University of Southern California Dissertations and Theses

Network influences in health initiatives: multimedia games for youth in Peru

(USC Thesis Other)

Network influences in health initiatives: multimedia games for youth in Peru

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

NETWORK INFLUENCES IN HEALTH INITIATIVES:
MULTIMEDIA GAMES FOR YOUTH IN PERU

by
Arul I. Chib

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(COMMUNICATION)

August 2007

Copyright 2007 Arul I. Chib
ii
DEDICATION
This dissertation is dedicated to my mother, who shall henceforth
be the only person on the planet required to address me as Doctor.
iii
ACKNOWLEDGEMENTS
This document represents the culmination of five wonderful years of learning
how to situate myself in relation to the world. The program at Annenberg offered me
opportunities I could never have imagined, and provided a richness of experience
that will be difficult to replicate. I attribute all this to the brilliance, understanding
and friendship of the faculty and staff, and to their appreciation and encouragement
of my work.
To Dr. Michael Cody, a particularly warm thanks; for giving me his time, for
his “critical” insights, and for inspiring me to forge a path less-wandered. With the
doors of his office and house wide-open, he welcomed my regular intrusions, offered
guidance in more ways than can be enumerated, and shared his enthusiasm for the
discipline. I also thank the other members of my committee: Dr. Tom Valente, who
introduced me to social network analysis and to Peru; Dr. Francois Bar, who
introduced me to his social network and to the world of ICTs. Their
recommendations and insights have helped me develop as a scholar, and will hold
me in good stead for the future.
I owe a huge debt to Dr. Daniel Aspilcueta of INPPARES for allowing me to
study the youth health program. I would never have accomplished the fieldwork
without the help of Marina Aguilar, the staff and youth volunteers of INPPARES,
and Merli. I treasure their support and friendship while in a foreign land. It is
iv
necessary to remember the contribution of the students of the schools of San Juan de
Lurigancho, whose participation was invaluable.
Thanks are also due to Dr. Peter Monge, whose rigor and scholarly energy I
shall always try to emulate. Dr. Sandra Ball-Rokeach shared her vast knowledge, and
Dr. Peter Vorderer gave generously of his time. For this I am grateful.
I am indebted to the staff at Annenberg for dealing with umpteen glitches,
and for their patience and good humor in chatting with me during my frequent
“walkabouts”. I thank my friends in the program, and in life, for being there each
step of the way.
Finally, I offer thanks and love to my family, the Chibs, for being Chibs. To
my sister, Sonali, for diligently proof-reading this document. And especially to Pa,
for the books.
v
TABLE OF CONTENTS

DEDICATION .............................................................................................................ii
ACKNOWLEDGEMENTS ........................................................................................iii
LIST OF FIGURES ...................................................................................................vii
LIST OF TABLES ....................................................................................................viii
ABSTRACT.................................................................................................................x
Outline..........................................................................................................................1
Chapter 1 ......................................................................................................................2
Introduction..............................................................................................................2
Information Communication Technologies for Development (ICT4D) ............10
Entertainment-Education....................................................................................16
Research Problem...............................................................................................18
Chapter 2 ....................................................................................................................30
Review of the Literature ........................................................................................30
Diffusion of Innovations ....................................................................................35
Social Cognitive Theory ....................................................................................37
Self-Efficacy ......................................................................................................40
Multiple Levels of Analysis...............................................................................43
Social Network Analysis....................................................................................47
Testing Framework and Hypotheses..................................................................52
Chapter 3 ....................................................................................................................61
Methodology ..........................................................................................................61
Study Site ...........................................................................................................61
Study Design ......................................................................................................62
Communication Materials..................................................................................63
Survey Sample ...................................................................................................69
vi
Instrumentation ..................................................................................................72
Chapter 4 ....................................................................................................................79
Results....................................................................................................................79
Descriptive Statistics..........................................................................................79
Reliability and Validity......................................................................................81
Sources of Health Information...........................................................................86
Preliminary Analysis..........................................................................................95
Summary of Hypothesis Testing........................................................................97
Summary of Multiple Regression Analysis .....................................................111
Chapter 5 ..................................................................................................................115
Discussion ............................................................................................................115
Theoretical and Methodological Significance .................................................115
Summary of Major Findings and Implications ................................................118
Limitations .......................................................................................................122
Suggestions for Future Research......................................................................124
REFERENCES.........................................................................................................126
APPENDICES .........................................................................................................139
English-language Questionnaire ......................................................................140
Spanish-language Questionnaire......................................................................149
vii
LIST OF FIGURES
Figure 1. Health Programs at the Schools..................................................................25
Figure 2: Theoretical model.......................................................................................53
Figure 3: Theoretical model for Intergenerational Transmission of Health (Rimal,
2003) ..................................................................................................................54
Figure 4: Proposed Theoretical Model for Health Interventions ...............................56
Figure 5: Co-playing of Computer Multimedia Game...............................................65
Figure 6: Co-playing of Traditional Board Game......................................................66
Figure 7: Filling out of Pre-Intervention Survey........................................................67
Figure 8: Co-playing of Computer Multimedia Game...............................................69
Figure 9. APJ School Advice Network (Gender).......................................................71
Figure 10. Most Important Ways of Obtaining Health Information from Media
and Social Resources..........................................................................................88
Figure 11. Frequency of Obtaining, and Trust in, Health Information from Media
and Social Resources..........................................................................................94
Figure 12. Independencia Americana School Advice Network (Social Efficacy)...106
Figure 13. Independencia Americana School Friendship Network (Social
Efficacy)...........................................................................................................107
Figure 14. Nicolas Copernico School Friendship Network (Attitudes)...................108
Figure 15. Daniel Alomias Robles School Advice Network (Knowledge) .............111

viii
LIST OF TABLES
Table 1. School Attended by Participants..................................................................63
Table 2. Gender of Participants..................................................................................70
Table 3. Ages of Participants .....................................................................................70
Table 4. Descriptive Statistics....................................................................................80
Table 5. Reliability and Item Total Correlation Indices of the Knowledge and
Attitude Scales ...................................................................................................82
Table 6. Factor matrix of Knowledge and Attitudes using Principal Components
Analysis..............................................................................................................84
Table 7. Reliability and Item Total Correlation Indices of the other Scales..............85
Table 8. Factor matrix of other factors using Principal Components Analysis .........86
Table 9. Most Important Ways of Obtaining Health Information from Media and
Social Resources ................................................................................................88
Table 10. Most Important Ways of Obtaining Health Information from Media and
Social Resources for differences between Genders ...........................................89
Table 11. Frequency of Obtaining Health Information from Media and Social
Resources ...........................................................................................................90
Table 12. Independent Samples T-test for Frequency of Accessing Health
Information from Media and Social Resources for differences between
Genders ..............................................................................................................91
Table 13. Trust Health Information from Media and Social Resources ....................92
Table 14. Independent Samples T-test for Trust in Health Information from
Media and Social Resources for differences between Genders .........................93
Table 15. Correlation Matrix for Frequency of, and Trust in, Sources for Health
Information and Knowledge Attitude, and Efficacy variables...........................95
Table 16. Correlations for the Pre-Intervention Sample ............................................96
Table 17. Correlations for the Post-Intervention Sample...........................................96
Table 18. Mean Differences and Standard Error between Pre and Post Samples......98
Table 19. Means Differences and Standard Error between Interactive and Health
game for the Pre-Intervention Sample ...............................................................99
Table 20. Means Differences and Standard Error between High and Low Social
Efficacy Groups for the Pre-Intervention Sample............................................100
ix
Table 21. Means Differences and Standard Error between High and Low Peer
Resistance Efficacy Groups for the Pre-Intervention Sample..........................100
Table 22. Means Differences and Standard Error between High and Low Social
Efficacy Groups for the Pre-Intervention Sample............................................100
Table 23. Means Differences and Standard Error between High and Low Peer
Resistance Efficacy Groups for the Pre-Intervention Sample..........................101
Table 24. Means Differences and Standard Error between High and Low Ease of
Game Play Groups for the Post-Intervention Sample......................................102
Table 25. Means Differences and Standard Error between High and Low Advice
Network Degree Groups for the Pre-Intervention Sample...............................103
Table 26. Means Differences and Standard Error between High and Low
Friendship Network Degree Groups for the Pre-Intervention Sample.............103
Table 27. Means Differences and Standard Error between High and Low
Personal Links of the Advice Network for the Pre-Intervention Sample ........104
Table 28. Means Differences and Standard Error between High and Low
Personal Links of the Friends Network for the Pre-Intervention Sample........105
Table 29. Correlations Between Network Alters and Individuals for the Pre-
Intervention Sample .........................................................................................109
Table 30. Means Differences and Standard Error between High and Low
Knowledge and Attitudes of Alters for the Pre-Intervention Sample..............110
Table 31. Standardized Coefficients, Bivariate and Partial Correlations of
Predictors with Attitudes Factor for the Pre-Intervention Sample ...................112
Table 32. Standardized Coefficients, Bivariate and Partial Correlations of
Predictors with Attitudes Factor for the Post-Intervention Sample..................114

