Strongly Disagree to 5 =
Strongly Agree): “finding information about safe migration is…easy,” and “…important,” and “a
hotline would be…useful in emergency,” “…helpful in providing information,” and “…useful in
emergency.” The Cronbach’s α ranges from .65 to .75 and all the items were averaged to form an
index.
63
Knowledge of human trafficking. Thirteen items were adapted from the Trafficking
Awareness Survey (Minnesota Advocates for Human Rights, 2003) and augmented by items
developed during our survey construction process. Participants chose either Yes or No if they
have heard of terms such as “human trafficking,” and “safe migration,” as well as statements
such as “A worker in trouble/human trafficking victim could be….people who left or were taken
away from their country or city and tricked or forced to do a job in which they were exploited,”
and “…someone who is forced to work longer hours than were written in the contract or
promised.” Each “Yes” answer was given one point and added to form a composite score out of
13. Higher scores showed greater knowledge about human trafficking.
Background Influences
Demographics. Variables such as sex (male = 0; female =1), age (number of years),
education (number of years of formal education completed), and annual household income (in
USD) were collected.
Prior experience with migration. Participants were asked if they had ever moved away
from home for work. Yes was coded as 1 and No was coded as 0.
Prior experience with human trafficking. This variable was measured by 15 items to
assess the degree to which the respondent has experienced conditions of human trafficking and
exploitation (Zhang, 2012b). Examples of the items are, “My employer confiscated my passport,
identification documents or other legal documents during my employment,” and “I was
psychologically abused with threat of violence against me and/or my loved ones.” Items that
described experiences that had happened to the survey participant were coded as 1, and no
experience items were coded as 0. All the items were summed to create an index that ranges
from 0-15.
64
Environmental constraints. Participants answered how likely each of these environmental
circumstances would influence them to migrate, using a 5-point Likert-type scale anchored at 1 =
Very unlikely and 5 = Very likely. Examples of the items include “death of a family member,”
“having large debts to pay off,” and “one of your parents lost his or her job.” The items were
averaged to form a composite score of 1-5. The Cronbach α is .81.
Communication Ecology
Integrated connection to the storytelling network (ICSN). This measure was developed by
Kim and Ball-Rokeach (2006b). It is computed using weighted summation of three interaction
terms between connections to local media (range = 1–10, M = 2.83, SD = 1.20), connections to
community organizations (range = 2–10, M = 2.48, SD = 1.46), and intensity of interpersonal
neighborhood storytelling (range=1–10, M = 2.72, SD = 1.55). Formula 1 was used for the
calculation of ICSN. In Formula 1, LC is the score for connections to local media, INS is the
score of intensity of interpersonal neighborhood storytelling, and OC is the score of connections
to community organizations (range 1–30, M = 7.69, SD = 2.82).
𝐼𝐶𝑆𝑁 = √𝐿𝐶 × 𝐼𝑁𝑆 + √𝐼𝑁𝑆 × 𝑂𝐶 + √𝑂𝐶 × 𝐿𝐶
2
(1)
Communication hotspots. This variable is an aggregate score, calculated by naming the
places in the community where the participants can get together and chat. Each place mentioned
(e.g., grocery store, mosque, and café) was coded with 1 point and the scores were summed.
Comfort zones. Participants were asked if they were comfortable talking about human
trafficking at the hotspot locations they had just mentioned. Each Yes answer was coded as 1 and
No was coded as 0. Scores were then summed.
21
Operationalization of ICSN can be found in Kim and Ball-Rokeach (2006a) and Kim et al. (2011).
65
Interpersonal discussion about trafficking. Participants were asked to list the people (e.g.,
“your spouse,” “your father,” and “your mother,”) with whom they had discussed about the
dangers of working away from home. Each person was given one point and the total number of
people mentioned were summed to create an aggregate score on this variable.
Migrant Network. Participants were asked “Did any of the following people move away
from home for work?” with 13 possible items (i.e., “my spouse,” “my mother,” “my daughter,”
and “other,”). Yes was coded as 1 and No was coded as 0. All the items were summed up to
create a score from 0-13.
Media exposure to trafficking. Exposure to human trafficking-related information via
different media channels was measured by the question, “How often have you seen or heard
information about workers in trouble (human trafficking) in the following way?” Participants
responded to the 7 items: “newspaper,” “magazine,” “radio,” “television,” “internet,” “social
media,” and “billboard” using a 5-point Likert-type scale anchored at 1 =Not at all and 5 = All
the time. Scores are averaged from the 7 items with the higher number indicating greater
exposure. The Cronbach’s α is .84.
Statistical Analysis
The analytic strategy for this study includes computing descriptive statistics of the
background influences and communication ecological variables. To test the effects of the MTV
EXIT documentary (RQ1), we ran General Linear Model Repeated Measures tests for each
outcome variable at the baseline, posttest 1 and posttest 2 and compare them between the exposed
and control conditions. To identify the determinant factors (RQ2-RQ4), multiple regression
analyses were conducted on each outcome variable. The IBM SPSS Statistics 25 Program was
used for all the statistical analyses.
66
Results
Table 3.1 displays descriptive statistics of the participant characteristics by group
assignment and data collection period. There were no drop outs during the baseline and posttest 1
data collection, and data in each variable was reported under the same column. At the baseline,
47% of the participants were men, and had an average age of 28.3 years old and 8.7 years of
formal education. In other words, the participants, on average, have not completed SMP (Grade
9) education. Average annual household income was $1,858.60.
To ensure similar distribution in participant demographic characteristics between the
control and the exposed groups, crosstab analysis of sex, and t-test analyses of age, annual
household income, and education were conducted at baseline/posttest 1 and posttest 2. There
were no significant differences in participant demographic characteristics between the control
and the exposed groups across all data collection points.
For attrition analysis, we conducted crosstab and t-test analyses between those who
completed baseline/posttest 1 and posttest 2 samples with those who dropped out after
baseline/posttest 1. No significant differences were found, and attrition should not affect other
statistical analyses in this study.
67
Table 3.1 Characteristics of the participants in Indramayu, by group assignment and data collection period
Categorical Variables
Baseline/Posttest 1 Posttest 2
Exposed
(N=319)
Control
(N=208)
Total
(N=527)
Exposed
(N=196)
Control
(N=142)
Total
(N=338)
n (%) n (%)
Background Influences
Sex
Men 150 (47.0) 100 (48.1) 250 (47.4) 92 (46.9) 69 (48.6) 161 (47.6)
Women 169 (53.0) 108 (51.9) 277 (52.6) 104 (53.1) 73 (51.4) 177 (52.4)
Experience with migration 81 (25.4) 68 (32.7) 149 (28.3) 49 (25.0) 44 (31.0) 93 (27.5)
Continuous Variables M (SD) M (SD)
Background Influences
Age (in years) 28.2 (6.5) 28.4 (6.9) 28.3 (6.6) 28.0 (6.3) 28.3 (6.7) 28.1 (6.5)
Education (in years) 8.8 (2.9) 8.6 (2.7) 8.7 (2.8) 8.9 (3.0) 8.9 (2.9) 8.9 (3.0)
Annual household income (USD) 1,758.8 (1,188.4) 2,011.7 (1,137.4) 1,858.6 (1,174.0) 1,704.2 (1,100.8) 2,009.4 (1,198.2) 1,832.4 (1,150.9)
Experience with trafficking 1.3 (0.9) 1.3 (1.0) 1.3 (1.0) 1.3 (0.8) 1.3 (0.9) 1.3 (0.9)
Environmental constraints (range 1-5) 3.2 (0.6) 3.2 (0.8) 3.2 (0.7) 3.1 (0.6) 3.2 (0.8) 3.2 (0.7)
Communication Ecology
ICSN (range 1-30) 7.8 (2.7) 7.5 (3.0) 7.7 (2.8) 7.6 (2.8) 7.5 (2.8) 7.6 (2.8)
Communication hotspots 2.1 (1.2) 1.9 (1.1) 2.0 (1.1) 2.1 (1.3) 1.9 (1.1) 2.0 (1.2)
Comfort zones 2.6 (1.2) 2.2 (1.0) 2.5 (1.2) 2.6 (1.3) 2.2 (1.0) 2.4 (1.2)
Interpersonal discussion about
trafficking
1.7 (1.5) 1.7 (1.6) 1.7 (1.6) 1.6 (1.6) 1.7 (1.5) 1.6 (1.6)
Migrant network size 2.4 (1.6) 2.3 (1.6) 2.3 (1.6) 2.4 (1.7) 2.3 (1.6) 2.3 (1.7)
Media exposure to trafficking
(range 1-5)
1.7 (0.6) 1.6 (0.7) 1.6 (0.6) 1.7 (0.6) 1.6 (0.7) 1.7 (0.7)
68
Effects of the MTV EXIT Documentary
Table 3.2 answers RQ1 by summarizing the efficacy of the MTV EXIT documentary on
the participants’ knowledge, attitudes, norms, efficacy, perceived risks, skills & abilities, intention,
and behavior related to human trafficking prevention at one-week (posttest 1) and four-month
follow-up (posttest 2). Overall, there was a significant difference between the exposed and control
groups on the outcome variables over time, F(20, 1328) = 3.592, p < .001, η
2
= .051. In particular,
we found significant documentary effects on three variables: knowledge, F(2, 672) = 4.44, p < .05,
η
2
= .013; perceived risks, F(2, 672) = 5.625, p <.05, η
2
= .016; and skills & abilities, F(2, 672)
= 14.517, p <.001, η
2
= .041. Moreover, moderate documentary effects were also found on
efficacy, F(2, 672) = 2.988, p = .051, η
2
= .009, and intention F(2, 672) = 2.584, p = .076, η
2
=
.008.
Further t-test analyses of all the outcome variables were conducted to assess the significant
differences between the two groups at each follow-up. At posttest 1, knowledge, t(525) = -3.54, p
< .001, and skills & abilities, t(525) = -5.122, p < .001, had significantly higher values in the
exposed condition. At posttest 2, three variables were found to have moderately significant higher
scores in the exposed group; attitudes, t(336) = -1.866, p = .063; perceived risks, with t(336) = -
1.711, p = .088, and intention, t(336) = -1.874, p = .062. We did not find any effect on the human
trafficking prevention behavior at posttest 1 nor posttest 2.
69
Table 3.2 Documentary effects on knowledge, attitudes, norms, efficacy, perceived risks, skills & abilities,
intention, and behavior at 1-week and 4-month follow-up
Human Trafficking Prevention
Outcomes
Baseline
Posttest 1
(1-Week Follow-up)
Posttest 2
(4-Month Follow-up)
Mean (95% CI)
Behavior (range 0-9)
Control 0.89 (0.65-1.14) 0.85 (0.60-1.09) 1.47** (1.08-1.85)
Exposed 1.03 (0.83-1.24) 0.97 (0.77-1.17) 1.07 (0.78-1.36)
Intention (range 1-25)
Control 12.32 (11.44-13.20) 13.52** (12.77-14.27) 14.53*** (13.93-15.13)
Exposed 11.95 (11.20-12.70) 14.43*** (13.79-15.07) 15.28***
a
(14.77-15.79)
Skills & Abilities (range 0-7)
Control 1.87 (1.63-2.12) 1.82 (1.55-2.10) 2.06 (1.76-2.36)
Exposed 1.92 (1.72-2.13) 2.51**
c
(2.27-2.74) 1.46***
b
(1.21-1.72)
Perceived Norms (range 1-5)
Control 2.61 (2.47-2.74) 2.59 (2.46-2.72) 3.21*** (3.07-3.35)
Exposed 2.91 (2.80-3.01) 2.85
b
(2.75-2.95) 3.23*** (3.12-3.35)
Efficacy (range 1-5)
Control 3.64 (3.51-3.78) 3.84** (3.75-3.93) 3.92*** (3.86-3.99)
Exposed 3.52 (3.40-3.63) 3.89*** (3.82-3.97) 3.98*** (3.93-4.03)
Perceived Risks (range 1-25)
Control 19.02 (18.29-19.74) 19.70 (18.97-20.44) 16.77*** (16.25-17.28)
Exposed 17.62 (17.00-18.23) 19.18*** (18.56-19.81) 17.36
a
(16.92-17.80)
Attitudes (range 1-5)
Control 3.65 (3.56-3.74) 3.85** (3.77-3.93) 3.88** (3.80-3.96)
Exposed 3.72 (3.64-3.80) 3.96*** (3.89-4.02) 3.98***
a
(3.91-4.05)
Knowledge (range 0-13)
Control 7.61 (7.07-8.16) 8.83*** (8.43-9.23) 10.23*** (9.9-10.56)
Exposed 7.82 (7.36-8.29) 9.62***
c
(9.28-9.96) 9.92*** (9.64-10.20)
*p < .10, **p < .05, ***p < .001, the value is significantly different when compared to the same variable from
the baseline (within effects)
a
p < .10,
b
p < .05,
c
p < .001, the value in the exposed condition is significantly different from the same variable in
the control condition within the same data collection point (between effects)
Determinants of Human Trafficking Prevention Outcomes
Table 3.3 answers RQ2-RQ4 by highlighting the determinants of human trafficking
prevention outcomes. These findings were obtained from the baseline data before participants
were assigned to exposed or control conditions.
70
For background influences (RQ2), age was found to be a significant determinant for skills
& abilities and intention. As participants got one year older, their skills and abilities increased on
average by .03 unit (p < .01) but their intention to practice human trafficking prevention behavior
decreased by .02 unit (p < .01). Annual household income was also a significant determinant for
skills & abilities (b = .00, p < .05), and intention (b = .00, p < .05). Prior experience with
migration was a positive determinant of perceived risks (b = 1.09, p < .05) and intention (b = .28,
p < .01), but a negative predictor for perceived norms (b = -.38, p < .05) and skills & abilities (b
= -.51, p < .01). Prior experience with human trafficking was a predictor of 3 outcome variables:
attitudes (b = -.06, p < .10), efficacy (b = -.08, p < .05), and intention (b = -.08, p < .05). It
should be noted that as participants had more experience with trafficking, they had less scores of
attitudes, efficacy, and intention. Environmental constraints were the strongest determinant of
prevention outcomes, as they were found to be a significant predictor for 7 outcomes: knowledge
(b = .39, p < .05), attitudes (b = .11, p < .001), perceived risks (b = 1.02, p < .001), efficacy (b =
.37, p < .001), perceived norms (b = .79, p < .001), skills & abilities (b = .26, p < .01), and
intention (b = .26, p < .001).
For communication ecological determinants (RQ3 and RQ4), ICSN was a significant
predictor of skills & abilities (b = .10, p < .05), and behavior (b = .27, p < .001). Participants who
lived in a community with more hotspots (public spaces to gather and socialize) had higher
efficacy (b = .08, p < .05) and skills to call the hotline number (b = .38, p < .001). Participants’
level of comfort in discussing about human trafficking in those hotspots was also predictive of
more positive attitudes (b = .01, p < .01), norms to migrate (b = .40, p < .001), and less abilities
to call the hotline (b = -.20, p < .01). Interpersonal discussion about human trafficking was also
shown to be one of the strongest facilitators of trafficking prevention. Participants who had more
71
discussion with their social network about human trafficking had more knowledge (b = .35, p <
.01), favorable attitudes (b = .05, p < .05), more perceived risks (b = .44, p < .05), a higher sense
of efficacy (b = .09, p < .05), more skills to use hotline (b = .18, p < .01), and the intention to
practice safe migration (b = .11, p < .01). The size of their migrant network was also beneficial
in preventing human trafficking as it helped to increase knowledge (b = .43, p < .001), positive
attitudes (b = .03, p < .05), perceived risks (b = .33, p < .01), raise personal efficacy (b = .09, p <
.001), and norms to migrate (b = .14, p < .01). Media exposure to human trafficking was found to
be a significant predictor for only knowledge (b = .84, p < .01), and attitudes (b = .16, p < .01).