x
ABSTRACT
The spread of contagious STDs, HIV/AIDS, and unintended pregnancies in
developing nations is a source of concern, especially for marginalized youth. This
study examined how information and communication technologies (ICTs) could
bridge gaps in their knowledge and attitudes about sexual and reproductive health.
The dissertation observed the use of a technology-mediated gaming system to
educate Peruvian youths. Working in collaboration with an NGO, Instituto Peruano
de Paternidad Responsable, 108 boys and 111 girls living in the barrios of Lima
were randomly assigned to two conditions, an interactive computer-based
multimedia game and a traditional board game. The research design consisted of pre-
and post-intervention surveys. The study utilized social network analysis to include
social influences in a mixed-influence model.
We find that technology-mediated game playing was equally effective as
traditional health interventions in producing significant improvements in
respondents’ knowledge, attitudes, and self-efficacy. Further, personal peer-
resistance self-efficacy was a positive influence in guarding against developing
negative attitudes. Social self-efficacy was in turn positively correlated with the
individuals’ position in the social network. Finally, the nature of the social link
determined the extent to which one’s associates could influence self knowledge and
attitudes. In this particular case, advice networks were more influential than
friendship networks. A framework for the multivariate relationships, based on
xi
Rimal’s (2003) model, is proposed. This extended the health model, based on social
cognitive theory, to the entire social network. ICTs can provide an advantage, in
terms of innovation, interactivity, and social networking, for use in health
interventions in developing countries.
Keywords: Social cognitive theory, diffusion of innovations, health, HIV,
ICT, network, multimedia, games, youth, Peru
1
Outline
Chapter 1: ICTs and Development
The potential benefits of ICTs in development change
Issues with realizing the potential of ICTs
Research Problem: Sexual and reproductive health gaps in Lima, Peru
ICT as an NGO-based response
Chapter 2: Theory
Entertainment-Education interactive games
Multiple Levels of Analysis
- Diffusion of Innovation
- Social Cognitive Theory and the role of Self-Efficacy
Hypotheses
Chapter 3: Methodology
Study Design: Site, Materials, and Implementation
Participants
Measures
Chapter 4: Results
Hypothesis Testing
Regression Analysis
Network Analysis
Chapter 5: Discussion
Summary of Findings and Implications
Theoretical and Methodological Significance
Limitations
Suggestions for Future Research
2

Chapter 1
Introduction
In recent years there has been growing interest expressed in incorporating
information and communication technologies (ICTs) to achieve development goals
in health, human rights and in social marketing in general. Academic scholars,
governments and advocates of health and civil society, among others, see great
potential in using ICTs to research and influence specific segments of society. For
example, the United Nations Millennium Declaration (United Nations, 2005)
includes ICTs as tools to achieve a number of important development objectives: to
eradicate extreme poverty, to spread basic universal education, to enable gender
empowerment, and to improve various health measures. Target 18 of the Millennium
Development Goals (MDGs) seeks to “make available the benefits of new
technologies, specifically information and communication” (United Nations, 2005,
Pg. 41). The United Nations proposes that the range of ICTs can lead to significant
improvements in these critical measures of development—indeed, the emphasis
remains on development objectives, rather than on the technology (OECD, 2003).
In this first chapter, I will overview arguments in support of the claim that
ICTs will provide an advantage in development work, as well as the
counterarguments that ICTs pose no real solution. I will go beyond the access issue
3
to examine the complex nature of conditions under which ICTs can prove useful, or
fail to achieve objectives, in developing nation contexts. In addition, I will discuss
the extent to which Peruvian adolescents are at risk for STDs and HIV infection, and
what needs to be done to address at-risk behaviors. The chapter concludes with a
description of the response of non-governmental organizations (NGOs) in general,
and the specific ICT intervention that is the focus of this dissertation.
The second chapter describes entertainment–education as a framework, and
the theoretical constructs within which the dissertation study can be situated—social
cognitive theory, diffusion of innovations, and social network analysis. Here,
theoretical implications of the inter-personal interaction of groups and communities
leads to posing research questions and the related hypotheses to be tested. In the third
chapter, I describe the research methodology. The fourth chapter presents the
findings from the data analysis. The final chapter summarizes the findings, discusses
implications, and provides directions for future research.
The potential benefits of ICT in development change
Scholars have defined ICTs from a variety of perspectives, including their
physical forms, the content that they encapsulate, the processes that they enable, by
the actors that they connect, and finally, by the advantages they confer upon users.
The technology itself has been described encompassing a variety of forms, ranging
from advanced modern technologies, such as the Internet, mobile telephony, GPS
4
navigation, computer-based applications, and satellite communication, along with
relatively older technologies such as radio, television, land-line telephones, video
and audio cassettes, multimedia CD-ROM, print, out-door, and theatre, etc. (OECD,
2001). Currently the buzz surrounds the more recent networked technologies,
especially wireless connectivity, as evidenced by a spate of projects involving the
Internet and mobile telephones (W2i, 2003).
ICTs for development have matured greatly in recent times, and projects
relying on these mature forms have been producing noteworthy results. One practice
currently in favor is using entertainment formats of varying levels of technology to
deliver educational information to selected segments of the population.
Entertainment-education (E-E) initiatives are a special case of utilizing ICTs in the
core areas of development. E-E is a strategy whereby entertainment media programs
incorporate educational content to influence and motivate the audience members’
attitudes and behavior, as well as provide knowledge about pro-social issues. Singhal
and Rogers (2002, p. 9) define E-E narrowly as “the process of purposely designing
and implementing a media message both to entertain and educate, in order to
increase audience members knowledge about an educational issue, create favorable
attitudes, and change overt behavior.”
E-E studies (see chapter 2) have primarily examined projects using broadcast
media technologies such as television, radio, and print (Bosch & Ogada, 2000).
5
Clearly, the attempt to achieve positive social behavior and attitude change at a
societal level via the rapid diffusion of information has meant using mass media
technologies. Nonetheless, a number of media forms are in evidence.
Another process-based approach suggests that ICTs can render electronically
a system of “capturing, processing, storing and disseminating information”
(Duncombe & Heeks, 1999, p.2). In contrast to the inclusive definitions of E-E, a
technological focus on ICTs prevents a discussion of the impact of other
communication mediums such as print, theatre, and games. It may be more useful to
utilize a content based approach—referring to the ability of ICTs to handle a variety
of data, including voice, video and text (Chowdhury, 2000); as well as to the
connections that ICTs enable, such as those mentioned by Chowdhury (2000, p. 3),
“among human agents, among humans and information systems, and among
information systems.” Such a definition that focuses on the outcomes that ICTs
enable, rather than on the technology itself, can allow an integration of E-E research
within the broader ICT for development (ICT4D) discipline.
Finally, the tools comprising ICTs are considered not only in terms of their
traditional advantages of high technology, global connectivity, and their interactive
nature (Castells, 2000), but also in terms of their ability to inform and educate (ITU,
2003), and to enhance social communication. Various scholars (Flanagan & Metzger,
2001; Koku, Nazer & Wellman, 2001; Mundorf & Laird, 2002; Haythornewaite &
6
Wellman, 1998) point to the specific characteristics of modern ICTs such as the
Internet that revolutionize the way we communicate. Qualities such as
asynchronicity, interactivity, and the ability to multicast, offer the potential for
enhanced interpersonal communication—opportunities for creating relationships
with a lower sense of social risk, opportunities to test multiple identities, a higher
degree of privacy, a lower sense of accountability, and the ability to strengthen
existing relationships with more frequent communication (Baym, 2001). The
enhanced connectivity, shared storage capabilities, and easing of space and time
constraints delivered by ICTs promises almost instantaneous access to a wealth of
information at relatively low costs (Baym, 2001; Castells, 2001b; Wellman, 2001).
This assumes, of course, that people can easily access these technologies.
Despite the potential, the growth of ICTs is constrained by the existing
communication infrastructure, institutional policies, and economic power of the
nations that the individuals reside in. Even though there may be marked differences
at the intra-national level, with pockets of high ICT development in the midst of dire
poverty, to a certain extent national-level phenomena act as barriers for all residents.
Given these constraints, it would appear that developing countries and the poorest
nations risk being marginalized in the race for globalization.
It may seem that ICTs are being accorded too much importance in
determining the level of development that nations can achieve. Yet, for some
7
(Castells, 2000; Hutton & Giddens, 2000; Rantanen, 2001), globalization is defined
in terms of access to the new communication technologies. Rantanen (2001) suggests
that the effects of globalization are often stated in quantitative measurement: the
more ICTs one country has per inhabitant, the more globalized it is deemed.
Information is indeed the new currency, and access to it is the measure of
development. Bell (1999), from a post-Marxist perspective, argues that theoretical
knowledge has supplanted labor and capital, embodying practical knowledge, as the
fundamental building block of modern economies and societies, compared to the
industrial economies preceding it. Since ICTs increasingly operate as repositories
and conduits of the new form of capital, information, access to them becomes critical
from a development perspective.
Scholars like Castells, Bell, and Ball-Rokeach (1998) agree that modern
information and communication technologies affect both international and domestic
spheres of the economic, social and political framework of nations. This, then, places
ICTs in a critical role for affecting the level of development for any nation embedded
in the inter-connected global system. From a development perspective, this suggests
communication campaigns containing information critical to maintenance of one’s
health would be as important as the delivery of medical services. In reality, no matter
how one views these technologies, making the link between ICT growth and key
indicators of development, especially poverty, has been difficult (Brown, 2001).
8
A key distinction has focused on the causal relationship between ICT
deployment and development (Trujillo, 2003). Some claim that national development
leads to increased usage of ICTs, while proponents of ICT4D argue that usage of
these technologies allows developing nations to catapult over certain sections of the
purportedly linear development path. Still others claim that there is no relationship
between ICT deployment and development. Negating the causal relationships
proposed, nay-sayers such as Chowdhury (2000) claims that “ICTs do not have any
more to do with poverty and food security in the developing countries than rain
dances have to do with rain.”
Furthering the scepticism is the notion that reliance on ICTs would act as a
decisive factor in exacerbating social and economic inequalities. This idea receives
theoretical support from the knowledge gap hypothesis (Tichenor, Donohue, & Olien,
1970), which postulates that given particular systemic change, such as the
introduction of ICTs, certain subsystems within the social system will benefit more
than others since they exhibit patterns of behavior conducive to change. This then
ultimately leads to a situation where certain subgroups adapt more rapidly to the
change, and distance themselves from those who are late adopters or resist adoption.
Unsurprisingly, in a social context, we find that those more capable of
exploiting change for their own benefit are already privileged, be they economic,
educational, or positional advantages. Thus, one would expect that the introduction
9
of new technologies would increase social and economic disparities, rather than
aiding society equitably. As Trujillo (2003) puts it,
Another aspect to consider is the sharp contrasts
that emerge when a hub ... develop at fast rates or
start using or producing ICTs when its surrounding
neighbors do not. If development does not
accelerate with relative evenness in the developing
world, or continues to accelerate even more in the
developed world, then a scenario emerges in which
islands of prosperity are surrounded by oceans of
poverty and frustration. (p. 8)
The disparities in ICT equality are captured in the debate surrounding the
“Digital Divide” (DD): ICTs such as the Internet (on which the debate is largely
focused, and to some extent, computers), are accessible only to a few. The vast
majority of the World’s population has little or no access to the technology, and this
access-gap is increasing over time (Haythornewaite, 2001; Hutton & Giddens, 2000;
Katz, & Rice, 2002; Lievrouw & Livingstone, 2002; Mansell, 1999; Norris, 2001;
OECD, 2001). It might also appear that the issues of the digital divide are not just
about access to communication technologies but deeply rooted in inequities of global
economic distributions and particularly persistent poverty (Servon, 2002). At the
global level, the digital divide is even more pronounced than that recorded in the
United States (UNHDR, 2003). This is further compounded by the suggestion that
social inequities would intensify as the use of the Internet proliferated (Jung, Qiu, &
Kim, 2001).
10
The United Nations Human Development report (UNHDR, 2003) indicates
that per 1000 people, high income countries had 397 Internet users, compared to
merely 6 for the low-income countries. Comparing 2000 data, almost 55% (currently
80%, according to NTIA, 2002) of the United States had Internet access, while the
figures for the Middle East, Sub-Saharan Africa, and South Asia ranged from 0.4 to
0.6 per cent (UNHDR, 2001). Clearly, some regions of the planet, especially the
economically disadvantaged nations, are particularly unable to install and maintain
communication technology infrastructures.
Information Communication Technologies for Development (ICT4D)
While evidence for the direct influence of ICTs in resolving or attenuating
the impact of poverty have been ambiguous so far, certain programs have
demonstrated the potential to improve the lives of marginalized communities by
impacting key programmatic areas such as health, education, livelihood generation,
etc.. The United Nations MDG suggested that ICTs can be used in order to attain a
broad swath of development objectives. Specific health objectives mentioned include
the need to “increase access to reproductive health information, including
information on AIDS prevention, through locally appropriate content in local
languages” (ITU, 2003, p. 81).
Various applications utilizing ICTs are in evidence within the arena of health
projects (Buller, Woodall, Hall, Borland, Ax, Brown, & Hines, 2000; Paisley, 2000;
11
Rogers, 2003). Numerous health communication campaigns have utilized modern
ICTs such as the Internet (Bull, McFarlan, & King, 2001; Buller et al, 2001;
Leiberman, 2001; McKee, Manoncourt, Yoon, & Carnegie, 2000; Rogers, 2004).
The use of ICT in health related projects using personal digital assitants (PDAs) for
data entry and transmission in rural areas includes the Indian Healthcare Delivery
project (Cecchini & Scott, 2003), Jiva’s Teledoc project and CoOptions Technology,
and Uganda’s Satellife program (Donner, 2005). The community-based health
program in Ghana (ITU, 2003) uses interactive centers in order to reduce maternal
mortality. A similar maternal and infant health project (Chib, 2006) uses cellphones
to connect rural midwives to the urban health infrastructure. Cellphones not only
provide voice support during birth complications, but are also a means to collect
health data via the short messaging system (SMS). Educational projects utilizing
ICTs include TARAhaat, whose telecenters are focused on rural India (Prahalad &
Hammond, 2002), and educational radio programs in countries ranging from Mexico
to Thailand (Kenny, 2001).
There has been a distinct euphoria attached to the networked technologies
such as the Internet and mobile telephony. However, less emphasis has been placed
on traditional ICTs such as television, radio and increasingly so, if labels such as
‘old’ can be applied to them; computer applications.
12
Certainly, there are advantages to utilize the Internet for health
communication campaigns and projects. There are three important benefits (Flanagin
& Metzger, 2001). First, the Internet offers the ability to sever hitherto time and
space constraints by delivering, on demand, information not available locally due to
economic, social, legal, and other constraints. Second, the Internet offers an ability to
both deliver targeted messages to specific segments of online users or multicast to a
relatively large number simultaneously. Third, the Internet offers an ability to utilize
the inherent interactivity of the medium to tailor messages. It is also possible that one
will be able to reduce the cost of accessing certain types of information. Nonetheless,
this strategy loses credibility, when a large percentage of the intended recipients of
health campaigns, and possibly, those at highest risk, do not have Internet access.
Further, even if the intended audience does have Internet access, albeit on an
intermittent basis and with poor connectivity, it is quite likely that they do not have
the requisite skills to utilize it properly and effectively. In response to a general
dissatisfaction with using access to technology and time spent using it as unqualified
barometers of social equality. Jung, Qui, and Kim (2001; also see Loges & Jung,
2001), created an Internet connectedness index (ICI) which reveals continuing
inequalities in terms of the intensity and satisfaction of Internet use despite the
narrowing gap in basic access to the technology. Clearly, providing a computer
terminal with Internet access is not going to do away with the complex issues of
literacy, technological familiarity, and perceived competence. Others (Mossberger,
13
Tolbert, & Stansbury, 2003) argue that the DD issue needs to be approached not in
terms of the difference between access among various sub-groups but in terms of
their inherent abilities and backgrounds to use the potential of the Internet to its
maximum.
It would seem that despite the advantages offered by web-based solutions,
their usage may be sub-optimal in the case of the very poor because the poor may be
relatively unfamiliar with, or unable to use, the new technology. As Kenny (World
Bank, 2002) argues:
The nature of extreme poverty in LDCs–very low
incomes, subsistence and unskilled wage labor as
the dominant income source, food as the dominant
consumption good, low education and high
illiteracy, minority language group status and rural
location—points to an unsustainably high cost and
relatively low benefit of direct Internet service
provision through telecenters to the very poor. (p. 1)
Indeed, cost can be a very important factor in determining the
appropriateness of an ICT for delivering social benefits such as health and education.
ICTs differ in the expense associated with them, and costs can be broadly broken
down into fixed start-up costs for infrastructure, and variable costs for running
expenses. Certainly, given state and private investments in the media infrastructure,
technologies such as television and radio have negligible variable costs compared to
connecting to the Internet, or telephony (Wolf, Castro, Navarro, & García 2002;
Dock & Helwig, 1999). It would thus seem that any modern ICT-based intervention
14
would not be sustainable over any extended length of time. However, there are
various initiatives and organizations that seek to tackle the supply-side problem of
high costs. Amongst these, computing technologies can be feasible if utilizing open-
source software (Galperin & Bar, 2006). For one, the cost of duplicating CD-ROMs
is rapidly declining (Proenza, 2002). Though unrealized, there is great public
enthusiasm for the MIT Labs One Laptop Per Child (olpc.com) project.
Notwithstanding barriers such as economic capability, it is only a question of
time when simpler, more accessible, ICTs will replace expensive, infrastructure-
hungry, and training-intense ICTs. Nonetheless, merely providing access is a
technological solution that seems overly simplistic, and that may not solve the
complex developmental issues faced (Keeble, 2003; Servon, 2002), even if it is at a
low-cost (Cecchini & Scott, 2003).
Access to ICTs is not a sufficient condition for improving development
metrics (Cecchini & Scott, 2003). Much depends on the capability of the users to
utilize the ICTs in a productive manner (Gerster & Zimmerman, 2003). Information
systems need to be examined in terms of content—how this translates to the
competencies required in the local communities, rather than traditional methods of
measurement. Prahalad and Hammond (2002) offer anecdotal evidence to suggest
that the poor welcome new ICTs, and quickly adapt to these new technologies.
Indeed, technological proficiency (e.g., computer literacy) is a critical source of self-
esteem and productivity for the audience (Gerster & Zimmerman, 2003).
15
Scholars such as Mansell & Wehn (1998) also caution against the replication
of Western models to developing nations, such as the standard model of ‘one person
– one telephone – one Internet access point.’ Users in economically advantaged
situations tend to have personal computers in multiple locations; their schools,
offices, and homes. Each one of these machines is usually used only by one user,
who is not constrained by having to share computing time with others, or by having
limited access to the technology. Such a user would also have access to educational
institutions, training and maintenance support, and to social resources such as tech-
savvy friends and relatives; all of whom can help the user in mastering the
technology. Clearly, this model would be hard to replicate in any developing nation,
where there are severe economic, infrastructural, personal, and sociostructural
constraints.
When a user in a marginalized community gets access to a computer, s/he
often has to deal with a lack of technological knowledge combined with shared usage
of the limited hardware resource. Without training, the likely outcome would be an
inability to use the technology optimally or to any significant advantage. Nonetheless,
the role of ICTs in development is often seen as a technological solution linked to
global networks, and the barriers most-often cited include infrastructural
insufficiencies at the local level (Cecchini & Scott, 2003). Compounding users’
ability to take optimal advantage of available technological resources are basic issues
related to the provision of electricity, and whether the existing conditions provide
16
adequate storage and periodic maintenance such that the machine hardware can work
more than a few months.
Issues related to technological illiteracy and inadequate infrastructural
support need not be insurmountable problems. For example, the constraint of having
to share a computer can be turned to the advantage of the user. There is a recognition
that shared access models and indigenous production of locally-relevant content
(Mansell & Wehn, 1998) can better serve needs, and in a more sustainable manner.
Users can learn from each other, maintenance can be provided by the community as
a whole, and shared ownership allows variable costs to be defrayed over multiple
users. However, the key issue of developing relevant, enjoyable and easily accessible
content remains. Principles of entertainment-education are outlined in the next
section as one way to develop such content, simultaneously fulfilling community
development needs.
Entertainment-Education
Lack of established modern ICT infrastructures in developing nations has led
social activists to turn to entertainment-education (E-E) products delivered via
traditional mass media, leading to their proliferation across the globe (Singhal, Cody,
Rogers, & Sabido, 2004). Program topics that dominate E-E programming are
health-related, especially with reproductive health, family planning, and the
prevention of HIV /AIDS, and gender empowerment. Music and soap operas were
17
the predominant media utilized in these E-E projects. These programs predominantly
use mass media ICTs such as radio, television, and print. These projects for social
change rely on an existing infrastructure. This then makes them economically viable,
given public accessibility bordering on negligible variable costs (Kenny, 2001).
There are non-economic advantages to using relatively mature technologies
such as radio, television, and increasingly, computing facilities. These include the
engaging nature of the content, the ability to consume or participate in groups in
addition to solo usage, and their ubiquitous reach in parts of the developing world.
Shared viewing or usage of ICTs is a particularly important feature from a theoretical
perspective of increasing project efficiency, as we shall examine in the next chapter.
ICTs can improve lives and harness the power of social networks by
enhancing communication without necessarily requiring Internet connectivity or
cutting-edge technological solutions. The scope of the use of ICTs in development,
and entertainment-education, is broadening from utilizing just the mass media to
include both more traditional participatory, such as folk theatre, and modern
interactive formats, such as multimedia technologies. This method approaches the
development model with an eye towards providing equality of advantage. The
argument is that introducing social change is not merely a top-down process initiated
by institutions controlling the means of production, and projecting out to a passive
yet pliant audience.
18
This dissertation examines both indigenous and popular media formats such
as board games in addition to modern communication technologies, such as
multimedia computer-based games. Skuse (2001) suggests that a mix of modern
ICTs and traditional media can be used collaboratively in order to achieve desired
social change.
Since pro-social objectives may not parallel the profit motives of
corporations, economic resources are essential in extending the benefits of ICTs to
those currently without access. Yet, money and access are not sufficient to translate
the advantages of ICTs into tangible benefits for the poor. An evolving partnership,
of governments and non-governmental bodies (Hawkins, 2002), needs to work hand-
in-hand with technology providers to solve problems holistically. Directed
development programs using ICTs are one such solution among a multi-pronged
multi-lateral effort. Thus, it will not be critical to make the crucial choice between
medicines or ICTs, or textbooks or ICTs; rather, the question is how to use ICTs
effectively in order to deliver better health and better education to those that need
them.
Research Problem
Issues related to sexual and reproductive health are global in magnitude.
Particularly relevant to developing nations is the spread of contagious sexual
diseases such as the Acquired Immunodeficiency Syndrome (AIDS) and social
19
problems arising from unintended pregnancies (UIP) (Chirinos, Salazar, & Brindis,
2000). AIDS is a disease resulting from the spread of the Human Immunodeficiency
Virus (HIV), and is indiscriminate, affecting everyone, regardless of their age,
gender, ethnicity, nationality, or social status. According to UNAIDS (2005), the
AIDS epidemic continues to ravage the planet, with three million people succumbing
to the disease annually, and an estimated 40 million people living with HIV/AIDS
(PLWHA). The disease continues unabated, with an estimated five million new cases
diagnosed each year (UNAIDS, 2005).
There are 1.8 million PLWHA in Latin America. In Peru specifically, the first
case of AIDS appeared as early as 1983. Currently, official reports from Peru of
PLWHA estimate that there are 23,657 persons with HIV, with an additional 17,678
AIDS cases (Ministry of Health, 2005). However, outside of the governmental health
system, estimates from experts are considerably higher—ranging from 76,000 to a
possible 144,328 (CIA World Factbook, 2005; Kusunoki, Gunaira, Navarro, &
Velásquez, 2005; Paho.org) Peruvians affected by HIV/AIDS. UNAIDS (2005)
estimates an annual 5,600 deaths due to the disease nationwide.
Despite the fact that AIDS affects everyone, researchers have found certain
populations, such as women, adolescents, and children, to be particularly vulnerable
(Kusunoki et al, 2005), possibly due to biological and social factors. The increased
20
vulnerability of women is reflected in the shift in the male/female ratio of HIV/AIDS
cases in Peru from 11/1 to 3/1 in 2000 (Paho.org).
While the most common form of transmitting the virus is via sexual contact
in Peru, accounting for 95.7% of the cases (Paho.org), there are social and religious
taboos surrounding the discussion of sexuality. There is distinct opposition to
discussing issues regarding sexual health, and especially to talking about the use of
modern contraceptives such as condoms, which is one of the key pillars of educating
the public about protecting themselves from contracting HIV. This has led to a
situation in which there is low awareness about HIV/AIDS and transmission of the
virus, with a significant part of the resistance coming from powerful political and
religious institutions (Hardee, Agarwal, Luke, Wilson, Pendzich, Farrell, & Cross,
1999). Of note is the fact that Roman Catholics comprise 81% of the population
(CIA World Factbook, 2005). Papal authorities continue to repeat misinformation
about the effectiveness of condoms, as part of an overall strategy to discourage usage
(Bergsjø, 2002). Thus, even if one were to achieve widespread awareness of
condom effectiveness, the translation of this knowledge to actual usage is in practice
undermined, especially in the vulnerable youth segment.
UNICEF suggests that young adults comprise the bulk of HIV/AIDS
infections, and that most of them represent an enormous health risk, being unaware
of their carrier status. This fact is vitally important in combating the disease, as
21
seventy per cent of the Peruvian population is under the age of 25. This represents a
significant and vulnerable segment of the population. UNAIDS (2005) reported that
there has been a steady rise in AIDS cases among 20 to 24 year olds for some years.
This suggests that teenagers are the ones contracting the virus, and highlights the
importance of reaching and educating teenagers about HIV and associated risks. The
estimated adult HIV prevalence rate (15+ years) in Peru in 2005 was 0.6.
Further, some scholars (Loli, Aramburú, & Paxman, 1987; Singh & Wulf,
1991) suggest that Peruvian adolescents are at risk of experiencing the negative
consequences of their not practicing safe sex. Estimates of modern methods of
contraceptive prevalence are 50 percent (Globalis.com), although total contraceptive
prevalence, including traditional remedies, may be as high as 64 percent (Hardee et
al, 1999). However, only 31% of sexually active adolescents have used a modern
method of contraception (Chirinos et al, 2000). Further, given the social and
religious stigma attached to pre-marital sex, there is a psychological barrier to
seeking sexual and reproductive health advice or contraceptives. Studies (Chirinos et
al, 2000) confirm that for many Peruvian male adolescents, knowledge about
sexuality is limited, with the consequence that risks of contracting an STD or causing
an UIP are quite high.
These vulnerabilities increase when adolescents are living in conditions of
social exclusion. Marginalized communities are defined not only in economic terms,
22
and in terms of their relative access to ICTs, but also in terms of their vulnerability to
external crises, including those arising from natural, social or economic forces
beyond their control. This leads to a powerlessness on the part of communities to
make or influence decisions that affect them, and results in their exclusion from
relevant decision-making debates and from the information-holding sections of
society (Gerster & Zimmerman, 2003).
In addition, there is a cyclical relationship between the spread of HIV/AIDS,
and marginalization (UNDP, 2001). Contributing factors such as poverty and
underdevelopment increase the risks associated with contracting the disease.
Conversely, the presence of an infected member has a high likelihood of contributing
to keeping the family in poverty. In Peru, 54.1% of the population live below the
poverty line (CIA World Factbook, 2005). Therefore, educating and enabling young
people living in economically disadvantaged situations is crucial for improvements
in sexual and reproductive health, and especially so in the fight to halt the spread of
HIV/AIDS. To address these issues, several state-sponsored and non-governmental
organizations (NGOs) have attempted to introduce, upgrade, and expand the health
services, and information received by marginalized communities.
In Peru, a number of governmental institutions organized by the Ministry of
Health, in conjunction with academic, religious and non-governmental organizations
address issues of sexual and reproductive health. Despite these efforts, the total
23
health expenditure per capita reported for 2004 was merely $235, compared to USA
at $6096. Even compared to a developing nation like Mexico ($655), and a country
within the region, Brazil ($1519), the expenditures per capita for health in Peru are
quite low. Nonetheless, the figure, given the context of the developing economy
status of Peru, is understandable, as it still comprises 4.4 percent of the gross
domestic product (WHO, 2005). Consequently, the health infrastructure finds it
difficult to deliver much-required services with an estimated density of only 1.17
physicians per 1,000 members of the population, with the nurses being even below
that at 0.067 per 1,000 members of the population (WHO, 2005).
Response
There are a number of established NGOs in Peru trying to fill the gap
encountered in the delivery of health services. However, results from the government
working in conjunction with the non-governmental sector are inconclusive (Hardee
et al, 1999). NGOs favor a more demand driven approach than governments, and
seek to collaborate with communities that they work with in a participatory
framework (Gerster & Zimmerman, 2003). Thus, compared to government services,
NGOs can leverage their role as grass-roots intermediaries working at the local level,
in conjunction with local leaders, fostering a sense of local ownership and
participation, to meet the needs of marginalized communities (Cecchini & Scott,
2003).
24
Since Peru is a predominantly Catholic country, the local leadership is
resistant to addressing issues relating to sexuality. For groups such as teenagers, the
constricting social atmosphere may not only be a barrier to receiving information
about sexual and reproductive health, it also acts as a deterrent to seeking out such
information from relevant health resources. Thus, if only a minority of teenagers
arrive at clinics, then NGOs in the health sector are compelled to reach out to them.
The scholastic infrastructure can serve a critical role in delivering health
programs to marginalized youth (See Figure 1). Schools can be an important partner
for health NGOs due to reliable access to this segment of the population (Bandura,
2003). Schools also provide access to instructors trained in delivering information.
However, in the case of sexual and reproductive health, the problems that apply to
priests hold true for teachers as well. Teenagers are no more likely to discuss their
sexual issues in a classroom setting than they are to bring up these issues with the
local clergyman. Sex educators are cautious and guarded, making oblique references,
leading to increased confusion for students, and possibly to a greater ignorance of the
facts (Bandura, 2006). One response to this predicament by NGOs has been to
eliminate the mediator, replacing instructors using various strategies, including peer
volunteers and game-based formats for health information delivery.
25
Figure 1. Health Programs at the Schools