72
Table 3.3 Unstandardized OLS regression coefficients (standard errors) predicting knowledge, attitudes, norms, efficacy, perceived risks, skills & abilities,
intention, and behavior
Behavior Intention
Skills &
Abilities
Perceived
Norms
Efficacy
Perceived
Risks
Attitudes Knowledge
Background Influences
Female -.18 (.22) .00 (.06) -.06 (.12) -.04 (.13) .00 (.06) -.40 (.36) .05 (.05) .08 (.24)
Age -.03* (.02) -.02*** (.01) .03*** (.01) .01 (.01) .00 (.01) -.03 (.03) -.00 (.00) -.01 (.02)
Education .01 (.04) .00 (.01) .00 (.02) -.03 (.02) .01 (.01) .09 (.07) .01 (.01) .01 (.05)
Annual household income .00 (.00) .00** (.00) .00** (.00) .00 (.00) .00 (.00) .00 (.00) .00 (.00) .00* (.00)
Experience with migration .14 (.27) .28*** (.09) -.51*** (.17) -.38** (.18) .09 (.09) 1.09** (.50) .11 (.07) .53 (.33)
Experience with trafficking .12 (.09) -.08** (.04) .07 (.07) .07 (.08) -.08** (.04) -.22 (.22) -.06* (.03) -.12 (.15)
Environmental constraints -.11 (.19) .26**** (.05) .26*** (.09) .79**** (.10) .37**** (.05) 1.02**** (.28) .11*** (.04) .39** (.19)
Communication Ecology
ICSN .27**** (.05) .01 (.02) .10** (.04) .05 (.04) .02 (.02) .09 (.12) -.01 (.02) -.15* (.08)
Communication hotspots .01 (.15) .04 (.04) .38**** (.08) -.16* (.08) .08** (.04) -.02 (.23) -.06** (.03) -.10 (.15)
Comfort zones -.05 (.14) .04 (.04) -.20*** (.07) .40**** (.08) -.01 (.04) -.31 (.22) .10*** (.03) -.23 (.14)
Interpersonal discussion about
trafficking
-.08 (.10) .11*** (.03) .18*** (.07) -.10 (.07) .09** (.03) .44** (.20) .05** (.03) .35*** (.13)
Migrant network size .02 (.06) .02 (.02) -.01 (.04) .14*** (.04) .09**** (.02) .33*** (.13) .03** (.02) .43**** (.08)
Media exposure to trafficking -.11 (.19) .13* (.07) .12 (.13) .25* (.14) .01 (.07) .21 (.38) .16*** (.05) .84*** (.25)
Note: OLS = ordinary least squares
*p < .10, **p<.05, ***p<.01, ****p<.001
73
Discussion
The results of this study indicated that there were significant effects of the MTV EXIT
documentary on the three outcome variables: knowledge, perceived risks, skills & abilities, and
two limited effects on efficacy and intention (see Figure 3.3). The documentary effects were
mostly diminished by the 4-month follow-up (posttest 2) as only knowledge and intention
variables could sustain the higher scores at a significant level. No effect was found for the
behavioral outcome.
Figure 3.3 Longitudinal mean scores of human trafficking prevention outcomes for the
exposed group
These results are not surprising as researchers have found that exposure to a mass media
campaign alone, in this case the MTV EXIT documentary, would have little impact in
influencing a person to adopt the recommended behavior (McDivitt et al., 1997). Mass media
often plays the role of increasing awareness and knowledge but has limited persuasive ability to
affect attitudes related to a new behavior. The behavioral change models that have been utilized
to create health campaigns often recommend the use of mass media to increase awareness and
knowledge, but to rely on interpersonal channels such as peer education, community organizing,
and social support as additional conditions to achieve behavioral change (Chaffee, 1982). This
74
prior research is consistent with the results from our study, which demonstrates that interpersonal
discussion, migrant network size, integrated connectedness to the storytelling network, and
communication hotspots are all facilitators of human trafficking prevention outcomes.
It should be noted, however, that the MTV EXIT Campaign in Indonesia included a
variety of communication activities beyond this single documentary. Future studies could utilize
a similar research design but include more treatment groups (e.g., participants who are exposed
to the documentary, attended the community awareness events, and participated in youth
activism training, etc.) to examine the effects of each treatment on the different target
populations. There is a growing body of research on strategic communication that argues each
stakeholder group needs separate message strategies that are coordinated and phased according
to an overarching strategic plan (Riley et al., 2014).
The limited effects of the documentary could also stem from its original development in
2012, which did not benefit from more rigorous message design from new approaches of
“edutainment” and the latest behavioral change theories such as Integrative Model used in this
study. The documentary could emphasize its messages on the factors that scored lower at the
baseline (e.g., attitudes, norms, efficacy, skills & abilities) to increase the likelihood that the
target audience would improve those scores and subsequently have intention and adopt the
prevention behavior. Since a portion of our participants did not watch the documentary in its
entirety, the current stories could be edited for brevity so that they are more compelling to a
younger audience that has grown up on action movies and video games and has a shorter
attention span.
For background determinants of human trafficking prevention outcomes, our study found
demographic variables such as sex, age, household income, and level of education to be weak
75
predictors. On the other hand, prior experience with trafficking and migration were stronger
predictors. It is worrisome to see participants with trafficking experience to have less favorable
attitudes, efficacy, and intention to perform preventive behavior. This is consistent with previous
research that finds trafficked victims to be more at-risk to be re-trafficked if they were not given
assistance beyond two years after being rescued from a trafficking situation (Jobe, 2010). The
campaign should make careful efforts in identifying and addressing this at-risk population of the
available resources and support they could access to prevent themselves from being re-trafficked.
Nevertheless, experience with migration was found to be a facilitator of human trafficking
prevention, indicating experienced migrants had more intention and mindset to take
precautionary measures to prevent themselves from becoming the victims of trafficking. Anti-
trafficking campaigns could utilize this protective factor by focusing its message around reaching
out to trusted peers and family members who have migrated safely for advice on trafficking
prevention.
Environmental constraints were also a strong determinant of human trafficking
prevention outcomes. Participants who had a better understanding of environmental constraints
(such as having debts to pay off, lack of job opportunities and unemployment) that push people
to migrate out of Indramayu for work had more knowledge, attitudes, perceived risks, efficacy,
perceived norms, skills & abilities and intention to practice prevention behavior. These
participants seem to be more informed and resourceful in understanding the push and pull factors
of migration, and the dangers associated with it. Environmental constraints factor could be used
as an audience segmentation strategy and more comprehensive messaging could be tailored for
participants who score less on this factor.
76
Most of communication ecological factors were found to be stronger determinants than
background influences. For example, ICSN is the only significant determinant of human
trafficking prevention behavior in this study. Individuals with stronger ICSN practiced more safe
migration and anti-trafficking actions because of their connection with communication resources
in their neighborhood, including family and neighbors, community-based organizations, and
local media. Since interpersonal discussion about human trafficking and migrant network size
are generally good predictors of anti-trafficking related outcomes, prevention campaigns should
focus on utilizing existing migrant networks to encourage peer and familial discussions about the
dangers of human trafficking and encourage safe migration steps to minimize future risk.
Limitations and Future Direction
Limitations in this study may help to explain some of the findings. This survey took up to
45 minutes to complete and respondent fatigue could impact data quality. In addition, high
response drop-out rates, as well as changing the survey format from face-to-face interviews to
telephone interviews at posttest 2 may all contribute to the reliability and validity of the survey
data.
This study provides evidence that the MTV EXIT documentary had positive effects on
relevant prevention knowledge, perceived risks, skills & abilities, and limited effects on efficacy
and intention to perform prevention behavior. To the author’s knowledge, this research project is
one of the first evaluation studies of anti-trafficking campaigns that used a randomized
experimental control trial design with longitudinal panel samples. We also identify prior
experience with human trafficking and migration, and communication ecological factors to be
strong determinants of trafficking prevention-related outcomes. This study supports the inclusion
of mass media campaigns as primary prevention strategies, and to utilize components of the
77
Multilevel Communication Model to identify determinants of prevention outcomes that can be
highlighted and incorporated into the campaign messages. Future studies should include more
intervention formats and approaches (e.g., events, peer education, and trainings) that will add to
our understanding of anti-trafficking campaign effects. Rigorous research designs should test the
efficacy of each component of a campaign across time with panel samples and control
conditions. Other message development approaches such as the use of narrative and persuasion
theories to better engage the target audience could be particularly valuable for practitioners to
plan more effective interventions to increase knowledge, attitudes, and behavioral outcomes of
human trafficking prevention. In addition, a follow-up study could utilize structural equation
modeling as a statistical analysis method to test the significance of relationships and pathways
among the variables in the Multilevel Communication Model and find the overall fit of the
model in predicting prevention behaviors.
78
CHAPTER 4: THE NARRATIVE EFFECTS AND
THE ROLE OF TRANSPORTATION AND IDENTIFICATION
IN CHANGING HUMAN TRAFFICKING PREVENTION OUTCOMES
Introduction
This chapter continues to explore the third step of the public health approach in human
trafficking prevention: designing and testing prevention strategies (Alpert & Chin, 2017). In
particular, this chapter focuses on exploring the role of narrative mechanisms as persuasion
strategies in changing the target audience’s knowledge, attitudes, perceived norms, intention, and
behaviors related to human trafficking prevention.
The power of narrative or storytelling has been documented and used by people for
thousands of years (Fisher, 1985, 1987). Narrative is often employed in entertainment education,
the process of designing and implementing a media message that both entertains and educates in
order to increase audience members’ knowledge about a particular issue, create desirable
attitudes, shift social norms, and change behavior (Singhal & Rogers, 1999, 2002). Anti-
trafficking organizations like MTV EXIT has utilized narrative in television programs such as
documentaries, public service announcements, and music videos to educate and entertain target
audience. So far, over $108 million have been invested for the MTV EXIT campaign to reach
millions of households in Asia and the Pacific (USAID, 2012). Research scholars have claimed
that engaging stories may decrease resistance, facilitate smooth processing of new and/or
difficult information, generate cognitive and emotional effects that influence favorable attitudes
and intentions, and encourage social discussion and role models for behavior change (Green,
2006; Kreuter et al., 2007).
79
Recent studies on narratives in health communication have paved the way to establish the
evidence that narrative communication in various media formats have significant, and varying
effects on the health-related outcomes, depending on the participants’ gender, prior experience,
and individual perception of the underlying mechanisms of the narratives. In one study that
focused on risky sexual behavior among university students, Moyer-Guse and Nabi (2011) found
that the effects of the narrative (video clips from the fictional television show the OC) varied
depending on gender and prior experience, with narratives generating the largest effects on
women with less direct sexual experience. In another study by Niederdeppe, Shapiro, and
Porticella (2011), the results found that the narrative condition (print material containing the
personal story of a patient) increased the belief that obesity can be partly attributed to societal
and environmental factors which are barriers to diet and exercise. A study by Murphy and her
colleagues (2013) on using narratives in cervical cancer-related communication discovered that
the cervical cancer-related film in a narrative format was more effective than the non-narrative
format in increasing knowledge and attitudes toward cervical cancer information, treatment, and
prevention. Nevertheless, there is a lack of studies to demonstrate an empirical evidence of the
effectiveness of anti-trafficking narratives for the target population.
Given the community of researchers who are advocating for the use of narrative forms of
communication in health promotion (Green, 2006; Hinyard & Kreuter, 2007; Kreuter et al.,
2007; Murphy et al., 2013), this study aims to add on to a growing body of evidence-based
research on the relative efficacy of narratives in health communication in the context of human
trafficking prevention. In addition, this chapter explores the theoretical mechanisms that underlie
narrative persuasion so we can inform researchers and practitioners to focus on the appropriate
mechanisms to increase the likelihood of program success. Our first hypothesis was that
80
compared to a control group, participants who receive the human trafficking prevention
information in a non-fictional narrative format will have…
H1a: increased knowledge of human trafficking risk and prevention;
H1b: more positive attitudes toward human trafficking prevention behaviors;
H1c: less favorable norm about migration;
H1d: increased behavioral intention to practice safe migration and human trafficking
prevention
H1e: increased behavior to seek human trafficking prevention information and call the
hotline
We also aim to explore if…
RQ1: there are different narrative effects between male and female participants on
knowledge, attitudes, norm about migration, intention, and behaviors.
Identification
According to a social cognitive theory, human beings do not only learn from direct
experience but also from observing others and modeling their behaviors (Bandura, 2004).
Moreover, people are more likely to adopt behaviors that are shown by people who are similar to
themselves (Bandura, 2002). Previous studies have shown that viewers who identified with
narrative characters of similar physical appearances such as race/ethnicity, biological sex and age
were more likely to be engaged in the story, leading to changes in belief, attitudes and/or
intention about the health issues depicted in the story. In Murphy et al.’s (2013) study of the
narrative effects of cervical cancer-related film that featured a cast of Latina actresses, Mexican
American participants had greater identification with the main Latina characters, and were more
transported into the story than European American participants. The study also found Mexican
81
Americans to have greater changes in knowledge, favorable attitudes, and intention to get a pap
test than European Americans. In another study by McQueen et al.’s (2011), African American
women who were exposed to a film with stories of African American breast cancer patients had
higher identification, affect, and engagement, and subsequently lower perceived barriers and
fatalistic beliefs about cancer. This study defines identification as perceived similarity in life
circumstances and situations with the main characters in the documentary film.
Transportation
One of the most cited theories of narrative engagement is created by Green and Brock’s
(2000) who term it as a transportation theory, a cognitive state in which viewers become highly
drawn and therefore ‘transported’ into a story. Green, Brock, and Kaufman (2004) state that
transportation into a narrative world is, ‘‘the process of temporarily leaving one’s reality behind
and emerging from the experience somehow different from the person one was before entering
the milieu of the narrative’’ (p. 315). Transportation was found to correlate with enjoyment of a
narrative and accepting the beliefs depicted in the narrative (Green & Brock, 2000; Green et al.,
2004). Another recent study discovered that greater transportation into the cancer storyline of a
Desperate Housewives program correlated with positive changes in knowledge, attitudes, and
behavior related to cancer and transportation of a cervical cancer-related film to be predictive of
knowledge about cervical cancer information, treatment, and prevention (Murphy et al., 2011).
This evidence is similar to Slater and Rouner’s (2002) extended elaboration likelihood model
(EELM), which claims that engaging in the narrative may stimulate deeper processing of a
particular message, leading the audience to adopt recommended attitudes and behaviors that are
shown in the message. Transportation into the story and identification with characters are two
important narrative constructs that can be measured to trigger responses that are indicative of the
82
goal of the intervention (Slater & Rouner, 2002). For this study, Green and Brock’s (2000)
transportation will be used to measure participants’ level of involvement into a narrative film.
Differences between men and women in processing narratives
The previous sections of identification and transportation discuss several studies that
demonstrate different effects of narrative-based interventions on different target populations
(e.g., female minorities). This research will also explore the role of sex in moderating the effects
of human trafficking prevention narratives among the target population. Since our narrative
stimuli consists of human trafficking stories of one male and two female victims, the next
hypotheses were…
H2a: female participants will be more transported into the documentary;
H2b: female participants will identify more strongly with the two female characters (Siti
and Ika);
H2c: the male participants will identify more with the male character (Ismail).
Finally, to establish empirical evidence of the relationships between transportation and
identification with the changes in knowledge, attitudes, norms, intention and behavior related to
human trafficking prevention, this study will explore…
RQ2: which transportation and identification constructs will produce an increase in
human trafficking prevention-related (a) knowledge, (b) attitudes, (c) norms, (d) behavioral
intention, and (e) behaviors.
RQ3: if the participant’s sex will have significant interactive effects with transportation
and identification on human trafficking prevention-related (a) knowledge, (b) attitudes, (c)
norms, (d) behavioral intention, and (e) behaviors.
83
Methods
This study is a mixed-method explanatory sequential design by using quantitative surveys
in a pretest-posttest control group experiment to capture the changes in human trafficking
prevention outcomes among men and women before and after watching a documentary, and
assess transportation and identification with characters in the documentary. Focus groups were
conducted as a qualitative follow-up study to help interpret the survey results (Creswell & Plano
Clark, 2011). Figure 4.1 describes the flowchart of the participants in this study. This mixed-
method design is appropriate to explain audience’s mixed reception to a complex narrative like
human trafficking, which may not be sufficiently analyzed and contextualized by only
quantitative data. The quantitative survey is a principal method while focus groups data is used
as a supporting component. This methodological triangulation or the use of multiple methods to
study a single topic provided this project with rich data that had different layers of depth and
perspectives (Denzin, 1978).
Figure 4.1 Flowchart of the participants in the mixed-method explanatory sequential design
84
Stimuli Material
The narrative material used in this study was a 24-minute documentary film titled
Enslaved: An MTV EXIT Special, hosted by Dian Sastrowardoyo, a popular actress in
Indonesia. The Bahasa Indonesian film was produced in 2012 by MTV EXIT and featured the
stories of three human trafficking victims; Siti, a woman who migrated to Malaysia and ended up
as a domestic worker in exploitative working conditions; Ismail, a man who was trapped in
bonded labor in the logging industry in Northern Sumatra; and Ika, a teenage girl who was
deceived into forced prostitution in Batam. The documentary started out with the three victims
narrating their poor living conditions, and aspirations to seek a better life. The story moved on to
document the process by which they were trafficked into exploitative work, and how they
eventually escaped from their situations. The film concluded with prevention messages that
advised the viewer to be skeptical of recruitment agents, to seek information and verify job
opportunities through someone they can trust, such as government agencies and NGOs, before
accepting the offer. They also suggested that anyone considering migration to have complete
possession of legal documents and identification before traveling abroad, to call the hotline
numbers if they need help or report suspicious activities, and to raise awareness of the issue by
sharing information with others. To ensure that our research participants have viewed the
narrative program, we asked the participants to watch the documentary in its entirety and only
included those who viewed at least 18 minutes into the film.