The literature reveals that games oriented around imparting health
information are widely used. In a world increasingly dominated by the storage of
information in electronic formats (Bandura, 2006), it is no surprise that computer-
based interactive games are being employed to provide more engaging content to
26
teenagers. The benefits of games will be reviewed in the next section. However,
major criticisms leveled at the usage of ICT based games have revolved around the
unfamiliarity of the technology, the great expense that it would involve providing a
computer to each beneficiary, and the long-term sustainability of such programs.
There are different ways to examine the expense and sustainability of these
health issues, including cost-benefit comparisons, engagement with corporate
partners, a subsidy-based vs. a pay model, etc. However, computing technologies are
advancing in schools, even in developing countries. A number of instances of this
abound, including the yet-to-be launched MIT Media Labs One Laptop Per Child
project. Further, the idea of shared usage dramatically changes the cost assumptions
of such approaches. One such set of programs, according to Servaes (1999) engages
community networks, an information resource using computer networking to
“reinvigorate the health and well-being of local communities”. These community
programs allow for economic viability with their ability to “disaggregate access from
ownership” (Prahalad & Hammond, 2002, p. 11), thus allowing for shared usage
across multiple members of the community.
While there are a number of information communication technologies for
development projects being conducted around the globe, there has been, and yet
remains, a paucity of documented evaluation reports about specific projects. Much of
the evidence is anecdotal, or at the global level of analysis, with national-level data,
27
and has in most cases, only recently begun (ITU, 2003). This dissertation aims to
provide scientifically grounded evidence to answer the question—can an ICT4D
health intervention be successful? Further it looks at how such an intervention can
compare versus a traditional method, again game-based, to deliver the same
objectives.
The local programs run by NGOs tend to emphasize equality of access, while
simultaneously delivering culturally relevant social services. Health organizations
aim to provide relevant information using a variety of appealing communication
mediums. These may be both traditional games as well as modern ICT-based
interactive games. The NGO chosen in this study developed activities based on both
these formats.
The Instituto Peruano de Paternidad Responsable (INPPARES), a health
agency based in Lima, Peru, provides information services to youth regarding sexual
and reproductive health. INPPARES is a member of the International Planned
Parenthood Federation (IPPF), which is a globally recognized health provider, with
establishments in over 140 nations.
A key area of emphasis for INPPARES is the youth segment, as it provides
information services regarding sexual and reproductive health, especially the
prevention of STDs such as HIV, and the elimination of obstacles to contraceptive
access. One of the programs targeted at youth is the “Yes® Educational Stations”.
28
This health campaign aims to involve the community in developing healthy sexual
behaviors in youth. One of the ICT-based components of this strategy is developing
multimedia games for educating youth about basic reproductive and sexual health,
including transmission of HIV and other STDs, usage of contraceptives, and other
safe-sex behaviors.
This study, in collaboration with the Instituto Peruano de Paternidad
Responsable, Peru, evaluates the effectiveness of multimedia materials in enhancing
participants’ knowledge about these subjects, and their enhanced ability to control
personal reproductive health choices. It compares these interactive materials with a
traditional board-based game delivering the same information. Note that while
INPPARES works in a broad spectrum of health-development related activities, yet
ICT4D is a new area of community support for them. Thus, their interest in
participating in a research project stems from the fact that evaluating ICTs as an
effective tool can improve their service delivery objectives.
Determining the mode of delivery of these technology-based health services
requires an understanding of the ICT infrastructure of the country. According to the
World Development Indicators database (Globalis.com; World Bank, 2004), in terms
of computing, 10.5% of Peruvians use the Internet; however, personal computers
aggregate only 5.2 per 100 people. This is due to the fact that Peru, exemplified by
the capital Lima, has seen an explosion in public cabinas, or Internet access centers.
29
ICT expenditure as a percentage of GDP is 6.9%, and only 3% of schools are
connected to the Internet. Thus, for an NGO to conduct an ICT-based intervention
targeting school students, computer-based and traditional games are better choices
than the Internet.
In this special case of ICT4D, entertainment-education, the literature suggests
(Sood, Menard & Witte, 2004; Witte, 1996) that the purposeful design of media
messages in the field is often conducted in an atheoretical manner. There is therefore
an urgent need (IDS, 2005; Trujillo, 2003) to transfer “best practices” learned in one
area to other areas and potential barriers across regions, and to have scientifically
designed evaluations, aided by sound theoretical fundamentals, at the field level to
judge program effectiveness. The next chapter explores the theoretical basis for
conducting ICT4D projects as health interventions, and proposes a framework to
analyze their effectiveness.
30
Chapter 2
Review of the Literature
In this chapter, I review research on entertainment-education and introduce
the fundamental theories underlying this dissertation, diffusion of innovations
research and social cognitive theory. I then define self-efficacy, discuss social
network analysis, and the need for conducting analysis at multiple levels. This
chapter ends with a presentation of a model integrating theories and proposing
hypotheses.
Entertainment-education offers the means to create positive social change.
This communication form should have tremendous value in a world that is
increasingly reliant on media technologies, and therefore understanding of its
underlying mechanisms has increasing import. As outlined in the previous chapter,
the impact of traditional communication technologies (television, radio, and print)
has been studied in detail, and there is an increasing emphasis on newer multimedia
technologies.
First, multimedia computer-based games have wide-ranging appeal for youth
(Gee, 2003; Vorderer, 2000). However, much of the research on the effects of games
has focused on the adverse impact that these games might have on youth (Anderson
& Dill, 2000; Anderson & Bushman, 2001; Sherry, 2001, 2007). Other scholars have
suggested that games may be beneficial to adolescents (Durkin & Barber, 2002),
31
especially in the arena of learning (Gee, 2003; Linderoth, Lindström, &
Alexandersson, 2004).
While most research has focused on commercial games, a few scholars have
begun to focus on the use of interactive games in the area of health communication
(Bandura, 2004; Brown, Lieberman, Gemeny, Fan, Wilson, & Pasta, 1997;
Lieberman, 2001; Thomas, Cahill, & Santini, 1997). These are specifically designed
to do more than entertain; in fact, to achieve identifiable health outcomes. These
games are not typical of their commercially available counterparts—with lower
production qualities in features such as resolution, fidelity, game-play, and usability
(Rizzo & McLaughlin, 2006). Yet whether or not their quality and feature-sets are
the same as store-bought games, the key questions revolve around their effectiveness
in meeting health goals, and about the long-term sustainability of such E-E
multimedia interventions.
Research Questions 1 and 2
Do health games lead to improvement in health knowledge and attitudes?
Are interactive entertainment-education games as effective as traditional games in
meeting health objectives?

There are two important limitations to traditional work in Entertainment-
Education. First, the research in E-E has concentrated on the effects of the
interventions rather than on developing valid theory explicating the mechanisms
32
behind the effects. Singhal and Rogers (2001) acknowledge that E-E research has
focused on formative evaluation, campaign monitoring and summative evaluation of
effects. There are historical precedents to this emphasis; since E-E as a strategy for
social change was discovered serendipitously, (Singhal & Rogers, 2001, 2004) as
summarized in the effects of the radio serial drama, Simplemente Mariá.
The emphasis on outcomes is probably due to the evaluatory requirements of
institutional donors that fund these interventions. This leads to a practitioner-bias
focusing on evidence of programs that work to ensure future funding from sponsors.
Despite this bias, not all the evidence found is positive. In fact, evaluating only the
benefits of socio-medical innovations, Mosteller (1981) found that of 28 innovations,
less than half (12) had positive results, and three even had negative outcomes.
The second problem with E-E research is its reliance on linearity of effects at
the individual level (Papa, Singhal, Law, Pant, Sood, Rogers, & Shefner-Rogers,
2000; Rogers & Kincaid, 1981). An examination of the E-E literature suggest that
the vast quantitative studies in the field have focused on individual-level behavior
(Brodie, Foehr, Rideout, Baer, Miller, Flournoy, & Altman, 2001; Cody, Fernandes
& Wilkin, 2004; Collins, Elliott, Berry, Kanouse, & Hunter, 2003; Kane, Gueye,
Speizer, Pacque-Margolis, & Baron, 1998; Piotrow & de Fossard, 2004; Rogers,
Vaughan, Swalehe, Rao, Svenkrud, & Sood, 1999; Vaughan & Rogers, 2000). This
focus on the individual level fails to recognize the influence of the community and
personal social relationships found in traditional social structures.
33
The process of social change brought about by the mass media is far more
complex, and involves interaction between members of the social system, in addition
to the direct effects paradigm. It is essential to understand this meta-level, i.e., the
role of the group, or community, in creating an atmosphere wherein learning can
occur, and further, where behavior modifications can succeed such learning.
As the field of E-E grows more sophisticated, scholars seek to understand the
effects on multiple levels . Some recent efforts (Bandura, 2004; Papa et al, 2000;
Rogers, 2003; Sharf, Freimuth, Greenspon, & Plotnick, 1996; Valente, 2005; Valente
& Saba, 1998) have been devoted to examining the processes at the level of the
community or social network, in addition to individual level effects.
A better understanding of the “process” involved in modern forms of E-E,
such as interactive multimedia games, may help improve programmatic success and
increase the chances of successfully meeting social objectives. A robust explication
of the mechanisms underlying the E-E experience can allow the development of
engaging programs with educational content that are, most importantly, effective.
Therefore, to shed light upon the mechanisms operating during the experience of
engaging with an E-E multimedia product, we need to turn to the theoretical models.
A word of caution is in order before proceeding. It has been suggested that a
methodology needs to be specific, sufficiently complete and clear in order to be
useful (Abrami, Cholmsky, & Gordon, 2000; Kerlinger & Lee, 2000). Others are of
the contrasting view that it would require integrating multiple theoretical
34
perspectives to explain complexity observed in multi-level phenomena such as the
one undertaken. Monge and Contractor (2003, p. 21)) suggest that “utilizing multiple
theories should ... significantly increase the amount of variance accounted for by
these theoretical mechanisms”. Using multiple theories to develop testing models can
therefore pose a problem to the researcher. One way to achieve simplicity and clarity
is to develop models that bring together the commonalities, and complementarities,
between theories.
Certain theoretical constructs were chosen to study here for three reasons: (1)
they link the experience of exposure to the subsequent effects; (2) they both include
as a process whereby information, attitudes, and ideas are spread; and (3) they deal
with phenomenon at the inter-personal level, such as relationships between
individuals, in addition to the psychological effects at the intra-personal level. While
these theoretical constructs are derived from earlier work (Bandura, 2003; Rogers,
2003), synthesizing them into an integrative framework and developing a
methodological tool to test them, via the relatively new field of network analysis, is
where this dissertation hopes to contribute to the field.
The purpose of this research is twofold. First, it aims to develop an
integrative model that links the mechanisms of the E-E experience to the educational
learning or subsequent effects. To do this it draws upon the theoretical constructs of
(1) diffusion of innovations (Rogers, 2003), and (2) social cognitive theory (Bandura,
1999, 2006). Secondly, the research proposes a methodological framework to test
35
such a model. This framework uses social network analysis to understand the
relationship between individual-level effects and the more methodologically
challenging inter-personal/social level of analysis.
Diffusion of Innovations
Behavior change theories have been used in a variety of health
communication studies (Elder, 2001; McKee, Manoncourt, Yoon, & Carnegie, 2000;
Rice & Atkin, 2001; Valente, 2002). These include the transtheoretical model
(Prochaska, DiClemente, & Norcross, 1992), the hierarchy of effects model
(McGuire, 1989) and diffusion of innovations (Rogers, 2003). One fundamental
construct in these theories is that effects occur linearly, with media affecting
knowledge, attitudes and practices (KAP). At a basic level, these KAP-based models,
also known as the learning hierarchy model, suggest that knowledge and belief
change occurs first, prior to behavioral change, and that knowledge and attitude
changes lead to positive behavior change, such as engaging in beneficial practices.
Others (Valente & Saba, 1998) challenge the linear progression implied in the KAP-
model, suggesting instead that products may often be tried without knowing exactly
how they work.
Diffusion of innovations research (Rogers & Shoemaker, 1971; Rogers,
2003) focuses on “overt behavior change”, as opposed to just change in knowledge
or attitudes, which are considered intermediate steps in the process. Rogers describes
36
this process in an innovation-decision model with five stages—that of knowledge,
persuasion, decision, implementation, and confirmation.
Rogers’ diffusion model (2003) classifies members of the social system
based on their innovativeness, or degree to which an individual is relatively earlier to
adopt an innovative idea than other members. The classification includes, in order of
adopting an innovation—innovators, early adopters, early majority, late majority,
and laggards. The rate of adoption is initially slow, with few responding to an
unfamiliar idea. As these early adopters get comfortable with the innovation, word-
of-mouth spreads the benefits, and the innovation adoption occurs at an increasing
pace. At this point, the majority of the social network adopts the new idea or practice.
Finally, with most people adopting, over time the rate of pace slows. Finally, the
laggards adopt, and the pace of adoption tapers off asymptotically to a point where
there are usually some non-adopters within the system. A cumulative frequency of
adoption results in the traditional S-shaped curve.
Rogers’ diffusion research model is fundamentally concerned with
understanding the processes whereby innovations are adopted and spread, and with
understanding the mechanisms underlying how innovations, broadly defined, are
diffused through a social system. Social change is said to have occurred within a
complex social environment—via interaction, experimenting with the new project,
its adoption, and the imitation in audience members from innovators to laggards
(Rogers & Kincaid, 1981). Further, early adopters are influenced by media and/or
37
external forces, while late adopters tend to gather information through social contact
with those who have already adopted (Rogers & Shoemaker, 1971). The model
considers effects at multiple levels – individual, interpersonal, and the community
network level.
Note that the current study does not measure adoption of beneficial behavior
due to the time constraint between exposure to the multimedia game and
measurement, and due to social constraints on asking teenagers about their sexual
behavior. However, the focus here is on the change in knowledge and attitudes, as
captured by specific measures of sexual and reproductive health. The process of
modeling change, via direct influence of the multimedia game and social influences,
is nonetheless expected to occur.
Social Cognitive Theory
Social cognitive theory (SCT) (Bandura, 1986, 1999, 2001, 2002, 2003,
2004) finds its roots in social learning theory, which dates to early developments in
the disciplines of behavioral and social psychology. SCT has formed the theoretical
backbone of many health interventions, especially those focused on adolescents
(Schwarzer & Luszczynska, 2006) since the initial academic interest in the
phenomenon. SCT suggests that the extent to which behaviors are retained will
depend on personal, environmental, and social determinants such as exposure to the
modeled outcomes, perceived personal and collective self-efficacy, and compatibility
between social and self-sanctioned behavior. This theory addresses the mimetic
38
learning based on observation of models, which relies heavily on punishments and
rewards. Social models may be “similar models” or “prestigious models” in
entertainment programs, and are either rewarded for engaging in positive behaviors
or punished for engaging in undesirable, bad, behavior. Ideally, the “good” or
desirable behavior will be emulated by the audience, while the bad behaviors are
rejected. Simply put, behavior is environmentally conditioned through observation
and enhanced by perceiving reward-punishment outcomes.
The theory does not imply a deterministic system, since individuals are
agentic, meaning that they retain the freedom to exert control over their own lives.
Importantly from the perspective of entertainment-education, people are capable of
both creating and using symbolic communication, and this allows them to learn
vicariously. Individuals utilize their cognitive abilities to assign symbols, which
allow adaptation and mastery over the environment. Symbols are used as a means of
communication, and for the expansion of knowledge by acting as a repository of
information. Information can thus be processed by matching possible courses of
action versus ones’ store of encoded information.
Vicarious learning, or learning through observation, allows for faster and less
tedious learning than that acquired through the trials and errors of direct experience.
This ability has added significance in modern times due to the increased penetration
of communication vehicles and increased global connectedness that allows for a
39
rapid diffusion of ideas. The notion of vicarious learning is an important one for E-E,
as media models are created with the purpose of acting as exemplars.
Thus, for high-risk behaviors such as those involving sexual and reproductive
health decisions, vicarious learning can be extremely beneficial. Rather than
exposing youth to potentially life-risking situations, behaviors such as using a
condom or negotiating safe-sex practices can occur through media modeling. In the
case of societies that stigmatize certain knowledge, such as that related to sexuality,
an interactive multimedia game can perform the role of the model, without the need
for potentially embarrassing human interaction. This then can be categorized as a
case of abstract modeling (Bandura, 2002), in which individuals develop behavioral
rules based on observation of social exemplars, such as media characters.
Theoretically, these rules are then replicated in one’s own actions across different
situations.
According to SCT, individuals also have self-regulatory and self-reflective
abilities. The former acts as a motivational mechanism affecting one’s decisions to
take action. This ability also acts as a moral indicator that inhibits inhumane
behavior and promotes humane action. The verification process for self-regulation is
theorized to occur via four routes of enactive (self-thoughts vs. results of action);
vicarious (self-thoughts vs. results of others actions); social (self-thoughts vs. others
thoughts); and logical (deductive reasoning). The study under examination is
designed to utilize the vicarious and social mechanisms of self-regulation in realizing
40
beneficial sexual and reproductive outcomes. While playing multimedia games
facilitates vicarious learning, social self-regulation arises from the act of co-playing
among participants.
Finally, the capability of self-reflection acts through a comparison of one’s
thoughts versus observation of reality. It is expected that the knowledge gained by
playing the games will allow youth to engage in a cognitive evaluation of the
outcomes. A successful health campaign will allow the individual to choose adoption
or rejection of the behavior in the future, based not only on their self-reflection, but
on the knowledge internalized prior to engaging in the behavior.
Self-Efficacy
According to Bandura (1986, 1997), perceived self-efficacy is what one
believes one can do, with what one has, and under various circumstances. Perceived
self-efficacy is an important mechanism in explaining adoption or non-adoption.
However, self-efficacy is not a generalized concept. Rather, it depends on whether
the function or task at hand is specific or broad in nature. People differ in their
perceptions of efficacy in various domains. For example, a student who is
particularly proficient in math may not necessarily feel extremely confident in her
ability to play field hockey. This implies that the measurement of efficacy should be
assessed within a specific domain to have relevance.
In the entertainment-education literature, there are a number of examples of
self-efficacy and collective efficacy predicting desired behaviors (Agha, 2003;
41
Rogers et al, 1999; Papa et al, 2000; Singhal, & Rogers, 2004; Singhal, Sharma,
Papa, & Witte, 2004; Soul City, 2001). Self-efficacy and collective efficacy are
thought to influence each other, and are both influenced by interpersonal discussions
and media exposure.
Vicarious experience is one source of self-efficacy information (Bandura,
1986, 2003), which suggests that the influence of social models can be felt through
their impact on an individual’s self-efficacy. Observing successful performances by
others leads to increased self-efficacy in mastering similar activities. This is
heightened for activities that the individual has little or no prior experience—which
holds true of teenagers entering their sexually aware period. Peer influence,
operationalized in the multimedia health game as a co-playing experience, has been
found to be significant in the area of sexual and reproductive health information
(Bandura, 1997; Schunk & Meece, 2006).
Modeled performance, both in real and virtual situations, is based on social
comparison processes (Bandura, 1986, 1997). Individuals compare the similarity of
models in terms of age, gender, race, and socio-economic characteristics. The
assumption is that similar personal characteristics would predict similar performance
attainment. Further, the literature suggests that credibility of the information source
plays a significant role in changing self-efficacy. Thus, for teenagers, there is a
tension inherent in treating peers as credible sources of health information, possibly
42
due to the realization that they have limited experiences and competences in matters
of sexuality.
Nonetheless, Bandura (1986, p. 438) argues “that perceived self-efficacy
mediates health behavior” (also see Rimal, 2003). It seems logical that people would
engage in beneficial health behaviors when they consider themselves capable of
executing them successfully. Numerous studies confirm this relationship in
theoretically-based health interventions (Amir, Roziner, Knoll, Neufeld, 1999; Li,
Mcauley, Fisher, Harmer, Chaumeton, & Wilson, 2002; Bradley, & Corwyn, 2001;
Jemmott, Jemmott, Fong, & McCaffree, 1999; O’Leary, 1992; Ott, Greening,
Palardy, Holderby, & DeBell, 2000; Prochaska et al, 1992).
Bandura (2004) proposes the use of four stages to design effective health
prevention programs—information delivery, development of social and self-
management skills, increasing perceived self-efficacy, and creating a social support
system. He suggests that educational institutions focus on the first, and by ignoring
subsequent stages, fail to provide youth with the requisite skills or a robust sense of
self-efficacy needed to protect the individual from adopting harmful health practices.
Thus, it is not only important to improve knowledge and attitudes, but also to
increase perceived self-efficacy. Previous studies on the use of contraception
demonstrates that efficacy beliefs help the individual manage relationships as well as
intentions to practice safe sex (Kasen, Vaughan, & Walter, 1992). On the other hand,
43
drug and alcohol use lowered the perceived self-efficacy, and thus raised the
possibility of engaging in high-risk unprotected sex.
At the group level, Bandura (2002) recognizes the importance of multiple
levels with his notion that exposure also engenders collective efficacy, or the degree
to which individuals in a system believe that they can organize and execute courses
of action required to achieve collective goals
Research Question 3
Do higher levels of self-efficacy positively impact key health indicators such as
knowledge and attitudes?