Experimental Procedures
The data from this study is drawn from the original survey administered in Indramayu,
Indonesia. The survey data was collected from a pretest-posttest control group design. Surveys
were implemented using multi-stage sampling from the regency level, to districts,
85
neighborhoods, and households. A random sample of 6 out of the 18 districts were chosen as the
study sites. Researchers visited 90 neighborhoods in the selected districts and conducted face-to-
face interviews with 527 households. A Kish Grid was used for each household where more than
one person was eligible to participate to make sure there was no systematic selection bias of the
respondents. To be eligible to participate, respondents had to be 18-39 years old, live in
Indramayu, completed at least elementary school, be able to read and write Bahasa Indonesian,
did not know anything about the MTV EXIT campaign, and, for the narrative condition group,
had the means to view the MTV EXIT VCD. In the end, 527 people agreed to participate in the
study and they were randomly assigned to a narrative condition (n = 319) or a control condition
(n = 208). Institutional review boards at the investigator’s university approved the study.
The survey data used in this study were collected in two waves from April to May 2014
through face-to-face interviews. At the pretest (T1), participants in the narrative condition were
given a VCD with the documentary Enslaved: an MTV EXIT Special after the interview was
completed. These participants were instructed to watch the documentary in its entirety for a
posttest (T2) interview about one week later. There were no drop outs at the posttest and all
participants were given an MTV EXIT-branded pen to thank them for their time.
Survey Participants
47% of the participants were men, and 34% were 18-24 years of age, followed by 29%
who were 25-30 years old, 19% aged 31-35, and 18% from the age range of 36-39. A majority of
the participants had annual household incomes between $1,241 and $2,482 (46%), and 34% had
less than $1,241 while only 4.4% had over $3,700. About 65% were married and 32% were
single. For their level of education, 40% reported that they had completed SD (elementary
school), 30% completed SMP (Grade 9), and 27% graduated from SMA (Grade 12). To ensure
86
similar distribution in participant demographic characteristics between the control and the
exposed groups, crosstab analyses of sex, age, annual household income, marital status and
education were conducted. There were no significant differences in participant demographic
characteristics between the control and the exposed groups.
Measures
For easier comparison, all of the five dependent variables (human trafficking prevention
outcomes) were scaled to a score of 0-10. These five variables were collected at pretest and
posttest in both narrative and control conditions. Independent variables (transportation and
identification) were collected only at the posttest with the participants in the narrative condition.
Dependent Variables
Knowledge about human trafficking. Thirteen items were adapted from the Trafficking
Awareness Survey (Minnesota Advocates for Human Rights, 2003) and augmented by items
developed during our survey construction process. Participants chose either Yes or No if they
have heard of terms such as “human trafficking,” and “safe migration,” as well as statements
such as “A worker in trouble/human trafficking victim could be….people who left or were taken
away from their country or city and tricked or forced to do a job in which they were exploited,”
and “…someone who is forced to work longer hours than were written in the contract or
promised.” Each “Yes” answer was given one point and summed to form a composite score out
of 13, and then scaled to out of 10. Higher scores showed greater knowledge about human
trafficking.
Attitudes regarding human trafficking prevention behavior. Participants were asked to
rate their level of agreement with five statements (on a scale from 1 = Strongly Disagree to 5 =
Strongly Agree): “finding information about safe migration is…easy,” and “…important,” and “a
87
hotline would be…useful in emergency,” “…helpful in providing information,” and “…useful in
emergency.” The Cronbach α ranges from .65 to .77 and all the items were averaged to form an
index.
Normative pressure to migrate. Participants were asked to rate their level of agreement
(scale from 1 = Strongly Disagree to 5 = Strongly Agree) with two statements: if they had an 18-
year old daughter, they would ask her to move away from home for work, and if they had an 18-
year old son, they would ask him to move away from home for work. The two items were
averaged to form an index and its Cronbach’s α is from .84 to .89.
Intention to practice human trafficking prevention behavior. Participants rated 6
possibilities to answer the question, “If you were to migrate for work in the next few years, how
likely would you be to do any of the following?” using a 5-point Likert scale anchored at 1 =
Very unlikely and 5 = Very likely. Two examples of the possibilities are: “become a member of
an NGO or migrants organization to receive advice and assistance on migration,” and “ask
friends and family for advice.” The items were aggregated to form a composite index. The
Cronbach’s α ranges from .83 to .90.
Human Trafficking Prevention Behavior. For this outcome variable, participants were
asked if they have ever tried to seek information to migrate safely for work for either: (1)
themselves (2) family members, and/or (3) their friends. A follow-up question of how they find
safe migration information such as through “family member,” “a sponsor,” “newspaper,” and
“government agency (BP3TKI or Disnakertrans).” Each mentioned answer was accorded 1 point.
Respondents also answered whether or not they had called a counter-trafficking hotline number.
Yes was coded as 1 and No coded as 0. The scores were then aggregated to form a composite
index for this variable.
88
Independent Variables
Transportation. This variable was measured in the follow-up survey using 10 of 11 items
from Green and Brock’s (2000) scale. Principle components factor analysis using varimax
rotation was run on 10 items and the final scale used in this study included five items that loaded
on a single factor. Examples of the items include “The events in the documentary are relevant to
my everyday life,” and “I wanted to learn what eventually happened to the people in the
documentary.” The items were measured by a 5-point Likert scale from 1 = Strongly disagree to
5 = Strongly agree, and the mean of the 5 items were calculated into the composite index (α =
.731).
Identification. Participants in the narrative condition were asked the extent to which they
identified with each of the three main characters (Siti, Ismail, and Ika). Two components of
identification were assessed on 5-point scales, “How similar do you think your life circumstances
are to the following people in the documentary?” (1=not similar at all to 5=very similar), and
“Do you think you could end up in the same situation as the following people in the
documentary?” (1=Very unlikely to 5=Very Likely). The reliabilities of identification with Siti,
Ismail, and Ika were Cronbach’s α of .71, .69, and .64, respectively.
Statistical Analysis
The analytic strategy included analysis of variance for H1, independent samples t-tests
between narrative and control conditions (RQ1), and between male and female for (RQ1 and
H2). RQ2 and RQ3 were analyzed using multiple linear regression analysis. For RQ3, four
additional sex interaction variables (Transportation X Female, Siti X Female, Ismail X Female,
89
and Ika X Female) were added into the multiple linear regression model. The alpha level was set
at 0.05 a priori.
Focus Group Discussion Study
Focus group, a carefully planned, interactive interview among a group of people
(normally 4-7) who participate in a guided discussion of a topic of interest, is one of the most
widely used qualitative tools in the field of social sciences (Stewart & Shamdasani, 1990).
Unlike face-to-face interviews, which rely on individual recollection of past experiences that
could be fallible to memory, focus group generates shared, magnified discussion on a topic that
triggers immediate evidence from a variety of similar or contrasting opinions and experiences
(Morgan, 1997). The direct evidence from the interactive group discussion can provide more
accurate, explanatory conclusions on the participants’ reception to the documentary than post
analyses of the survey data. Through surveys and focus groups, this study capitalized their
strengths and minimizes their weaknesses, and therefore should generate rich, insightful data
with varied depth and valiant consensus.
As I do not speak Bahasa Indonesian language, I contacted a field manager from a
national migrant organization in Indonesia to help me in forming a small team of three
researchers to recruit participants, organize, and moderate the focus group discussion. The field
manager has known me from the time I was a regional campaign coordinator for the MTV EXIT
Foundation in 2009. In addition to agreeing to share research data and final report, I conducted a
3-day workshop on planning and implementing a focus group for interested employees of this
migrant organization. Focus group discussions (FGDs) were held with 2 groups of women and 2
groups of men (a total of 4 groups with 8 participants in each FGD) in Indramayu in June 2017
to explore reasoning behind some of the survey results. Recruitment criteria included people who
90
had participated in the previous surveys and had migrated for work or knew a family member
who had migrated. The focus group was set up at an open lounge area in front of the meeting
room on the first floor of a local migrant organization building. Each FGD lasted for about two
hours. Participants were briefed about the purpose of the study and all agreed to sign consent
form. The entire session was audio recorded on a digital player. FGD moderator asked the
participants to watch the documentary before the discussion. They were then asked about their
reaction toward the stories of each of the three trafficked victims, how they identified with any of
the characters, how they feel about the prevention messages in the documentary, especially in
comparison to their own migration experiences.
The FGD moderator received 5 hours of training from me on how to moderate the group
discussion, which started out with an open-ended question that ask them to describe their
reaction toward the documentary. The FGD moderator made significant efforts to elicit
responses from every participant by encouraging less active members to voice their opinions. As
a non-Indonesian speaker, I had a translator sat next to me to inform me what conversation was
going on during each FGD. I would then ask a translator to raise additional questions to the
participants where necessary. I made reflective memos to describe the FGD settings, non-verbal
reactions, valiant narratives, contradictions, limitations (e.g., at one point, the moderator was
dominating the group discussion by informing his participants about the importance of this
documentary, potentially misleading them to agree with the anti-trafficking message). These
memos helped me to reflect significant interruptions that may obfuscate my interpretations of the
data and enhance my self-reflexivity during and after meeting with the research participants.
After four FGDs were complete, I sent recorded audio files to a local translator to transcribe and
translate the file into a written document for me to analyze in this study.
91
Results
The first hypothesis predicted the documentary film effects for knowledge, attitudes,
norms, intention and behavior. This hypothesis was tested using a 2 (film: narrative or control) ×
2 (sex: male or female) analysis of variance. Means and standard deviations are shown in Table
4.1. There was a main effect of the documentary film on change in knowledge, F(1, 523) = 9.35,
p < .01, such that viewers of the narrative (M = 1.56, SD = 2.29) learned significantly more
information about human trafficking than those in the control condition (M = 0.94, SD = 2.16).
Similarly, change in intention toward practicing safe migration and human trafficking prevention
varied significantly between the narrative and control conditions, F(1, 523) = 4.90, p < .05.
Participants who were exposed to the narrative film showed higher change in intention (M =
0.92, SD = 1.96) than those in the control condition (M = 0.58, SD = 1.56). Narrative effects
were not found in attitudes, norms, and behavior related to human trafficking prevention.
Table 4.1 Means and standard deviations of change in knowledge, attitudes, norms,
intention and behavior by sex and experimental condition (narrative or control)
Male Female Total
(N=250) (N=277) (N=527)
Change in knowledge
Narrative
b
1.43 (2.39) 1.68
b
(2.20) 1.56
b
(2.29)
Control 1.02 (1.98) 0.85 (2.32) 0.94 (2.16)
Change in attitudes
Narrative 0.49 (0.94) 0.37 (1.23) 0.43 (1.11)
Control 0.30 (0.88) 0.51 (1.04) 0.41 (0.97)
Change in norms
Narrative* -0.33* (1.67) 0.08* (1.83) -0.11 (1.77)
Control -0.15 (1.18) 0.08 (1.39) -0.03 (1.30)
Change in intention
Narrative
a
1.00
b
(1.69) 0.86 (2.18) 0.92
a
(1.96)
Control* 0.31* (1.37) 0.83* (1.69) 0.58 (1.56)
Change in behavior
Narrative -0.15 (2.02) -0.17 (2.34) -0.16 (2.20)
Control -0.09 (1.96) -0.08 (2.38) -0.09 (2.18)
*p < .05, there's a significant difference between male and female in the same variable
92
a
p < .05,
b
p < .01, the value in the narrative condition is significantly different from the
same variable in the control condition
Note: All the variable scores could range from −10 to 10.
For the first research question that explore if sex interacted with any of the five outcome
variables, we found that women had slightly increased favorable norm about migration (M =
0.08, SD = 1.83) while men had less favorable norm about migration in the narrative condition
(M = -0.33, SD = 1.67), t(317) = -2.11, p < .05. It was also interesting to note that there was a
significant sex interaction with the change in intention only in the control group: women had
higher change in intention (M = 0.83, SD = 1.69) than men (M = 0.31, SD = 1.37), t(206) = -
2.44, p < .05. No significant sex differences were found for change in knowledge, attitudes, and
behavior in both conditions.
The second hypotheses predicted a sex difference in transportation, and identification
with each of the three characters in response to the narrative film (see Table 4.2). Contrary to our
hypotheses, women were not transported into the documentary (M = 3.21, SD = 0.63) any more
than men (M = 3.28, SD = 0.71). Nevertheless, for identification with characters, we found that
men (M = 2.22, SD = 0.81) had higher identification with Ismail (the male trafficked victim)
than women (M = 1.80, SD = 0.68). However, no significant differences were found between
men and women for identification with Siti and Ika, the two female trafficked victims. It should
also be noted that participants generally did not find much identification with the three characters
as all of the mean scores were less than 2.5 out 5.0.
93
Table 4.2 Means and standard deviations of transportation and identification with
characters of the documentary by sex
Male Female Total
(N=150) (N=169) (N=319)
Transportation 3.28 (0.71) 3.21 (0.63) 3.24 (0.67)
Identification
with Siti 1.96 (0.77) 2.07 (0.82) 2.01 (0.80)
with Ismail 2.22* (0.81) 1.80* (0.68) 2.00 (0.77)
with Ika 1.91 (0.76) 1.94 (0.76) 1.92 (0.76)
Note: Scores could range from 1 to 5. Standard deviations are in parentheses. Means
between male and female in the same row with * differs at p < .05 by the independent
samples t-test method.
Finally, RQ2 and RQ3 explored how the theoretical constructs of transportation and
identification with Siti, Ismail and Ika characters and their sex interaction variables were related
to the dependent variables of knowledge, attitudes, norms, intention, and behaviors. A variety of
demographic and migration variables (including sex, age, education, household income,
migration experience) were entered into the regression analyses as control variables because they
might relate to human trafficking prevention-related outcomes. Most control variables did not
have a significant relationship with any of the dependent variables, nor did they shift the results
pattern. Table 4.3 reports the regression analyses without these variables included.
Table 4.3 Effects of transportation and identification with characters on knowledge, attitudes, norms,
intention, and behavior by sex (n=319)
Knowledge Attitudes Norms Intention Behavior
Pretest level 0.42*** 0.10* 0.60*** 0.26*** 0.45***
Transportation 0.13 0.44*** -0.29** 0.26** 0.08
Transportation x Female -0.39 -0.11 0.42 0.06 -0.43
Identification
with Siti 0.86*** 0.51* 0.50* 0.23 -0.38
with Siti x Female -1.58*** -0.91* 1.00** 0.35 0.81*
with Ismail -0.36*** 0.09 -0.05 0.01 -0.01
with Ismail x Female .90*** -0.20 -0.22 -0.26 0.22
with Ika -0.32 -0.66*** 0.33 0.10 0.36*
with Ika x Female 0.38 1.01** -0.36 0.09 -0.75*
Adjusted R
2
0.28 0.40 0.41 0.41 0.39
*p < .05, **p < .01, ***p < .001
Note: Standardized beta coefficients from regression models.
94
The posttest levels of knowledge, attitudes, norms, intention and behaviors were
regressed on their pretest variables, transportation, identification with each character, and the
four sex interaction variables (Transportation X Female, Siti X Female, Ismail X Female, and Ika
X Female). This regression explained 28% of the variance in knowledge at posttest, F(19, 299) =
7.54, p < .001. Transportation was not a significant predictor of knowledge (β = 0.13, p = .18).
Women who were more transported by the film did not have greater level of knowledge than
men (β = -0.39, p = .23). In addition, people who identified more with Siti had significantly
higher level of knowledge (β = 0.86, p < .001) but those who identified more with Ismail had
significantly lower level of knowledge (β = -0.36, p <.001). Interestingly, sex has a significant
reverse interaction effect on identification with Siti and Ismail. Women who identified more with
Siti (the Indonesian maid who was abused by her employers) had less knowledge (β = -1.58, p <
.001) when compared to men. On the other hand, women who identified more with Ismail (the
male trafficked victim) had significantly more knowledge (β = 0.90, p < .001) than their male
counterparts. Identification with Ika did not result in significant effect on knowledge.