Multiple Levels of Analysis
It seems that both these theoretical approaches, diffusion of innovations
research and social cognitive theory, acknowledge the dual pathway through which
change may occur—the direct path, and that mediated by inter-personal networks.
This direction continues in the tradition of the two-step flow of communication
(Katz, 1957; Katz & Lazarsfeld, 1955). It has long been theorized that while an
external influence like the mass media leads to awareness, actual adoption occurs via
influence from interpersonal contact. Katz’s hypothesis states that ideas flow from
the mass media to “opinion leaders and from these to the less active sections of the
population” (Katz, 1957, p. 61). Rogers (2003) broadens this role to include specific
individuals with varying degrees of ability to influence change in the diffusion
44
framework, other than opinion leaders, such as change agents and aides. Bandura
(2002) echoes the thought when he states that
In short, there is no single pattern of social influence.
The media can implant ideas either directly or through
adopters. Analyses of the role of mass media in social
diffusion must distinguish between their effect in
learning modeled activities and on their adoptive use
and examine how media and interpersonal influences
affect these separable processes. (p.142)
Singhal and Rogers (1999) contend that prompting interpersonal discourse
about educational content is one of the main processes through which E-E
interventions influence adoption of behavior. There is sufficient evidence that E-E
programs engender interpersonal conversations (Collins et al., 2003; Rogers &
Kincaid, 1981; Rogers et al, 1999; Valente, Kim, Lettenmaier, Glass, & Dibba, 1994;
Vaughan, Regis, & St. Catherine, 2000; Vaughan & Rogers, 2000), which in turn
lead to some form of attitudinal and behavioral change. However, comparative
testing of the mixed influence of mass media and interpersonal communication, and
the process by which this occurs, has rarely been clarified in much detail, or linked to
theoretical constructs.
Given proof of network influences operating, in terms of interpersonal
discussions in the community, it seems evident that educational content embedded in
entertainment programs generates and stimulates discourse in communication
networks. As with much of the effort in E-E, practitioners have seized upon concepts
and utilized them before they are fully explicated by researchers. Practitioners in
45
health diffusion have mobilized community networks as part of an effort to develop
integrative entertainment education campaigns. Storey and Jacobson (2004), for
example, describe a population program in Nepal that focused on community
activities in an effort to harness the network level discussions that aid decision-
making. Data from the Twende na Wakati campaign (Rogers et al, 1999) show that
the intervention “had strong behavioral effects on family planning adoption; it
increased listeners’ self-efficacy regarding family planning adoption and influenced
listeners to talk with their spouses and peers about contraception” (p.193).
The importance of inter-personal discussions, especially for negotiating
sexual practices or relationships, is acknowledged. However, we need to understand
the processes whereby these information exchanges aid in the diffusion of adoption
through a communication network. We first need to examine the explication of these
mechanisms by the theoretical models under analysis. However, the underlying
mechanisms producing these outcomes have yet to be fully addressed.
According to Rogers (2002), while exposure was an important determinant, it
was less important in influencing the adoption of the target behaviors than was
interpersonal communication with a friend, or more importantly, with a spouse or a
sexual partner. Indeed, Rogers (2003) believes that the mass media act primarily in
diffusing information, and it is the interpersonal networks that are “more important
in persuading individuals to adopt or reject” (p. 305).
46
Social cognitive theory acknowledges multiple levels of analysis (Bandura,
1997, 2006). However, Bandura contests the view that interpersonal communication
channels are more effective than the media in diffusing new ideas (Klapper, 1960;
Sherry, 2002). Bandura (2002, 2004) contends that there is an inter-related pattern of
influence through which media systems influence social change, with mass media
modeling impacting individuals directly in conjunction with being mediated by
social networks.
However, it may not be sufficient to isolate the independent effects of the
media at the personal level, and the inter-personal effects at the social level. Indeed,
personal psychological and sociostructural factors need to be integrated into an
integrated framework (Bandura, 2006). Bandura (1986) calls for determining the
interdependencies “between personal and environmental influences” (p. 29). The
nature of the social network, roles occupied by various members of the network, and
their individual nodal level attributes; all contribute to the inter-dependency of
influences. Further, the social structure can influence one negatively; thus impeding
positive outcomes from health campaigns (Bandura, 1986; Chib, 2005)
Research Question 4
What is the role of interpersonal networks in affecting key health indicators such as
knowledge and attitudes?