The regression explained 40% of the variance in posttest attitudes, F(19, 299)=12.1, p <
.001. People who were more transported had higher positive attitudes toward human trafficking
prevention behaviors (β = 0.44, p < .001). Identification with Siti are associated with
significantly positive attitudes (β = 0.51, p < .05). However, women who identified more with
Siti had less favorable attitudes (β = -0.91, p < .05) than men. Identification with Ika (a female
sex trafficking victim) was found to be associated with significantly less favorable attitudes (β =
-0.66, p < .001). On the contrary, women who identified more with Ika had higher level of
favorable attitudes (β=1.01, p<.01) than men.
95
For norms related to migration, the regression model explained 41% of the variance in
posttest, F(19, 299)=12.38, p < .001. People with higher level of transportation were found to
have less favorable norm in moving away from home to work (β=-0.29, p < .01). We also found
that identification with Siti had the effect of increasing the norm to migrate (β = 0.50, p < .05).
Women who identified more with Siti had even higher level of norm to migrate (β = 1.00, p <
.01).
The regression explained 41% of the variance in posttest behavioral intention, F(19,
299)=12.51 p < .001, but only transportation was found to have a significant association with
intention (β = 0.26, p < .01).
For human trafficking prevention-related behaviors, the regression model explained
39% of the variance in posttest, F(19, 299)=11.87, p < .001. Women who identified with Siti
were found to practice more anti-trafficking behaviors (β = 0.81, p < .05) than men.
Identification with Ika was found to be significantly associated with more behaviors (β = 0.36, p
< .05). However, there was a sex interaction which reverse the effect as women who identified
with Ika had significantly less human trafficking prevention behaviors (β = -0.75, p < .05).
Discussion
The main goal of this chapter was to empirically test whether a narrative film produces an
impact on human trafficking prevention-related knowledge, attitudes, norms, intention and
behaviors. We were also interested to explore if the narrative film would affect men and women
differently. 527 Indonesian men and women participated in this study and they were assigned to
either a narrative condition (n=319) or a control condition (n=208). Quantitative data from this
study came from the pretest and posttest surveys, taken one week after narrative participants
were instructed to watch the documentary Enslaved.
96
A series of 2 (narrative vs. control) × 2 (sex) analyses of variance revealed that the
documentary narrative was effective in increasing human trafficking knowledge and intention to
practice safe migration and human trafficking prevention. However, no significant differences
were found on changes in attitudes toward trafficking prevention behaviors, perceived norms
about migration, and actual anti-trafficking behaviors. Hence, only H1a and H1d are accepted.
While narrative effect was not found on norm to migrate, we found that women had slightly
increased norm to migrate than men who had decreased norm to migrate after viewing the
documentary. This finding suggested that Indonesian women were under more normative
pressure to migrate out of their hometown to seek work in other cities in Indonesia or abroad.
Narrative Mechanisms and Their Influence on Human Trafficking Prevention
Additional analyses were made to explore different effects of narrative mechanisms
among male and female participants using the measures of transportation and identification with
the three main characters in the documentary. Despite more emphasis on the female victims (two
out of three) in the documentary, we did not find any significant difference on the level of
transportation into the documentary among male and female participants. Women did not
necessarily identify more with the two female victims (Siti and Ika) while men identified more
with the male victim (Ismail). This study found relatively weak identification with all of the
characters as the mean scores of all identifications were below 2.5 (out of 5.0). Characters with
the same sex as the audience did not automatically trigger more identification. In this study,
identification was operationalized as the participants’ perceived similarity of life situations and
experiences faced by the characters. To increase identification with the characters, the
documentary should ensure that the stories of the three victims were more similar to the
everyday situation and experience of the viewers. A 25-year-old female participant from the
97
focus group summed up her reaction toward Siti, a domestic worker who was imprisoned and
abused by her employer in Malaysia,
“I could not relate to Siti because we have nothing in common. I am sorry that she lost
both of her parents when she was young, and she has no siblings and is alone in the
world. But why did she migrate to Malaysia and expose herself to that risk to begin with?
She can just stay in Indonesia and work for herself. It is not like she needs to take care of
anyone. Some of us migrate and seek work abroad because we have parents, younger
siblings, or children to take care of.”
Without high level of identification with the main characters, prior research indicates that
participants are less likely to feel the situations depicted in the film and the fate of the characters
are relevant to them (Murphy et al., 2011, 2013). In order to create more identification with the
female audience, the documentary could have chosen a different female character with more
familial ties at home who migrates abroad to earn and remit money back home.
In addition to creating similarity between the narrative characters and the audience,
stories should be perceived as realistic to engage and increase identification. According to
Cho, Shen and Wilson (2014), realistic stories consist of high plausibility (events that could
happen in real life), typicality (stories that are similar to actual experiences), and factuality
(events that actually took place). Another female focus groups participant lamented on the
unrealistic situations of both Siti and Ika that reflected her reasoning for having little
identification with these characters:
“The chance for me to experience what happened to Siti or Ika is very low. I do not know
a single person who has moved away for work who were drugged and raped like these
two women. Their stories were so extreme that it made me question the intention of this
98
documentary. Is it to scare us to not migrate to work and make a better life for our
family?”
Based on the survey results from Chapter 2, most human trafficking victims experienced
labor exploitative conditions such as not being allowed to communicate with loved ones, not
being allowed to keep the money earned, confiscation of identification documents, and being
forced to work excessively long hours without any days off. On the other hand, the MTV EXIT
documentary showcased three extreme trafficking cases where the characters undergo severe,
life-threatening physical, sexual, and mental abuses by traffickers and employers. These three
characters likely did not reflect the common experiences of human trafficking that the
participants or people in their social networks had faced. Therefore, most of the participants had
weak identification with the victim characters. In order to achieve higher identification with the
characters, practitioners should present more realistic vignettes of everyday cases of labor
exploitation (e.g., cases where employers violate labor rights by engaging in bonded labor,
withholding wages, enforcing longer working hours than the contract, and limiting
communication with loved ones) instead of featuring extreme human trafficking cases that may
create more interesting narratives, but generate little identification with the viewers.
Our study found transportation to significantly increase the level of favorable attitudes
toward performing anti-trafficking behaviors, decrease the norm to migrate, and increase the
intention to practice safe migration and human trafficking prevention. However, people who
experienced higher transportation into the documentary did not have higher level of human
trafficking-related knowledge nor take more anti-trafficking actions. Nevertheless, transportation
was a strong predictor for three out of five human trafficking prevention outcomes. This
99
demonstrated that engaging narrative had a better persuasive impact to change viewer’s attitudes,
perceived norms and intention to prevent human trafficking.
We found mixed effects of identification with Siti, Ismail and Ika on human trafficking
prevention outcomes. Women who identified with Siti showed lower level of knowledge, more
negative attitudes toward trafficking prevention, and higher magnitude of the perceived norm to
migrate out of town. This study highlighted that women who perceived themselves to share
similar life experiences and situation as Siti were more vulnerable to being a victim of human
trafficking, given their lack of knowledge, negative attitudes, and strong pressure to migrate out
of where they were. Interestingly, these women also showed more anti-trafficking behaviors.
These contradictory results could be explained from a focus group participant who was a former
female migrant and had attempted to call the hotline to claim lost wages from her employer in
Saudi Arabia:
“I really feel for Siti, the domestic worker in the story. But I disagree with the simple
solutions that they show us. They said call this hotline number if you need help. Last
year, I tried to call this hotline number after I came back from Saudi Arabia and I needed
to report that my employer owed me six months of pay. The hotline staff was so useless
and could not help me at all. He said my problem was not related to human trafficking
and I need to contact somebody else. I feel I was lied to and all of them must be in a
scheme to cheat me.”
This insight reveals the complex reality of anti-trafficking behaviors, defined in this study
as seeking information on safe migration and human trafficking prevention, and calling the
hotline. Women who identified with Siti had the motivation to seek domestic work abroad due to
their perceived norm to migrate. In addition, the lack of human trafficking-related knowledge
100
propelled them to seek more information and call the hotline to report exploitative labor
situations (in this case, withheld payment from the employer). However, they might not have
good experience in using the hotline due to incompetent resources or undesirable encounters
with the hotline staff. This led to negative attitudes toward anti-trafficking behaviors.
The pattern of inverse relationship between attitudes and behaviors could also be found
among female audience who identified with Ika, the 15-year-old female victim of sex trafficking.
From the focus groups, women who mentioned Ika’s story tended to be younger, and expressed
more favorable attitudes toward anti-trafficking behaviors, but less need to perform those
prevention behaviors because they did not plan to migrate out of Indramayu in the future. An 18-
year-old female FGD participant who had never migrated out of Indramayu shared her reaction
from Ika’s story:
“I find Ika’s story the most interesting because I was closer to her age, and like her I
aspire to get out of this rural town and seek a more exciting life in big cities like Jakarta. I
agree with the message that there are bad people out there who want to lure young
women like us by offering us a waitress job in the city, but their real intention is to trick
us into sex work. It is important for me to register myself with the migrant organization
and call the hotline to check if those job opportunities are real. Anyway, I do not see the
need to seek more information or call the hotline right now because I feel I am not old
enough to move into the city on my own. I will probably do it in the future when I plan to
move.”
The focus group data helped to explain the mixed effects of identification with Siti and
Ika on human trafficking prevention outcomes among female participants. These findings display
a cautionary tale for anti-trafficking organizations on the unintended consequence of choosing to
101
communicate extreme human trafficking cases in the campaign narrative. While the viewers may
be transported into the narrative, they would develop weak identification with the characters.
Among those that identified more with the characters, in this case the women who identified with
Siti, they might counter-argue and develop reactance to the message in the narrative based on
their real-life negative experience with the recommended behaviors.
This study raises an ethical issue in designing a human trafficking prevention campaign
with the focus on assigning personal responsibility to avoid being a victim of trafficking rather
than changing structural factors that push people to migrate and improving community resources
that protect and support the victims. The MTV EXIT documentary was developed based on the
assumption that human trafficking could be prevented if at-risk populations were exposed to the
documentary, made aware of the risks and be savvy enough to seek help on their own through
the hotline or related organizations when they found themselves in an exploitative situation. The
focus on personal responsibility overlooks other structural factors that may push someone to
become victims, despite being knowledgeable about the risks. Poverty, lack of job opportunities
in their community, desire to move to urban cities, pressure by parents to migrate and secure jobs
so as to remit money back home are all contributing factors that will continue to push people to
migrate regardless of their awareness of human trafficking. It may also be unethical to promote
the use of hotline if there are no competent personnel to offer the right answers or referrals to the
victims. Since human trafficking is a complex phenomenon that requires ecological perspectives,
anti-trafficking campaigns should consider addressing structural factors at individual,
interpersonal, community, and societal levels. The intervention should address structural
problems such as improving protection services and justice (e.g., competent and high-quality
hotline service), strengthening law enforcement efforts against unscrupulous recruitment agents
102
and exploitative employers, and creating more job opportunities in the source areas to reduce
outgoing migration. Efforts should be made, whenever possible, to include local police force,
immigration office, public schools and local business owners to be involved in the design of the
human trafficking prevention intervention. Partnerships with these organizations could sensitize
law enforcers to understand the injustice and restrictions faced by at-risk populations, encourage
school teachers to institutionalize peer education and anti-trafficking campaigns into the formal
education system and educate employers on what constitutes labor exploitation and how to adopt
best practices against it.
Limitations and Future Direction
The survey data used for this chapter was part of a larger survey that assessed the
prevalence of human trafficking, its determinants, and the effects of the MTV EXIT
documentary on human trafficking prevention-related outcomes other than those measured in
this chapter. Given the extensive length of the survey, there may be survey fatigue that could
affect validity and reliability. More comprehensive measures of identification that
operationalized other dimensions of liking, wishful identification, and parasocial interaction were
not included (Moyer-Guse, 2008; Tal-Or and Cohen, 2010). Emotion, a variable that scholars
have found to be associated with narrative processing and mediate health-related outcomes, was
also not included (DeSteno, Petty, Rucker, Wegener & Braverman, 2004; Dillard & Nabi, 2006;
Nabi, 2002). Future studies for this research could include more identification and emotion
measures to further explore the underlying mechanisms of narrative persuasion.
Given the limited budget of this study, we did not create more film versions with
different characters and storylines to further compare different narrative effects. The conclusion
of this study may claim that the MTV EXIT documentary that utilized narrative approach had
103
effects on increased knowledge and behavioral intention on its viewers. The limited narrative
effects could stem from its original development (produced and released in 2012). The
documentary should include more rigorous message design that focuses on victim cases that are
more identifiable with the target audience. The documentary could also focus more time on
showing the main characters performing human trafficking prevention behaviors. The current
documentary spent only the last six minutes on recommending human trafficking prevention
behaviors with the host informing the viewers on what they should do to stop human trafficking
rather than demonstrating anti-trafficking actions performed on camera by the characters. For
example, the documentary could show potential migrants walking to the nearest migrant
organization to register themselves with the migrant network that could help them migrate safely,
asking government officials in labor agencies to obtain accurate information before making a
decision to migrate, and calling the hotline to verify the authenticity of the job offer abroad.
Researchers should be cautious in extrapolating the results of this study to other human
trafficking prevention campaigns as the narrative effects are heavily dependent upon the specific
types of human trafficking and labor exploitation featured in the film (this study focused on sex
trafficking, domestic worker abuses, and forced labor in the logging industry), geographical
context, and target populations. Greater narrative effects could be measured if the narrative is
tailored for a different group of participants, e.g., only potential migrants who are looking to
migrate abroad, thereby, making the intervention more relevant and applicable to their situations.
104
CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS
The chapter discusses how a multilevel communication approach and findings from this
project can be used as a guidance for researchers, practitioners, and policy makers to plan,
implement and scale up their human trafficking prevention programs.
This dissertation has described an overview of the scope, scale and consequences of the
human trafficking problem in Indonesia. Current knowledge and research gaps within the fields
of communication and human trafficking prevention from the perspectives of critical and
empirical paradigms are also discussed. This project contributes to a small but growing area of
research on prevalence and experiences of human trafficking victims, determinants of human
trafficking prevention, effects of an anti-trafficking campaign, and narrative persuasion
mechanisms that influence knowledge, attitudes, and behaviour related to human trafficking
prevention.
Findings from the prevalence study in chapter two suggest that there were approximately
98,654 victims of human trafficking (15.4% of the 18-39 years old population) in Indramayu,
Indonesia. Most common human trafficking experiences include not being allowed to
communicate with loved ones, not being allowed to keep the money earned, confiscation of
identification documents, and being forced to work without compensation to repay debts or work
excessively long hours without any days off. In addition, women experienced higher rate of
human trafficking than men (17.3% compared to 13.2%), as well as encountered more restriction
on their movements and exploitative labor practices. Migrant women who worked abroad had
less options to escape their situations. Rescued victims also struggled with mental health
problems such as anxiety and depression.
105
The study in chapter three provides evidence that the MTV EXIT documentary had
positive effects on knowledge, perceived risks, skills & abilities, and limited effects on efficacy
and intention to perform prevention behavior. Prior experience with migration, environmental
constraints, and communication ecological factors such as ICSN, interpersonal discussion about
trafficking, and migrant network were found to be strong protective determinants of trafficking
prevention outcomes. On the other hand, prior experience with trafficking is a risk factor for
prevention outcomes. A Multilevel Communication Model could be used as a theoretical
framework for practitioners to identify risk and protective factors among target populations, and
to evaluate the campaign effects on their knowledge, attitudes, efficacy, perceived norms,
perceived risks, skills & abilities, intention, and prevention behavior.
Findings from chapter four demonstrate that transportation and identification with
characters in a non-fictional narrative film play a role in influencing human trafficking
prevention outcomes. Viewers who were more transported into the narrative were more likely to
increase their favorable attitudes and intention to prevent human trafficking as well as feel less
normative pressure to migrate. Meanwhile, both male and female viewers had weak
identification with all of the victim characters in the documentary due to the extreme depiction of
their trafficking situations that were perceived as unrealistic and untypical of the experiences that
the audience or other migrants that they knew had. In particular, women who identified more
with Siti, the domestic worker character, had higher norms to migrate and less knowledge,
attitudes, and preventive behaviors. These unintended results could be explained by the female
audience’s counter-argument and reactance to the prevention messages (e.g., calling the hotline)
that were not effective based on their real-life experience.
106
Recommendations
Despite definitional and methodological challenges of estimating the number of human
trafficking victims, prevalence studies with reliable methods and valid measures are still needed
to generate data for policy makers to track human trafficking trends and allocate appropriate
resources to address the issue in a given region. Before carrying out a prevalence study, these
points should be considered:
Microlevel Research. Rather than obtaining national and global estimates, more efforts
should be focused on gathering prevalence data at a local level with manageable parameters
(e.g., specific locations and industries where human trafficking is most rampant, and actors other
than victims for more comprehensive insights to address the problem). Local data with more
precise information on common human trafficking conditions, industries, districts, and
demographics of the at-risk populations can be helpful to identify different target audience
segments and design interventions that are most appropriate for each segment.
Defining human trafficking. This study defines and operationalizes human trafficking
based on ILO’s definition of someone who has experienced involuntary labor and were under the
menace of penalty (Zhang, 2012a). Depending on the objectives of each study, researchers are
encouraged to follow this definition but should also incorporate definitions from a local law
where the study is carried out.
Choosing appropriate research methods. Given the nature of the victim population in
Indramayu who are mostly returned migrants and have experienced human trafficking from other
cities and countries, this study utilizes probability sampling for valid statistical inference and
extrapolation. For a study that focuses on estimating the number of victims who are hard to
reach, hidden, and currently working under exploitative conditions, non-probability sampling
107
such as respondent-driven sampling could be a better option. Data should be collected over time
to capture trends and establish more validity.
Considering unintended consequences. While prevalence data should be used to prioritize
and allocate resources, careful consideration about the unintended consequences of the misuse of
data should be discussed among relevant stakeholders. Prevalence data can be shared and
disseminated with full transparency for peer review and further collaboration. Stakeholders
should also make faithful attempts to include the voices of the victim populations to make use of
the data to advocate for policies that increase their safety, well-being, and economic
opportunities. Nevertheless, data can also be misused to enact harmful policies to ban at-risk
population from traveling, or to arrest and deport undocumented workers. Steps should be made
to protect the privacy and safety of research participants.
In addition to conducting research to gather prevalence data, efforts should be taken to
identify risk and protective factors that would help anti-trafficking organizations in designing
and testing human trafficking prevention strategies.
Using the Multilevel Communication Model to design, implement and evaluate
interventions. The MCM from chapter three could be utilized as a theoretical framework to
measure risk and protective factors of human trafficking, and assess the effects of human
trafficking prevention interventions. The MCM provides a guideline for researchers and
practitioners to identify communication determinants such as integrated connection to the
storytelling network (interpersonal communication with family, friends, and other residents,
consumption of local media, and connection to community-based organizations), communication
hotspots and comfort zones, as well as media exposure to trafficking information, discussion
108
about the issue, and size of migrant network). In addition, background influences (e.g.,
demographics and socioeconomic status) can be collected.
Addressing risk and protective factors. Findings from this study indicate that
communication ecological factors such as ICSN, interpersonal discussion, and migrant network
are strong protective factors against human trafficking. Intervention activities should take
advantage of these communication channels and resources to promote safe migration and
prevention behavior among at-risk populations. On the other hand, people who have been
trafficked are more likely to be re-trafficked again. Activities such as providing social support
and mentorship from trustworthy migrant worker organizations, and offering rehabilitative and
justice services (e.g., counselling and filing restorative claims against unscrupulous employers)
should be offered to the victim population. The results of these risk and protective factors can
also be used for audience segmentation. For example, people with lower communication
resources and demographic segments which are more susceptible to trafficking can be identified
and targeted for campaign efforts. More intensive, tailored activities should be developed for
these target audience with high vulnerability.
Evaluation research design. Evaluation should consist of a pretest/posttest design with a
control group, include probability sampling to increase the representativeness of the findings,
and whenever possible employ panel samples with longitudinal data collection to track short-
term and long-term effects, as well as increase the validity of the findings. Survey should be
developed, pretested, localized, and translated with the help of relevant stakeholders. Qualitative
methods such as in-depth interviews and focus groups can be included to help explain
unexpected survey results and other underlying mechanisms that may hinder the adoption of the
desired trafficking prevention behaviors.
109
Evaluating intervention effects. Knowledge, attitudes, norms, efficacy, perceived risks,
skills & abilities, intention, and anti-trafficking behaviors should be collected as baseline data,
and subsequently after exposure to intervention activities. Variables with lower scores should be
highlighted at the baseline and incorporated into campaign messages and programmatic activities
to influence more effective adoption of prevention behaviors.
Narrative persuasion strategies. For messaging in anti-trafficking campaigns, a model of
culture-centric narratives in health promotion could be adopted to guide the design of culturally
sensitive narratives that audience will be transported into the story, identify with the characters,
model their prevention behaviors, discuss and share the stories with the people in their network
(Larkey & Hecht, 2010). Be prepared to offer prevention behaviors that are easy to follow and
effective to the audience’s needs in the real world. Avoid promoting services such as a hotline
number that is not equipped with qualified personnel to provide correct solutions for the victims.
Intervention channels and activities. This study indicates that a mass media campaign like
broadcasting a documentary has limited effects on knowledge, perceived risks, skills & abilities,
efficacy and intention. To ensure more effective behavior change, communication activities
should also focus on utilizing ICSN, community hotspots, and migrant network of the target
audience (e.g., community theaters, social media applications, local mosques, and schools). For
example, returned migrants could be recruited to do outreach and hold discussions with potential
migrants to educate about the dangers of human trafficking and encourage safe migration.
Holding community gatherings in local “hotspots” could also be effective.
Partnerships. It is important to form partnerships with academic institutions, law
enforcement, labor department, migrant and community organizations to be involved in
designing the research agenda and prioritize issues to tackle in their community. Anti-trafficking
110
agenda should not be imposed on a community that may want focus on other pressing issues
such as sanitation, road safety, or other disease prevention. Collaboration with relevant
organizations could increase shared responsibilities, and widespread adoption of the human
trafficking prevention activities. The partnership approach should result in more sustainable
outcomes as community members are invested in the program. In addition, through the process
of research and project implementation, knowledge transfer as well as capacity building between
researchers and community participants would empower the community to strengthen their
communication networks to solve other issues that may not even be related to anti-trafficking
programs.
111
REFERENCES
Ackerson, L. K. & Viswanath, K. (2009). The social context of interpersonal communication and
health. Journal of Health Communication, 14(Suppl. 1), 5–17.
doi:10.1080/10810730902806836
Alpert, E. J., & Chin, S. E. (2017). Human trafficking: Perspectives on prevention. In M.
Chisolm-Straker & H. Stoklosa (Eds) Human trafficking is a public health issue (pp. 379-
400). Cham: Springer International Publishing. doi:10.1007/978-3-319-47824-1_22
Andrijasevic, R. (2007). Beautiful dead bodies: Gender, migration and representation in anti-
trafficking campaigns. Feminist Review, 86(1), 24-44.
Anti-Slavery International. (2006). Trafficking in woman, forced labor and domestic work in the
context of the Middle East and Gulf region. Working paper. Retrieved from
http://www.antislavery.org/ includes/documents/cm_docs/2009/t/traffic_
women_forced_labour_domestic_2006.pdf
Aradau, C. (2004) Neoliberal borders: Review of migration, agency and citizenship in sex
trafficking. Feminist Review, 99, e7-e9.
Archer, D., Boittin, M, & Mo, C. H. (2016). Reducing vulnerability to human trafficking: An
experimental intervention using anti-trafficking campaigns to change knowledge, attitudes,
beliefs, and practices in Nepal. Retrieved from https://www.iie.org/Research-and-
Insights/Publications/DFG-Vanderbilt-Publication
Bajari, A. (2013). Women as commodities, the analysis of local culture factor and
communication approach of women trafficking in West Java, Indonesia. Research on
Humanities and Social Sciences, 3 (5), 193-200.
112
Bales, K. (2012). Disposable people: New slavery in the global economy (3rd ed.). Berkeley,
CA: University of California Press.
Ball-Rokeach, S. J., Kim, Y. C., & Matei, S. (2001). Storytelling neighborhood: Paths to
belonging in diverse urban environments. Communication Research, 28(4), 392-428
Bandura, A. (2002). Social cognitive theory of mass communication. In J. Bryant & D. Zillman
(Eds.), Media effects: Advances in theory and research (2nd ed., pp. 121–153). Mahwah, NJ:
Erlbaum.
Bandura, A. (2004). Health promotion by social cognitive means. Health Education & Behavior,
31(2), 143–164. doi:10.1177/1090198104263660
Belser, P. (2005). Forced labour and human trafficking: Estimating the profits. Geneva:
International Labour Organization
Bernstein, E. (2014). Militarized humanitarianism meets carceral feminism: The politics of sex,
rights, and freedom in contemporary antitrafficking campaigns. Signs, 40(1).
Bleakley, A., Hennessy, M., Fishbein, M., & Jordan, A. (2011). Using the integrative model to
explain how exposure to sexual media content influences adolescent sexual behavior. Health
Education & Behavior, 38(5), 530-540. doi:10.1177/1090198110385775
Central Statistics Agency of Indramayu Regency. (2016) Population statistics according to gender in
Kabupaten Indramayu, 2013-2015. Retrieved from
https://indramayukab.bps.go.id/dynamictable/2016/08/03/16/jumlah-penduduk-menurut-jenis-
kelamin-di-kabupaten-indramayu-2013-2015.html
Centre for Research on Environment Health and Population Activities. (2004). The anti-
trafficking programme in rural Nepal: Assessment of change in awareness and
113
communication among adolescent girls, peers and parents in Baglug District. Gairidhare,
Kathmandu: Centre for Development and Population Activities.
Chaffee, S. H. (1982). Mass media and interpersonal channels: Competitive, convergent, or
complementary? In G. Gumber & R. Cathar (Eds.), Inter/media: Interpersonal communication in a
media world. New York: Oxford University Press
Cho, H., Shen, L., & Wilson, K. (2014). Perceived realism: Dimensions and roles in narrative
persuasion. Communication Research, 41(6), 828-851.
Cleland, J. (1973). A critique of KAP studies and some suggestions for their improvement.
Studies in Family Planning, 4(2), 42-47.
Creswell, J.W., & Plano Clark, V.L. (2011). Designing and conducting mixed methods research
(2nd edn). Thousand Oaks, CA: Sage Publications.
Curtis, R., Terry, K., Dank, M., Dombrowski, K., & Khan, B. (2008). The commercial sexual
exploitation of children in New York City, volume one: The CSEC population in New York
City: Size, characteristics, and needs (Final Report submitted to the National Institute of
Justice). Retrieved from https://www.ncjrs.gov/pdffiles1/nij/grants/225083.pdf
David, F. (2017). Building the infrastructure of anti-trafficking, part II: Why measurement
matters. Anti-Trafficking Review, 8, 148-152. doi:10.14197/atr.20121789
Davison, W. P. (1983). The third-person effect in communication. Public Opinion Quarterly, 47,
1-15.
Davidson, O. J. (2010). New slavery, old binaries: Human trafficking and the borders of
freedom. Global Networks-A Journal of Transnational Affairs, 10(2), 244-261.
Department of State. (2013). Trafficking in persons report. Retrieved July 14, 2014, from
http://www.state.gov/j/tip/rls/tiprpt/2013
114
DeSteno, D., Petty, R. E., Rucker, D. D., Wegener, D. T., & Braverman, J. (2004). Operation of
thought systems. In R. S. Wyer, Jr. & T. Srull (Eds.), Advances in Social Cognition (Vol. 4,
pp. 1–78). Hillsdale, NJ: Erlbaum.
Dillard, J. P., & Nabi, R. L. (2006). The persuasive influence of emotion in cancer prevention
and detection messages. Journal of Communication, 56, 123–139. doi:10.1111/j.1460-
2466.2006.00286.x
Dobransky, K., & Hargittai, E. (2012). Inquiring minds acquiring wellness: Uses of online and offline
sources for health information. Health Communication, 27(4), 331-343.
Doezema, J. (2000). Loose women or lost women? The re-emergence of the myth of white
slavery in contemporary discourses of trafficking in women. Gender Issues, 18(1), 23-50.
Dottridge, M. (2017). Global trafficking prevalence data distorts efforts to stop patterns of human
trafficking. Anti-Trafficking Review, 8, 161-164. doi:10.14197/atr.201217812
Downman, S. (2014). The role of media-based interventions in combating human trafficking in
Thailand. In S. Yea (Ed) Human trafficking in Asia (pp 200-216). London: Routledge.
Emmers, R. (2003). The threat of transnational crime in Southeast Asia: Drug trafficking, human
smuggling and trafficking and sea piracy. UNISCI Discussion Papers, 2, 1-11.
Fedina, L. & DeForge, B. R. (2017) Estimating the trafficked population: Public-health research
methodologies may be the answer, Journal of Human Trafficking, 3(1), 21-38.
doi:10.1080/23322705.2017.1280316
Feeny, N., Foa, E., Treadwell, K. et al. (2004). Posttraumatic stress disorder in youth: A critical review
of the cognitive and behavioural treatment outcome literature. Professional Psychology: Research
and Practice, 35(5), 466–476.
Feingold, D. A. (2005). Think Again: Human trafficking. Foreign Policy, 26-32.
115
Feingold, D. A. (2017). Playing the numbers: The spurious promise of global trafficking statistics’, Anti-
Trafficking Review, 8, 153-156. doi:10.14197/atr.201217810
Fishbein, M. (2000). The role of theory in HIV prevention. AIDS Care, 12, 273–278.
Fishbein, M., & Ajzen, I. (2010). Predicting and changing behavior: The reasoned action
approach. New York: Psychology Press.
Fisher, W. R. (1985). The narrative paradigm: In the beginning. Journal of Communication,
35(4), 74–89. doi:10.1111/j.1460-2466.1985.tb02974.x
Fisher, W. R. (1987). Human communication as narration: Toward a philosophy of reason,
value, and action. Columbia, SC: University of South Carolina Press.
Galarce, E. M., Ramanadhan, S., Weeks, J., Schneider, E. C., Gray, S. W., & Viswanath, K.
(2011). Class, race, ethnicity and information needs in post-treatment cancer patients. Patient
Education and Counseling, 85, 432–439.
Gallagher, A. T. (2014). The global slavery index: Seduction and obfuscation. Retrieved from
https://www.opendemocracy.net/5050/anne-gallagher/global-slavery-index-seduction-and-
obfuscation/
Gallagher, A. T. (2017). What’s wrong with the Global Slavery Index?’, Anti-Trafficking Review, 8, 90-
112. doi:10.14197/atr.20121786
Global Alliance against Traffic in Women (GAATW). (2010). Feeling good about feeling bad: A
global review of anti-trafficking initiatives. Retrieved November 15, 2011 from
http://www.gaatw.org/publications/GAATW_Global_Review.FeelingGood.AboutFeelingBad.
pdf
116
Global Alliance Against Traffic in Women (GAATW). (2011). Moving beyond ‘supply and
demand’ catchphrases. Assessing the uses and limitations of demand-based approaches in
anti-trafficking. Bangkok: GAATW
Global Slavery Index. (2014). The global slavery index 2014. Retrieved March 19, 2015 from
http://www.globalslaveryindex.org/country/indonesia/
Gordis, L. (2009). Epidemiology (4th ed.). Philadelphia, PA: Saunders, Elsevier.
Green, M. C. (2006). Narratives and cancer communication. Journal of Communication, 56(S1),
S163–S183. doi:10.1111/j.1460-2466.2006.00288.x
Green, M. C., & Brock, T. C. (2000). The role of transportation in the persuasiveness of public
narratives. Journal of Personality and Social Psychology, 79, 701–721. doi:10.1037/ 0022-
3514.79.5.701
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.
doi:10.1111/j.1468-2885.2004.tb00317.x
Gulati, G. (2010). Media representation of human trafficking in the United States, Great Britain,
and Canada. Informally published manuscript, Bentley University, Waltham, MA.
Hames, C., Dewar, F. & Napier-Moore, R. (2010). Feeling good about feeling bad: A global
review of anti-trafficking initiatives. Retrieved November 15, 2011 from
http://www.gaatw.org/publications/GAATW_Global_Review.FeelingGood.AboutFeelingBa
d.pdf
Hashash, Y. (2007). Evaluation report. Trafficking project. Hotline for migrant workers and
Isha, L'Isha.
117
Hinyard, L., & Kreuter, M. W. (2007). Using narrative communication as a tool for health
behavior change: A conceptual, theoretical, and empirical overview. Health Education &
Behavior, 34, 777–792. doi:10.1177/1090198106291963
Human Rights Watch. (2004). Help wanted: Abused against female migrant domestic workers in
Indonesia and Malaysia. Retrieved from http://www.hrw.org/reports/2004/07/21/help-
wanted
Human Rights Watch. (2006). Swept under the rug: Abuses against domestic workers around the
world. New York: Human Rights Watch
Human Rights Watch. (2010). Slow reform: Protection of migrant domestic workers in Asia and
the Middle East. New York: Human Rights Watch
Indramayu District Government. (2014). Population statistics. Retrieved October 1, 2014 from
http://www.indramayukab.go.id/statistik/99-kependudukan.html
International Labor Organization. (2005). Global alliance against forced labour: Global report under
the follow-up to the ILO Declaration on Fundamental Principles and Rights at Work. Geneva:
International Labour Office.