47
It is clear that both theories, diffusion of innovations and SCT, address the
influence of behavior-adoption at multiple levels—individual and inter-personal. We
can utilize a multi-level approach (Monge & Contractor, 2003) to understand the
dynamics of the interactions among and between media models, audiences, and a
person’s social networks. Rogers (2003) seems to advocate a network approach to
these two theories when he addresses their common points. He says that
Both theories seek to explain how individuals change
their overt behavior as a result of communication with
other individuals. Both theories stress information
exchange as essential to behavior change, and view
network links as the main explanation of how
individuals alter their behavior. (p. 342)
While diffusion of innovation has sparked new interest in network terms as
computational abilities have advanced and become more sophisticated (Monge &
Contractor, 2003; Rogers, 2003; Valente, 2003), there is a need to integrate these
theories into a framework that increases understanding of the links between
experience and effects, and that functions at multiple levels. To do so we turn to the
heuristic of social network analysis.
Social Network Analysis
Testing a multi-level framework requires a methodology that allows for
theoretical integration at different levels of analysis. Social network analysis (SNA)
is particularly useful for this (Knoke & Kuklinski, 1986; Scott, 2000; Wasserman &
Faust, 1994; Wellman & Berkowitz, 1988), since it is inherently multi-level, with the
48
possibilities of collecting and analyzing data at different levels of analysis—
individual, dyadic, triadic, subgroup and network. Recent advances in multi-level
modeling techniques allow for examination of higher levels of analysis, like network
centrality, while simultaneously controlling for lower level phenomenon, such as
nodal attributes (Monge & Contractor, 2003; Valente, 2005).
Network analysis is capable of capturing the principal mechanisms of social
influence processes. The central premise is that links between individuals give rise to
a structure with particular non-random properties. These properties arise from
relational and structural characteristics (Burt, 1987; Freeman, 1979; Scott, 2000;
Valente, 1995; Wellman & Berkowitz, 1988). Relational properties examine the
links between individuals, while structural properties examine the individual
positions within the network as well as properties of the network itself, such as
centralization.
SNA can measure the personal networks under consideration for
transformation; for example, health advice, friendship, sexual relations, etc. in order
to better understand the communication flows, and influence flows from, according
to the diffusion literature, early adopters. The type of communication link determines
the type of influence that one can have. For example, an individual’s friendship
network may be very different from a health advice network. Individuals who
occupy prominent positions in the former network may be in the periphery or totally
excluded from the latter. The question is whether individuals get health information
49
from their regular everyday friends, or from people who have specific expertise in
the topic (Bandura, 1986).
Research Question 5
– Does type of communication link influence key health indicators such as
knowledge and attitudes?