International Labour Organization. (2006a). Domestic workers in Southeast Asia: A decent work
priority. ILO project on mobilising action for the protection of domestic workers from forced
labour and trafficking in Southeast Asia. Jakarta: ILO.
International Labour Organization. (2006b). Using Indonesian law to protect and empower
Indonesian migrant workers: Some lessons from the Philippines. Jakarta: ILO.
International Labour Organization. (2009). Operational indicators of trafficking in human
beings. Results from a Delphi survey implemented by the ILO and the European Commission.
Retrieved from
118
http://www.ilo.org/sapfl/Informationresources/Factsheetsandbrochures/WCMS_105023/lang-
- en/index.htm
International Labour Organization. (2012a). ILO 2012 global estimate of forced labour. Retrieved from
http://www.ilo.org/wcmsp5/groups/public/@ed_norm/@declaration/documents/publication/wcms_18
1953.pdf
International Labour Organization. (2012b). Hard to see, harder to count: Survey guidelines to estimate
forced labour of adults and children. Geneva: ILO Publications.
International Labour Organization. (2014). Profits and poverty: The economics of forced labour.
Geneva: ILO Publications.
International Labour Organization. (2017a). Global estimates of modern slavery: Forced labour and
forced marriages. Geneva: ILO Publications.
International Labour Organization. (2017b). Methodology of the global estimates of modern slavery:
Forced labour and forced marriage. Geneva: ILO Publications.
International Organization for Migration. (2010). Labour migration from Indonesia: An overview
of Indonesian migration to selected destinations in Asia and the Middle East. Jakarta: IOM
International Organization for Migration. (2011). Human Trafficking in Indonesia. Retrieved January
20, 2013 from http://www.humantrafficking.org/countries/indonesia
International Organization for Migration. (2012). Study on public attitudes to migrant workers.
Retrieved from http://www.ilo.org/asia/info/WCMS_159851/lang--en/index.htm
International Organization for Migration. (2014). Human Trafficking in Indonesia. Jakarta: IOM
Presentation.
Jobe, A. (2010). The causes and consequences of re-trafficking: Evidence from the IOM human
trafficking database. Geneva: IOM Publications.
119
Kim, Y.C. & Ball-Rokeach, S. J. (2006a). Civic engagement from a communication
infrastructure perspective. Communication Theory, 16(2), 173-197.
Kim, Y.C. & Ball-Rokeach, S. J. (2006b). Community storytelling network, neighborhood
context, and civic engagement: A multilevel approach. Human Communication Research,
32(4), 411-439.
Kim, Y. C., Moran, M. B., Wilkin, H. A., & Ball-Rokeach, S. J. (2011). Integrated connection to
neighborhood storytelling network, education, and chronic disease knowledge among
African Americans and Latinos in Los Angeles. Journal of health communication, 16(4),
393-415.
Kistyarini. (2013). Bertaruh nyawa menjadi TKI [Bet life to become migrant worker]. Retrieved from
http://regional.kompas.com/read/2013/10/16/1745182/Bertaruh.Nyawa.Menjadi.TKI
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(3), 221–235.
doi:10.1007/BF02879904
Laczko, F. (2005). Data and Research on Human Trafficking. International Migration, 43(1-2),
5-16.
Lee, E. S. (1966). A theory of migration. Demography, 3(1), 47-57.
Lim, L.L. (1998). The sex sector: The economic and social bases of prostitution in Southeast
Asia. Geneva, International Labour Office.
Lindgren, D. (2010). Event evaluation for human trafficking in Indonesia (Medan & Jakarta). Bangkok:
Rapid Asia
120
Listiani, T. (2017). An analysis of gendered labor migration in rural Indonesia: A case study of
the Indramayu Regency, West Java, Indonesia. 政 策 科学/Policy Science, 25(1), 153-169.
Liu, Y. (2010). The commodification of human life: Human trafficking in the age of
globalization. Macalester International, 25(1), 11.
MacPhail, C., Pettifor, A., Moyo, W., & Rees, H. (2009). Factors associated with HIV testing
among sexually active South African youth aged 15- 24 years. AIDS Care, 21, 456–467.
Magenta Consulting. (2007). Evaluation of the educational campaign: Human trafficking phenomenon –
part II. Retrieved from
http://arhiva.lastrada.md/date/rapoarte/Evaluation_of_the_educational_campaign_(part_2).pdf
Marshall, P. (2011). Rethinking trafficking prevention: A guide to applying behavioral theory.
Retrieved from http://www.ungift.org/doc/knowledgehub/resource-
centre/rethinking_Trafficking_Prevention.pdf
Mashud, M. (2010). Perspektif fenomenologi tentang trafficking TKW [Phenomenological perspective
on trafficking experienced by Indonesian woman migrant workers]. Masyarakat, Kebudayaan, dan
Politik, 21(2), 146-154.
Matsaganis, M. D., & Wilkin, H. A. (2015). Communicative social capital and collective
efficacy as determinants of access to health-enhancing resources in residential
communities. Journal of health communication, 20(4), 377-386.
Mattar, M. Y. (2004). Trafficking in persons: An annotated bibliography. Law Library Journal, 96(4),
669-726.
McDivitt, J. A., Zimicki, S., & Hornik, R. C. (1997). Explaining the impact of a
communication campaign to change vaccination knowledge and coverage in the Philippines.
Health Communication, 9(2), 95-118.
121
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(6), 674–682.
doi:10.1037/a0025395
Mercy, J. A., Rosenberg, M. L., Powell, K. E., Broome, C. V., & Roper, W. L. (1993). Public
health policy for preventing violence. Health Affairs, 12(4), 7-29.
Merrigan, G., & Huston, C. L. (2009). Communication research methods (2nd Ed). Wadsworth:
Belmont, CA.
Minnesota Advocates for Human Rights. (2003). Testing your knowledge about trafficking.
Retrieved March 8, 2014 from
http://www1.umn.edu/humanrts/svaw/trafficking/training/materials/TestKnowledge.PDF
Morgan, D. (1997). The focus group guidebook (Vol. 1). Thousand Oaks, CA: Sage publications.
Morton, T. A. & Duck, J. M. (2001). Communication and health beliefs: Mass and interpersonal
influences on perceptions of risk to self and others. Communication Research, 28(5), 602–
626. doi:10.1177/009365001028005002
Moyer-Guse, E. (2008). Toward a theory of entertainment persuasion: Explaining the persuasive
effects of entertainment–education messages. Communication Theory, 18(3), 407–425.
doi:10.1111/j.1460-2466.2008.01408.x
Moyer-Guse, E., & Nabi, R. L. (2011). Comparing the effects of entertainment and educational
television programming on risky sexual behavior. Health Communication, 26(5), 416–426.
doi:10.1080/10410236.2011.552481
122
Moyer-Guse, 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. doi:10.1111/j.1460-2466.2011.01551.x
Murphy, S. T., Frank, L. B., Chatterjee, J. S., & Baezconde-Garbanati, L. (2013). Narrative
versus non-narrative: The role of identification, transportation, and emotion in reducing health
disparities. Journal of Communication, 63(1), 116-137.
Murphy, S. T., Frank, L. B., Moran, M. B., & Patnoe-Woodley, P. (2011). Involved, transported,
or emotional? Exploring the determinants of change in knowledge, attitudes, and behavior in
entertainment-education. Journal of Communication, 61(3), 407–431. doi:10.1111/j.1460-
2466.2011.01554.x
Nabi, R. L. (2002). Discrete emotions and persuasion. In J. Dillard & M. Pfau (Eds.), The
persuasion handbook (pp. 289–308). Thousand Oaks, CA: Sage.
Nichter, M. 1993. Social science lessons from diarrhea research and their application to ARI.
Human Organization, 52(1), 53-67.
Niederdeppe, J., Shapiro, M. A., & Porticella, N. (2011). Attributions of responsibility for
obesity: Narrative communication reduces reactive counterarguing among Liberals. Human
Communication Research, 37(3), 295–323. doi:10.1111/j.1468-2958.2011. 01409.x
Nieuwenhuys, C., & Pécoud, A. (2007). Human trafficking, information campaigns, and
strategies of migration control. American Behavioral Scientist, 50(12), 1674-1695.
Orphant, M. (2001). Trafficking in persons: Myths, methods, and human rights. Population
Reference Bureau. Retrieved from
http://www.prb.org/Publications/Articles/2001ATraffickinginPersonsMythsMethodsandHuma
nRights.aspx
123
Park, D. C. (2017). The diaspora of Indonesian female domestic workers in Saudi Arabia: A case of
abuse and mistreatment. The Journal of Contemporary Asian Societies, 2(1).
Pelto, J. P. & Pelto, G. H. (1997). Studying knowledge, culture, and behavior in applied medical
anthropology. Medical Anthropology Quarterly 11(2), 147- 163.
Peltzer, K., Matseke, G., Mzolo, T., & Majaja, M. (2009). Determinants of knowledge of HIV
status in South Africa: Results from a population-based HIV survey. BMC Public Health, 9,
174. doi:10.1186/ 1471-2458-9-174
Pennington, J. R., Ball, A. D., Hampton, R. D., & Soulakova, J. N. (2009). The cross-national
market in human beings. Journal of Macromarketing, 29(2), 119-134.
doi:10.1177/0276146708327630
Perry, K. M. & McEwing, L. (2013). How do social determinants affect human trafficking in
Southeast Asia, and what can we do about it? A systematic review. Health and Human Rights,
15(2), 138-159.
Picarelli, J. T. (2015). Science versus slavery: The National Institute of Justice and the future
directions of knowledge put to the service of the anti-trafficking movement. Journal of
Human Trafficking, 1(1), 39–55. doi:10.1080/23322705.2015.1013780
Pinquart, M., & Duberstein, P. R. (2004). Information needs and decision-making processes in
older cancer patients. Critical Reviews in Oncology/Hematology, 51, 69–80.
Pitts, W. J., Barrick, K., Zhang, S. X., & Lattimore, P. K. (2015). Estimating labor trafficking
among farmworkers: An inverse sampling strategy based on reliable housing
predictions. Journal of Human Trafficking, 1(2), 117-135.
Pollock, G. (1998). Modernity and the spaces of femininity. In N. Mirzoeff (Ed), The visual
culture reader. London and New York: Routledge.
124
Priasto, A. (2015). Summary of Indonesia’s poverty analysis: ADB papers on Indonesia.
Retrieved from https://www.adb.org/sites/default/files/publication/177017/ino-paper-04-
2015.pdf
Pusdalisbang Office. (2014). Population and labor force. Retrieved from
http://pusdalisbang.jabarprov.go.id/pusdalisbang/berkas/jabardalamangka/524277_penduduk.pdf
Raphael, J. (2017). Returning trafficking prevalence to the public policy debate: Introduction to
the special issue, Journal of Human Trafficking, 3(1), 1-20.
doi:10.1080/23322705.2017.1280314
Raymond, J. G. (2002). The new UN trafficking protocol. Women’s Studies International Forum,
25(5), 491–502.
Riley, P., Thomas, G. F., Weintraub, R., Noyes, A., & Dixon, S. (2014). Good governance and
strategic communication: A communication capital approach. In D. Holtzhausen & A. Zerfass
(Eds), The Routledge handbook of strategic communication (pp. 201-213). New York, NY:
Routledge.
Rizkytha, I. Z., & Abraham, J. (2015). Attitudes toward risk, social dominance orientation and perceived
scarcity of the opposite sex on Indonesian woman migrant workers. International Journal of
Research Studies in Psychology, 4(2).
Rhodes, F., Stein, J. A., Fishbein, M., Goldstein, R. B., & Rotheram-Borus, M. J. (2007). Using
theory to understand how interventions work: Project RESPECT, condom use, and the
integrative model. AIDS and Behavior, 11(3), 393-407.
Robinson, C., Branchini, C., & Thame, C. (2017). Global trafficking prevalence data advances the fight
against trafficking in persons. Anti-Trafficking Review, 8, 157-160. doi:10.14197/atr.201217811
125
Rogers, E. & Storey, D. (1987) Communication campaigns. In C. Berger & S. Chaffee (Eds.),
Handbook of communication science (pp. 817-846). Newbury Park, CA: Sage.
Rothman, E. F., Stoklosa, H., Baldwin, S. B., Chisolm-Straker, M., Kato Price, R., Atkinson, H.
G., & HEAL Trafficking. (2017). Public health research priorities to address US human
trafficking. American Journal of Public Health, 107(7):1045-1047.
doi:10.2105/AJPH.2017.303858.
Salmon, C. & Atkin, C. (2003) Using media campaigns for health promotion. In T. Thompson,
A. Dorsey, K. Miller and R. Parrott, Handbook for Health Communication (pp. 449-472). NY:
Routledge.
Schauer, E. J. & Wheaton, E. M. (2006). Sex trafficking into the United States: A literature
review. Criminal Justice Review, 31(2), 146-169.
Seo, M., & Matsaganis, M. D. (2013). How interpersonal communication mediates the
relationship of multichannel communication connections to health-enhancing and health-
threatening behaviors. Journal of health communication, 18(8), 1002-1020.
Shalit, D.A., Heynen, R., & Van Der Meulen, E. (2014). Human trafficking and media myths:
Federal funding, communication strategies, and Canadian anti-trafficking
programs. Canadian Journal of Communication, 39(3), 385-412.
Shehadeh, N., & McCoy, H. V. (2014). Risky sexual behaviors: The role of ethnic identity in HIV risk
in migrant workers. Journal of the Association of Nurses in AIDS Care, 25(4), 330-340.
Silvey, R. (2007). Mobilizing piety: Gendered morality and Indonesian–Saudi transnational
migration. Mobilities, 2(2), 219-229.
Singhal, A., & Rogers, E. M. (1999). Entertainment education: A communication strategy for
social change. Mahwah, NJ: Erlbaum.
126
Singhal, A., & Rogers, E. M. (2002). A theoretical agenda for entertainment-
education. Communication theory, 12(2), 117-135.
Slater, M. D., & Rouner, D. (2002). Entertainment education and elaboration likelihood:
Understanding the processing of narrative persuasion. Communication Theory, 12, 173–191.
doi:10.1111/j.1468-2885.2002.tb00265.x
Stark, C. & Hodgson, C. (2003). Sister oppressions: A comparison of wife battering and
prostitution in M. Farley (ed), Prostitution, trafficking, and traumatic stress. Binghamtom,
NY: The Hawthorne Press.
Steinfatt, T. M., & Baker, S. (2011). Measuring the extent of sex trafficking in Cambodia–2008. United
Nations Inter-Agency Project (UNIAP) on Human Trafficking. Retrieved from http://www.no-
trafficking.org/reports_docs/estimates/uniap_cb_estimates_report.pdf
Stewart, D. W., & Shamdasani, P. N. (2014). Focus groups: Theory and practice (3
rd
Ed). Los Angeles,
CA: Sage Publications.
Supriyoko (1990, July). Indonesian woman migrant workers: Background and notes. Paper presented at
Panel Discussion Forum of Indonesian Woman’s Dignity as Labor Abroad, Woman Tamansiswa
Branch Sarjanawiyata, Yogyakarta. Retrieved from
http://journal.amikom.ac.id/index.php/KIDA/article/viewFile/5146/2819
Synder, L. (2001). How effective are mediated health campaigns? In R. E. Rice & C. K. Atkin
(Eds), Public communication campaigns (3
rd
ed. pp. 181-190). Thousand Oaks, CA: Sage.
Tal-Or, N., & Cohen, J. (2010). Understanding audience involvement: Conceptualizing and
manipulating identification and transportation. Poetics, 38(4), 402–418.
doi:10.1016/j.poetic.2010.05.004
127
Taylor-Clark, K. A., Viswanath, K., & Blendon, R. J. (2010). Communication inequalities during
public health disasters: Katrina’s wake. Health Communication, 25, 221–229.
Tenkorang, E.Y. & Owusub, G.A.(2010).Correlates of HIV testing among women in Ghana:
Some evidence from the Demographic and Health Surveys. AIDS Care, 22, 296–307.
Thainiyom, P. (2011). A media campaign to increase awareness and prevention of human trafficking in
Asia: Background strategies and outcome evaluation of the MTV EXIT Campaign (End Exploitation
and Trafficking), Paper presented at the 2011 Interdisciplinary Conference on Human Trafficking,
Lincoln, Nebraska.
Tomkinson, S. (2012). The multiplicity of truths about human trafficking: Beyond “the sex
slave” discourse. Political Science Journal, 7(1), 50-67.
Tyldum, G., & Brunovskis, A. (2005). Describing the unobserved: methodological challenges in
empirical studies on human trafficking. International Migration, 43(1/2), 17–34.
Tyldum, G., Tveit, M., & Brunovskis, A. (2005). Taking stock: a review of the existing research
on trafficking for sexual exploitation. Fafo-report 493. Oslo: Fafo.
United Nations Development Programme. (2018). About Indonesia. Retrieved from
http://www.id.undp.org/content/indonesia/en/home/countryinfo.html
UNODC. (2004). United Nations convention against transnational organized crime and the
protocols thereto. Retrieved October 14, 2014 from
http://www.unodc.org/unodc/treaties/CTOC/
UNODC. (2008a). Toolkit to Combat Trafficking in Persons, Awareness Raising. Retrieved from
http://www.unodc.org/documents/human-trafficking/Toolkit-files/08-58296_tool_9-8.pdf
128
UNODC. (2008b). Compendium of best practices on human trafficking by non-governmental
organizations. Retrieved from http://www.unodc.org/documents/human-
trafficking/India_Training_material/Compendium_of_Best_Practices_by_NGOs.pdf
UNODC. (2009). Global report on trafficking in persons. Retrieved from
http://www.unodc.org/documents/Global_Report_on_TIP.pdf
UNODC. (2016). Global report on trafficking in persons 2016. Vienna: United Nations
Publication.
USAID (2012). Counter trafficking in persons policy. Retrieved October 9, 2013 from
http://www.usaid.gov/sites/default/files/documents/2496/C-
TIP_Field_Guide_Final_April%205%202013.pdf
Utami, N. (2012). Judicial review of the implementation of employment insurance in providing
protection for the placement of Indonesian workers abroad. Retrieved from
http://pustaka.unpad.ac.id/archives/117820/#
Valente, T. W. & Saba, W. P. (1998). Mass media and interpersonal influence in a reproductive
health communication campaign in Bolivia. Communication Research, 25(1), 96–124.
doi:10.1177/009365098025001004
Van der Laan, P., Smit, M., Busschers, I., & Aarten, P. (2011). Cross-border trafficking in
human beings: prevention and intervention strategies for reducing sexual exploitation.
Campbell Systematic Reviews, 9. doi:10.4073/csr.2011.9
Walk Free Foundation. (2016). The global slavery index 2016. Retrieved from
https://www.globalslaveryindex.org/download/
Webb, T. L., Joseph, J., Yardley, L., & Michie, S. (2010). Using the Internet to promote health
behavior change: A systematic review and meta-analysis of the impact of theoretical basis,
129
use of behavior change techniques, and mode of delivery on efficacy. Journal of Medical
Internet Research, 12(1), e4.
Weber, E. U., Blais, A.-R., & Betz, N. E. (2002). A domain-specific risk-attitude scale: Measuring risk
perceptions and risk behaviors. Journal of Behavioral Decision Making, 15, 263-290.
doi:10.1002/bdm.414
Weitzer, R. (2010). The movement to criminalize sex work in the United States. Journal of Law
and Society, 37, 61–84. doi:10.1111/jols.2010.37.issue-1
Weitzer, R. (2012). Sex trafficking and the sex industry: The need for evidence-based theory and
legislation. The Journal of Criminal Law & Criminology, 101, 1337–1369.
Weitzer, R. (2014a). Miscounting human trafficking and slavery. Retrieved from
https://www.opendemocracy.net/beyondslavery/ronald-weitzer-miscounting-human-
trafficking-and-slavery/
Weitzer, R. (2014b). New directions in research on human trafficking. The Annals of the
American Academy of Political and Social Science, 653, 6–24.
doi:10.1177/0002716214521562
Wilkin, H. A. (2013). Exploring the potential of communication infrastructure theory for
informing efforts to reduce health disparities. Journal of Communication, 63(1), 181-200.
Wilkin, H. A. and Ball-Rokeach, S. J. (2006). Reaching at risk groups: The importance of health
storytelling in Los Angeles Latino media. Journalism: Theory, Practice, 7(3), 299-320.
Wilkin, H. A. & Ball-Rokeach, S. J. (2011). Hard-to-reach? Using health access status as a way
to more effectively target segments of the Latino audience. Health Education Research,
26(2), 239–253. doi:10.1093/her/cyq090
130
Wilkin, H. A., Moran, M. B., Ball-Rokeach, S. J., Gonzalez, C. & Kim, Y.-C. (2010).
Applications of communication infrastructure theory. Health Communication, 25(6–7), 611–
612. doi:10.1080/10410236.2010.496839
Wilkin, H. A., Stringer, K. A., O’Quin, K., Montgomery, S. A. & Hunt, K. (2011). Using
communication infrastructure theory to formulate a strategy to locate ‘‘hard-to-reach’’
research participants. Journal of Applied Communication Research, 39(2), 201–213.
doi:10.1080/00909882.2011.556140
Williamson, C., Perdue, T., Belton, L., & Burns, O. (2012). Domestic sex trafficking in Ohio
(Ohio Human Trafficking Commission Final Report). Retrieved from
http://www.ohioattorneygeneral.gov/getattachment/1bc0e815-71b6-43f5-ba45-
c667840d4a93/2012-Domestic-Sex-Trafficking-in-Ohio-Report.aspx
World Bank (2017). Indonesia’s global worker: Juggling opportunities and risks. Retrieved from
http://pubdocs.worldbank.org/en/357131511778676366/Indonesias-Global-Workers-Juggling-
Opportunities-Risks.pdf
Wulan, T. R., Street, H. B., & Indonesia, P. (2014). Employment policy and its implications for
protection of Indonesian migrant workers (IMW): A comparative study between Hong Kong
and Malaysia. In 2014 International Conference on Public Management. Atlantis Press.
Yea, S. (2017). Editorial: The politics of evidence, data and research in anti-trafficking work.
Anti-Trafficking Review, 8, 1-13. doi:10.14197/atr.20121781
Yoder, P. S. (1997). Negotiating relevance: Beliefs, knowledge and practice in international
health projects. Medical Anthropology Quarterly, 11(2), 131-146.
Yzer, M. (2012). The integrative model of behavioral prediction as a tool for designing health
messages. Health Communication Message Design, Theory and Practice, 21-40.
131
Zhang, S. X. (2012a). Measuring labor trafficking: A research note. Crime, Law, & Social
Change, 58, 469-482. doi:10.1007/s10611-012-9393-y
Zhang, S. X. (2012b). Trafficking of migrant laborers in San Diego County: Looking for a
hidden population. Retrieved from https://www.ncjrs.gov/pdffiles1/nij/grants/240223.pdf
Zhang, S. X., Spiller, M. W., Finch, B. K., & Qin, Y. (2014). Estimating labor trafficking among
unauthorized migrant workers in San Diego. The Annals of the American Academy of Political
and Social Science, 653(1), 65-86.
Zimmerman, C., Hossain, M., Yun, K., Roche, B., Morison, L. and Watts, C. (2006). Stolen Smiles: The
physical and psychological health consequences of women and adolescents trafficked in Europe.
London School of Hygiene and Tropical Medicine (LSHTM). Retrieved November 25, 2014,
from http://www.lshtm.ac.uk/genderviolence/recent.htm
132
APPENDIX I: SURVEY INSTRUMENTS
Demographic and Socioeconomic Variables
Sex:
Male 1
Female 2
Age: What is your year of birth? (WRITE IN A YEAR E.G. 1985)
Residence Length: How long have you lived in this community/? IN YEARS (Not including time
spent working AWAY)
Ethnicity: What is your ethnicity? (Select one)
Javanese 1
Sundanese 2
Madura 3
Batak 4
Padang 5
Betawi 6
Other 7
Language: What is your spoken language? (Select all that apply)
Bahasa 1
Sundanese 2
Javanese 3
Cirebonan 4
Betawi 5
Other 6
Marital Status: What is your current relationship status? (Select one)
Married 1
Single 2
Divorced 3
Separated 4
Widowed 5
133
Years of Education: How many years of formal education do you have?
Years
Level of Education: What is the highest level of education you have completed? (Select one)
No formal education 1
Graduated from SD 2
Some SMP 3
Graduated from SMP 4
Some SMA 5
Graduated from SMA 6
Some university 7
Graduated from a
university
8
Postgraduate 9
Household Size: How many of your immediate family members (father, mother, and their
children) are currently living in your home?
Dependents: How many dependents do you have to support?
Household Income: Considering the combined income for all household members from all
sources, what is your best estimate of your household income for the past 12 months?
Mr. Rupiah
Mrs./Ms. Rupiah
Child Rupiah
Sibling Rupiah
Other Rupiah
Total Rupiah
Religion: Do you consider yourself a member of any of the following faiths? (Select one)
Islam 1
Hindu 2
Buddhism 3
Catholic 4
Other Christians 5
134
Other 6
None 7
Employment: What is your current employment status? (Select one)
Employed full-time 1
Employed part-time 2
Self-employed 3
Temporarily laid off 4
Unemployed 5
Home duties 6
Retired 7
Student 8
Permanently disabled 9
Experience with migration
Have you ever moved away from home for work?
Yes 1
No 2
Human trafficking
Do you think you were a victim of human trafficking or exploitation?
Yes 1
No 2
Did you experience any of the following during your employment away from home? (Select all
that apply)
Involuntary labor
I was working with reduced pay or without pay to repay my loan to the employer
and/or recruitment agency.
1
Menace of penalty
Threats and physical harm
I was psychologically abused with threat of violence against me and/or my loved
ones.
2
I was physically abused at any point during my travel to the workplace. 3
I was physically abused at my workplace. 4
I was sexually abused at my workplace 5
Restriction and depravation
My employer confiscated my passport, identification documents or other legal
documents during my employment.
6
135
I was not allowed to communicate freely with my family and friends. 7
I was not allowed to quit my job. 8
I was imprisoned at my workplace 9
I was denied medical treatment when I was sick or had an injury 10
I was denied proper food and/or water 11
Exploitative Labor Practices
I was not allowed to keep the money I earned. 12
I was not given a day off during my period of work. 13
I was forced to work excessively long hours. 14
I was forced to consume alcohol by my employers 15
Escape or rescue
How did you respond to the situations you faced from Q11c? (Select all that apply)
I called the anti-human trafficking/NGO hotline for help. 1
I called the embassy/consulate for help. 2
I called to passersby near the workplace for help. 3
I called my fellow workers for help 4
I called my families for help 5
I managed to escape from the workplace on my own. 6
I stopped working. 7
I went to an NGO/migrant organization for help. 8
I went to the police/local authority to file a report. 9
I did not seek any help. 10
How did you get out of the troubles that you experienced from Q11c? (Select all that apply)
I was rescued from the workplace by the police, NGO, and/or
embassy employees.
1
I escaped from the workplace on my own and went to the police,
NGO or embassy for help.
2
I was eventually dismissed by my employer. 3
Protection and justice
Did you do any of the following after you were rescued and/or left your employment? (Select all
that apply)
I cooperated with the police to testify against the employer and/or
sponsor in a criminal lawsuit.
1
I had a lawyer file a civil lawsuit against the employer and/or
sponsor.
2
I filed a worker insurance claim to receive my compensation/lost
wages.
3
136
I received my worker insurance payment. 4
I received financial assistance from the government. 5
I was placed in a rehabilitation shelter. 6
I had a medical check-up and treatment. 7
Health problems
Did you experience any of the following health issues as a result of your employment? (Select all
that apply)
Physical injury 1
Chronic pain in the body 2
Permanent physical disability 3
Depression 4
Anxiety 5
Suicide attempts 6
Other mental health problems 7
HIV/AIDS 8
Other Sexually Transmitted Diseases 9
Outcome Variables
Human Trafficking Prevention Behavior
Have you tried to find information about how to migrate safely for work? (Select all the apply)
No 1
Yes, for myself 2
Yes, for a family member 3
Yes, for a friend 4
How did you find safe migration information? (Select all that apply)
Family member 1 Radio 12
Relative 2 Internet website 13
Friend 3 Social Media 14
Coworker 4 Recruitment agency (PPTKIS or
PJTKI)
15
Teacher 5 Government agency (BP3TKI or
Disnakertrans)
16
Religious leader 6 Migrant/Labor Union Organization 17
Community leader 7 Non-governmental organization
(NGO)
18
Sponsor 8 Police 19
Brochures/Pamphlets 9 MTV EXIT 20
137
Magazines 10 Awareness-raising events 21
Newspaper 11 Other 22
Have you ever called a Human Trafficking Hotline Number?
Yes 1
No 2
Intention to practice human trafficking prevention behavior
If you were to migrate for work in the next few years, how likely would you be to do any of the
following? (Select one per row)
Very
Unlikely
Unlikely Somewhat
likely
Likely Very
likely
A Seek information on safe migration
and human trafficking from an
NGO/migrants organization.
1 2 3 4 5
B Seek information from the local
and/or national government (BP3TKI
and/or Disnakertrans)
1 2 3 4 5
C
Call a hotline to seek advice on safe
migration for work.
1 2 3 4 5
D
Become a member of an NGO or
migrants organization to receive
advice and assistance on migration.
1 2 3 4 5
E Ask friends and family for advice. 1 2 3 4 5
F Make sure I keep my legal and
identification documents in my
possession.
1 2 3 4 5
Skills and abilities
Do you remember seeing a human trafficking hotline number in the documentary?
Yes 1
No 2
What was this hotline number?
What would make you call the human trafficking hotline? (Select all that apply)
To seek information about a recruitment agency 1
To verify the legitimacy of an employer 2
138
To report suspicious activity 3
To seek advice about an employment contract 4
To seek advice on a wage dispute with an employer 5
To seek advice about my traveling documents 6
To seek help regarding abuse in the workplace 7
Would not call 8
Other 9
Perceived norms
If you had an 18-year old daughter, would you ask her to move away from home for work?
(Select one)
Definitely No Probably No Probably Yes Definitely Yes
1 2 3 4
If you had an 18-year old son, would you ask him to move away from home for work? (Select
one)
Definitely No Probably No Probably Yes Definitely Yes
1 2 3 4
Efficacy
How helpful do you think each the following would be in preventing a worker from getting in
trouble when working away from home…. (Select one per row)
Not at
all
helpful
Not
helpful
Somewhat
helpful
Helpful Extremely
helpful
A Seeking information on safe
migration and human trafficking
from an NGO/migrants
organization.
1 2 3 4 5
B Seeking information from the
local and/or national government
(BP3TKI and/or Disnakertrans)
1 2 3 4 5
C Calling a hotline to seek advice
on safe migration for work.
1 2 3 4 5
D Becoming a member of a
migrant organization.
1 2 3 4 5
E Asking friends and family for
advice.
1 2 3 4 5
F Keeping their legal and
identification documents in their
possession.
1 2 3 4 5
139
Perceived risks
Think about workers in trouble (human trafficking). Please indicate whether you agree or
disagree with the following statements: (Select one per row)
Strongly
disagree
disagree Neither agree
nor disagree
Agree Strongly
agree
A I believe that this is a
serious problem in
Indonesia.
1 2 3 4 5
B I believe that a worker
in trouble suffers serious
negative consequences
in his/her life.
1 2 3 4 5
C I believe that we should
NOT be concerned
about this issue in this
country.
1 2 3 4 5
D I could become a worker
in trouble.
1 2 3 4 5
Attitudes regarding human trafficking prevention behavior
How much do you agree or disagree that finding information about safe migration is… (Select
one per row)
Strongly
disagree
disagree Neither
agree nor
disagree
Agree Strongly
agree
A … easy 1 2 3 4 5
B … important 1 2 3 4 5
Do you agree or disagree that a hotline would be… (Select one per row)
Strongly
disagree
disagree Neither
agree nor
disagree
Agree Strongly
agree
A …easy to remember. 1 2 3 4 5
B …useful in emergency. 1 2 3 4 5
C …helpful in providing
information.
1 2 3 4 5
140
Knowledge of human trafficking
Have you heard of the following phrases? (Select one per row)
A Human trafficking Yes 1 No 2
B Safe migration Yes 1 No 2
C Modern-day slavery Yes 1 No 2
D Exploitation Yes 1 No 2
Which of the following statements describes human trafficking? (Select all that apply)
A A worker in trouble/human trafficking victim could
be……
Yes No
A1 ………people who left or were taken away from their
country or city and tricked or forced to do a job in
which they were exploited.
1 2
A2 ……… lured by people they personally know and trust. 1 2
A3 ………recruited through fake job opportunities. 1 2
A4 ………someone who is forced to work to repay a loan. 1 2
A5 ………a domestic worker who is abused by his/her
employer.
1 2
A6 …..…someone who is forced to work longer hours than
were written in the contract or promised.
1 2
A7 …..…people who are not receiving wages and/or
having their salary withheld by their employers.
1 2
B Human trafficking can consist of the forced labor of
men, women, and children.
1 2
C The confiscation of someone’s passport or legal
identity can be a part of the human trafficking process.
1 2
Environmental constraints
Would the following be more likely to make you move away from home for work? (Select one
per row.)
Very
unlikely
Unlikely Somewhat
likely
Likely Very
likely
A Death of a close family
member
1 2 3 4 5
B Having large debts to pay off 1 2 3 4 5
C One of your parents lost his
or her job
1 2 3 4 5
D You were unemployed for a
long time
1 2 3 4 5
141
E You have a child or other
family member to support
1 2 3 4 5
F You get a divorce (if you are
still single, imagine that you
are married)
1 2 3 4 5
G No job opportunities in my
community
1 2 3 4 5
H You were abused by your
family members at home
1 2 3 4 5
Communication Ecology
Integrated connection to the storytelling network (ICSN)
Connections to local media
In the past seven days, on how many days did you … (Select one frequency per row)
A read a national newspaper? 0 1 2 3 4 5 6 7
B read a local newspaper? 0 1 2 3 4 5 6 7
C watch news on television? 0 1 2 3 4 5 6 7
D listen to radio talk shows or
news?
0 1 2 3 4 5 6 7
E use the Internet, other than email? 0 1 2 3 4 5 6 7
F use Facebook? 0 1 2 3 4 5 6 7
G use twitter? 0 1 2 3 4 5 6 7
H talk on the phone? 0 1 2 3 4 5 6 7
I text on the phone? 0 1 2 3 4 5 6 7
J use texting application on the
phone?
0 1 2 3 4 5 6 7
K use instant messaging 0 1 2 3 4 5 6 7
L use email 0 1 2 3 4 5 6 7
Connections to community organizations
Are you currently a member of the following community
organizations?
No Yes
1
Sports, gym, fitness, or recreational groups (such as
football, badminton, saepa takro, volleyball, book
club, flower arrangement, video games, chess, etc.)
0 1
2 Cultural, religious, or ethnic centers 0 1
3 Village committee member (e.g., village fund) 0 1
4
Educational group/center (e.g., university, non-formal
education, school clubs)
0 1
5 Other, please specify: ________ 0 1
142
Intensity of interpersonal neighborhood storytelling
How often do you have discussions with other people about things happening in their
neighborhood?
Never All the Time
1 2 3 4 5 6 7 8 9 10
Communication hotspots
What is one place in your community where people get together and chat?
A Grocery store 1
B Traditional grocery 2
C Rice stalls/stalls 3
D Vegetable stalls 4
E Cafe 5
F Market 7
G School 8
H Library 9
I Park 10
J Neighborhood watch base 11
K Bus stop 12
L Community Center 13
M Barber shop 14
N Community organization 15
O Mosque 16
P Business 17
Q Gym/Recreational Center 18
R Restaurant/Coffee Shop 19
S Shopping Mall 20
T Local Government Office 21
U Local Leader’s House 22
V Other (Specify)______________ 23
Comfort zones
Are you comfortable to talk about human trafficking with others at the following locations?
(Select one per row)
No Yes
A Grocery store 0 1
B Traditional grocery 0 1
C Rice stalls/stalls 0 1
143
D Vegetable stalls 0 1
E Cafe 0 1
F Market 0 1
G School 0 1
H Library 0 1
I Park 0 1
J Neighborhood watch base 0 1
K Bus stop 0 1
L Community Center 0 1
M Barber shop 0 1
N Community organization 0 1
O Mosque 0 1
P Business 0 1
Q Gym/Recreational Center 0 1
R Restaurant/Coffee Shop 0 1
S Shopping Mall 0 1
T Local Government Office 0 1
U Local Leader’s House 0 1
V
Other
(Specify)______________
0 1
Interpersonal discussion about trafficking
With whom have you discussed the dangers of working away from home? (Select all that apply)
a sponsor 1
your spouse 2
your mother 3
your father 4
your daughter 5
your son 6
your uncle 7
your aunt 8
your male friend 9
your female friend 10
your coworker 11
your religious leader 12
someone else? (Please specify) 13
no one 14
Migrant Network
Do any of the following people have moved away from home for work? (Select all that apply)?
My spouse 1
144
My mother 2
My father 3
My daughter 4
My son 5
My brother 6
My sister 7
My uncle 8
My aunt 9
My male friend 10
My female friend 11
My coworker 12
Other people? (Please
specify)
13
No one 14
Media exposure to trafficking
How often have you seen or heard information about workers in trouble (human trafficking) in
the following way? (Select one per row)
Media Not at
all
Rarely Sometimes Often All the
time
A Newspaper 1 2 3 4 5
B Magazine 1 2 3 4 5
C Radio 1 2 3 4 5
D Television 1 2 3 4 5
E Internet 1 2 3 4 5
F Social media 1 2 3 4 5
G Billboard 1 2 3 4 5
Transportation
Do you agree or disagree with the following statements about the documentary? (Select one per
row)
Strongly
disagree
disagree Neither
agree nor
disagree
Agree Strongly
agree
I After I finished
watching the
documentary, I found it
easy to put it out of my
mind.
1 2 3 4 5
145
II I could picture myself in
the scenes shown in the
documentary.
1 2 3 4 5
III I found my mind
wandering while
watching the
documentary.
1 2 3 4 5
IV I found myself thinking
of ways the characters in
the documentary could
have behaved
differently.
1 2 3 4 5
V I wanted to learn what
eventually happened to
the people in the
documentary.
1 2 3 4 5
VI I was thinking intensely
while watching the
documentary.
1 2 3 4 5
VII The events in the
documentary are
relevant to my everyday
life.
1 2 3 4 5
VIII The events shown in the
documentary have
changed how I live my
life.
1 2 3 4 5
IX The information in this
documentary was
important to me.
1 2 3 4 5
X I understood what the
documentary was asking
me to do.
1 2 3 4 5
Identification
How similar do you think your life circumstances are to the following people in the
documentary? (Select one per row)
Not
similar at
all
Not
similar
Somewhat
similar
Similar Very
similar
I Siti (a woman who went
to Malaysia and worked
as a maid from house to
house)
1 2 3 4 5
146
II Ismail (a man who went
to Northern Sumatra and
had to work in the
forest)
1 2 3 4 5
III Ika (a girl who was
deceived into sex work
in Batam)
1 2 3 4 5
Do you think you could end up in the same situation as the following people in the documentary?
(Select one per row)
Very
unlikely
Unlikely Somewhat
likely
Likely Very
likely
I Siti (a woman who went
to Malaysia and worked
as a maid from house to
house)
1 2 3 4 5
II Ismail (a man who went
to Northern Sumatra and
had to work in the
forest)
1 2 3 4 5
III Ika (a girl who was
deceived into sex work
in Batam)
1 2 3 4 5
147
APPENDIX II: FOCUS GROUP QUESTIONS
The goals of the focus groups are to explore the participants' understanding of what
human trafficking is, what safe migration is, who is at risk, what motivates and pushes people to
migrate legally or illegally, the role of social networks in influencing and/or pressuring friends or
family members to migrate to look for work, their ideas on the protective factors and preventive
behaviors against human trafficking that can be realistically adopted, and related attitudes about
those behaviors. We're hoping that the results can be used to inform us their contextual
understanding of the problem and adding cultural factors that we have not really touched upon.
The focus groups questions are:
Knowledge on migration
1. What are your expectations when you think about migrating for work abroad?
a. Which countries do you think Indonesian migrants travel for work?
b. What kind of work do you think they do?
2. What about migrating to other cities in Indonesia?
a. Which cities do people travel to for work?
b. What kind of work do you think they do?
c. Do you think there is any difference in whether someone migrates abroad or
elsewhere in Indonesia? If so, which is better?
3. Are you aware of the differences between working abroad legally or illegally?
a. What are the differences?
4. What are the risks when you migrate for work?
a. How might you minimize these risks?
5. What do you think of when you hear the term "safe migration"?
148
On factors that cause people to migrate
6. What are the factors that push people in your community to migrate for work?
(Note to the moderator: try not to mention the points below. Let the participants raise
these issues or other points themselves.)
a. Poverty
b. Family instability? (divorced parents, death in a family, domestic violence,
unexpected debts to pay off, others?)
c. Lack of education & job opportunities (for both men and women?)
d. Desire for more material possessions (what do people want to possess in your
community that they do not already have?)
e. Perception of a better life elsewhere?
f. Success stories from returned migrants or other families who have someone who
migrated for work
7. Which family members have the power to decide who shall migrate for work?
a. What are the main reasons for asking a family member to migrate for work?
b. Do you think daughters have the choice to not migrate for work when asked to do
so by their own parents? Why or why not?
c. What about the wife if her husband asks her to migrate for work?
d. Are birth positions (first born, second born, etc.) a factor in causing a daughter to
migrate for work? Explain why.
e. Who do you think normally approach the parents or the head of the households to
ask their family members to migrate for work?
149
f. What about other people (friends, relatives, teachers, coworkers, neighbors, etc.)?
Recruitment & Decision-Making Process
8. How do people hear about the job opportunities in other provinces? What about in other
countries?
9. What makes people decide to accept a job offer and migrate abroad? What about
migrating to other cities in Indonesia?
10. Who would you trust to give out information about legitimate job offers?
a. What about sources from the media? Can you name specific media sources?
b. Any new media channels like websites, social media, and mobile phone apps?
Knowledge on human trafficking
11. What is human trafficking? Could you tell me some examples of human trafficking?
(The moderator can hand out a piece of paper and ask the participant to write down 3-4
words that come to mind when they thought about human trafficking)
After the participants have answered the first question, the moderator should read the definition
of human trafficking:
Human trafficking is the recruitment and transport of a person with the use of threat, force,
coercion, fraud or deception for the purpose of exploitation. Exploitation includes forced
prostitution, forced labor, slavery or practices similar to slavery, servitude or the removal
of organs.
12. Where can you find human trafficking taking place? (Probe if they think mainly within or
outside of Indonesia. Also probe for specific locations where Indonesians live and work
abroad.)
150
13. What should be the roles of the government on reducing human trafficking?
a. Probe the participants’ attitudes toward the local government officials (which
agencies?) especially on trustworthiness, ability, skills, and willingness to take on
human trafficking
b. Do you think the government could reduce human trafficking if they really
wanted to?
c. If so, why haven’t they done so? (see if they mention kick-backs)
Awareness of anti-trafficking organizations
14. Have you seen any anti-trafficking information in the media in the past 6 months?
a. If yes, can you name any persons or organizations that are working to protect your
community against human trafficking?
b. Have you heard of MTV EXIT?
Counter-trafficking behaviors
15. Imagine you are about to travel abroad for work, what should you prepare to make sure
you're safe?
a. What if they're undocumented or don't have legal traveling documents? What can
these people do?
16. Again, imagine you’re now working abroad and found yourself not being paid by your
employer who forces you to worker much longer hours, and your passport is taken away.
What would you do?
a. What about if this thing happen to you when you are working in other cities in
Indonesia?
17. What can you do to reduce human trafficking in your community?
151
a. What are the barriers that might stop you from performing these behaviors?
b. What kind of skills do you need to be able to do to perform these behaviors?
c. What are the resources needed to perform these behaviors?
Reaction toward Enslaved Documentary
Now, let’s talk about your experience from seeing Enslaved Documentary:
18. What is your impression after watching this?
a. What did you like about the program? What did you dislike about the show?
b. Did the show meet your expectation? Why do you feel that way?
c. What can you remember the most?
d. Did you learn anything new?
e. Is there anything different from your experience? If yes, what is it? How is this
difference important to you?
19. Which elements in the documentary were executed well?
a. Which elements should be improved? Specific examples? (e.g., pace, storyline,
scene settings, character stories, etc.)
20. Do you think the stories are realistic? Why or why not?
a. Which scenes convince you to believe and which part cannot convince you to
believe? Why?
21. Which characters do you identify with the most? Why?
22. Anything you didn’t understand? Irrelevant? Why?
23. What are the main messages in the show that you think the creator wants to communicate
to you?
24. Why do you think the creator made this documentary?
152
25. How would you compare this documentary to other things that you have seen (e.g., news,
films, etc.) about human trafficking?
26. Will you think differently about human trafficking from now? Why?
a. If yes, what will you think differently?
b. Which scenes drive you to change?
c. If no, why not? Are there ways to convince you to think differently?
27. If you had a chance to watch the documentary again, would you watch it? Why or why
not?
28. Will you recommend this to others? Who are they?
29. Any other comments?
Abstract (if available)
Abstract
This dissertation responds to the need to explore knowledge, attitudes, and behaviors related to human trafficking prevention through a multilevel communication perspective. The project is divided into three studies. The first study focuses on estimating the prevalence of human trafficking victims in Indramayu, Indonesia. Results demonstrate that 15.4 percent of Indramayu residents were estimated to be victims of human trafficking. Female migrants also experienced higher prevalence of human trafficking, as well as encountered more restriction on their movement, and exploitative labor practices. The second study evaluates the effects of an anti-trafficking campaign and explores the determinants of human trafficking prevention through a Multilevel Communication Model. Results reveal that campaign exposure was associated with increase in knowledge, perceived risks, skills & abilities, efficacy and intention. Prior experience with human trafficking and migration, and communication ecological factors were found to be strong determinants of trafficking prevention-related outcomes. The third study investigates the role of transportation and identification of the characters in the documentary in changing human trafficking prevention outcomes. Results demonstrate that transportation increased favorable attitudes and intention to prevent trafficking as well as decreased norms to migrate. Meanwhile, identification with characters had differing effects on male and female participants. Specifically, women who identified more with the female characters increased norms to migrate, and had less knowledge, attitudes, and preventive behaviors. Recommendations for future research direction and policy implications on the planning, implementation, and evaluation of human trafficking prevention campaigns are also discussed.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Immigrant health in context: a communication ecology approach to understanding health behavior
PDF
Understanding normative influence of neighborhoods: a multilevel approach to promoting Latinas’ cervical cancer prevention behaviors in urban ethnic communities
PDF
The influence of a humorous intervention on parent-child sexual communication
PDF
Contagious: social norms about health in work group networks
PDF
The power of social media narratives in raising mental health awareness for anti-stigma campaigns
PDF
A spatial analysis of human trafficking in Greater Los Angeles
PDF
From victim to subject: dramatization as research in Thailand's anti-trafficking movement
PDF
Using stories to communicate electronic cigarette risk to young adults: explicating similarity in narrative persuasion
PDF
Intermedia agenda setting in an era of fragmentation: applications of network science in the study of mass communication
PDF
Braver together: exploring the communicative and psychological experiences of LGBTQ youth and families
PDF
Understanding alliance building and communication across multiple technology platforms among California's immigrant-serving NGOs
PDF
Communicating organizational knowledge in a sociomaterial network: the influences of communication load, legitimacy, and credibility on health care best-practice communication
PDF
Hard to argue against coherence: a metacognitive approach to correction of misinformation
PDF
The role of social control in promoting healthy eating behavior among Chinese immigrants: an ecological approach
PDF
Stage of readiness and learning styles -- a tailored HIV/AIDS prevention intervention for youth detained in the Los Angeles County Juvenile Justice system
PDF
"There are enough people here to wage that war:" a communication infrastructure theory approach to identifying factors of anti-gentrification mobilization
PDF
Rediscovering the communication engine of neighborhood effects: how the interaction of residents and community institutions impacts health literacy and how it can be leveraged to improve health c...
PDF
Civic games with 'local fit': embedding with real‐world neighborhoods and place‐based networks
PDF
Exploring geography and multi-level factors associated with disparities in HPV vaccination among Latino adolescents in Los Angeles County
PDF
The impact of engaged communication scholarship on the re-imagination of space and place in South L.A.
Asset Metadata
Creator
Thainiyom, Prawit
(author)
Core Title
A multilevel communication approach to understanding human trafficking prevention behaviors in Indonesia
School
Annenberg School for Communication
Degree
Doctor of Philosophy
Degree Program
Communication
Publication Date
07/26/2018
Defense Date
05/08/2018
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
communication ecology,evaluation, trafficking prevalence,human trafficking,Indonesia,mass media campaigns,narrative persuasion,OAI-PMH Harvest
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Riley, Patricia (
committee chair
), Baezconde-Garbanati, Lourdes (
committee member
), Ball-Rokeach, Sandra (
committee member
), Murphy, Sheila (
committee member
)
Creator Email
pthainiyom@gmail.com,thainiyo@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c89-25112
Unique identifier
UC11670710
Identifier
etd-ThainiyomP-6499.pdf (filename),usctheses-c89-25112 (legacy record id)
Legacy Identifier
etd-ThainiyomP-6499.pdf
Dmrecord
25112
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Thainiyom, Prawit
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
communication ecology
evaluation, trafficking prevalence
human trafficking
mass media campaigns
narrative persuasion