Knoke & Kuklinski (1986) suggest that this phenomenon can be explained by
a structural argument. They believe that advantaged individuals, such as early
adopters, are more likely to occupy central positions in the given networks, and thus
have greater opportunities to access information flows within it (Robertson, 1971),
and act upon them earlier than others. The attitudes and behavior of individuals who
occupy positions of prestige and influence in the community should have great
impact on others in the social network to which they are directly or indirectly
connected (Bandura, 1986).
Some scholars (Bandura, 1986; Rogers & Kincaid, 1981) find evidence that
those with greater tie density will be earlier adopters of innovation. Various reasons
to explicate this phenomenon include greater social connectedness leading to better
information, increased awareness of other adopters, and inherent receptiveness to
innovations by those individuals with many social ties.
ICTs then become critical for diffusing new knowledge to, and influencing
behaviors of, populations where there are individuals who lack personal networks
50
comprising adopters, and who by virtue of their network positions, are continuously
being disadvantaged. The fact that personal network structures play an important part
in adoption and that ICTs can compensate for those individuals with less structurally
advantageous positions, suggests that social promotion campaigns using ICTs may
have an important role to play.
Network analysis can thus explain greater variance by examining various
influences upon the change process, and in the case of longitudinal studies, their
varying contributions at different time-points. Monge and Contractor (2003)
emphasize the complex structure of networks by noting the inter-dependence of
endogenous and exogenous variables. According to them
Endogenous variables are characteristics of the
relations within the network that are themselves used
to explain the structural tendency of that relation.
Exogenous variables are characteristics of the network,
other than the relation itself, that are used to explain
the structural tendencies. (p. 55)
This multi-level approach of network theorizing allows for testing theories
structurally—endogenous and exogenous—and at various levels of analysis. Thus,
hypotheses can be formulated and tested for a single theory (or multiple theories) at
different levels of analysis. For example, social cognitive theory can test hypotheses
at the exogenous individual level (i.e., an individual with a higher educational
background has a greater likelihood of self-regulating his/her own behavior), or test
hypotheses at the exogenous dyadic level (i.e., individuals have a greater likelihood
51
of adopting similar attitudes as others with similar attributes). A structural argument
at the endogenous dyadic level would distinguish itself from the latter argument by
hypothesizing that individuals at similar positions in the network have a greater
likelihood of sharing similar attitudes.
Innovativeness, stated in network terms, can also be distinguished on the
basis of whether individuals are innovative with respect to their personal social
system or whether they are innovative with respect to the entire network (Valente,
1996). Threshold models of diffusion of innovations (Abrahamson, 1997;
Granovetter, 1978; Rogers, 1995; Valente, 1996; Valente & Saba, 1998) argue that
individuals have varying thresholds of adoption. This model argues that standard
exposure may not be enough for an individual to adopt, and different personal
thresholds may explain the variance in different times taken for people to adopt after
exposure.
While the effects at the individual, interpersonal, and social system level are
studied within the model, we need to extend the model to consider influence on
adoption by targeted health campaigns, such as interactive multimedia games, an
external force that influences the realization of the network.
However, in interpreting this result one should bear in mind the criticism
from Singhal and Rogers (2002), who indicate that much of the research has
involved monitoring mainly individual-level, short-term behavior changes. Clearly, a
problem might arise in measuring immediate changes in knowledge, attitudes and
52
self-efficacy without considering a host of barriers that might actually hinder
adoption and performance of benefical behaviors. Clearly, in this case, systemic
issues such as unavailability of birth control methods, lack of support from the social
structure; and personal issues such as resistance from one’s sexual partner may be
more of a problem for vulnerable segments such as teenage girls.
The traditional measures of social network analysis and diffusion analysis
rely on self-reports from individuals of their communication patterns. The concern
surrounding this methodology is the transient nature of memory, such that these
reconstructions may or may not be the most reliable. One solution is to observe the
true nature of communication links; yet this is not practical in field research. The
current study aims to augment the social structure as elicited by respondent recall
with actual observation of co-playing of games. This is done in order to understand
the influence of observed communication pattern, and therefore objective, versus the
more subjective measures of the structural links provided by participants. It should
be noted that this is a mere snapshot in time of the social communication, and
therefore should not be expected to substitute for more complex, time-bound
interactions.
Testing Framework and Hypotheses
A general framework of social influences in health interventions, shown in
Figure 2, is based on the theoretical discussion.
53
Figure 2: Theoretical model

This is derived from a model of intergenerational transmission of health
shown in Figure 3, (Rimal, 2003), that aims to integrate interpersonal, intrapersonal
and communicative factors. Rimal’s model aims to determine the health behaviors of
children as influenced by their parents.
Discussion
Health
Information
Knowledge
Self
Efficacy

Distal variables

Demographic
variables

Culture

Individual
differences
Behavior/
Attitudes
Knowledge
Self
Efficacy
Intrapersonal Influence
Interpersonal Influence
Behavior/
Attitudes
54
Figure 3: Theoretical model for Intergenerational Transmission of Health (Rimal,
2003)

The model developed in this dissertation aims to incorporate social influences
beyond the paired parent-child relationship developed by Rimal (2003). Thus
influences from multiple alters in the individuals immediate environment are
hypothesized to influence the individual. The corresponding influences occur for
knowledge, self-efficacy, and attitudes.
In summarizing the theoretical discussion, the model in Figure 4 posits
effects at the individual, interpersonal, and social system level. The model assumes a
dual pathway, via direct influence and interpersonal networks, to influencing
55
individual attitudes and behavior via the mass media. Specifically, self-efficacy is
thought to mediate between the knowledge and attitudes, although knowledge also
has a direct impact on attitudes. Information about health from various sources
influences knowledge, self-efficacy, and attitudes. Interpersonal discussion
influences attitudes directly. Finally, the knowledge, self-efficacy, and attitudes of an
individual’s alters influence the individuals knowledge, self-efficacy, and attitudes
respectively.
56
Figure 4: Proposed Theoretical Model for Health Interventions

Discussions
Information
from Media
Behavior/
Attitudes
Knowledge
Self
Efficacy

Distal variables

Demographic
variables

Culture

Individual
differences
Information
from Social
Sources
Social
Efficacy
Discussions
Information
from Media
Behavior/
Attitudes
Knowledge
Self
Efficacy
Information
from Social
Sources
Social
Efficacy
Interpersonal Influence
Alter Effects
Self Effects

Distal variables

Demographic
variables

Culture

Individual
differences
57
Given such a framework, the next step is to test the relationships that have
been developed. Statistical analysis and social network models are methodologies
used for structuring the relationships under consideration into a testable format. The
following hypotheses test these relationships. In light of the significance of this
study, the research problem, and the review of the theoretical literature, the following
hypotheses were tested:
1. For all respondents under both conditions (interactive game and board game),
there will be a positive impact on key indicators of health knowledge and
attitudes, and efficacy measures. Specifically, the hypothesis states that there
will be:
a. An increase in the variable Knowledge pre- and post-intervention.
b. A decrease in the variable Barriers pre- and post-intervention.
c. An increase in the variable Attitudes pre- and post-intervention.
d. An increase in the variable Condoms pre- and post-intervention.
e. An increase in the variable Social Efficacy pre- and post-intervention.
f. An increase in the variable Peer Resistance Efficacy pre- and post-
intervention.
2. The interactive health game will be more effective than the traditional game
when measured by key indicators of health knowledge and attitudes, and
efficacy measures. Specifically, the hypothesis states that comparing the
interactive computer games and the traditional board game, the interactive
game will have:
58
a. A greater increase in the variable Knowledge pre- and post-
intervention.
b. A greater decrease in the variable Barriers pre- and post-intervention.
c. A greater increase in the variable Attitudes pre- and post-intervention.
d. A greater increase in the variable Condoms pre- and post-intervention.
e. A greater increase in the variable Social Efficacy pre- and post-
intervention.
f. A greater increase in the variable Peer Resistance Efficacy pre- and
post-intervention.
3. Respondents with high efficacy will have greater health knowledge and
attitudes than low efficacy players.
a. Respondents with high Social Efficacy will have greater Knowledge
than those with low efficacy.
b. Respondents with high Social Efficacy will have lower Barriers than
those with low efficacy.
c. Respondents with high Social Efficacy will have greater Attitudes than
those with low efficacy.
d. Respondents with high Peer Resistance Efficacy will have greater
Knowledge than those with low efficacy.
e. Respondents with high Peer Resistance Efficacy will have lower
Barriers than those with low efficacy.
f. Respondents with high Peer Resistance Efficacy will have greater
Attitudes than those with low efficacy.
4. Respondents who find greater familiarity with the game will have greater health
knowledge and attitudes than those are less familiar with the games.
59
a. Respondents with a higher Game score will have greater Knowledge
than those with a lower Game score.
b. Respondents with a higher Game score will have lower Barriers than
those with a lower Game score.
c. Respondents with a higher Game score will have greater Attitudes
than those with a lower Game score.
5. Players with more central positions (centrality) will have better health
knowledge, attitudes, and self-efficacy compared to players on the periphery.
a. Players with more central positions will exhibit greater Knowledge
compared to players on the periphery.
b. Players with more central positions will exhibit lower Barriers than
compared to players on the periphery.
c. Players with more central positions will exhibit greater Attitudes
compared to players on the periphery.
d. Players with more central positions will have greater Social Efficacy
compared to players on the periphery.
e. Players with more central positions will have greater Peer Resistance
Efficacy compared to players on the periphery.
6. Players with greater links in the network will have more health knowledge and
attitudes compared to players with fewer links.
a. Players with greater links in the network will exhibit better
Knowledge compared to players with fewer links.
b. Players with greater links in the network will exhibit lower Barriers
compared to players with fewer links.
c. Players with greater links in the network will exhibit better Attitudes
compared to players with fewer links.
60
d. Players with greater links in the network will exhibit greater Social
Efficacy compared to players with fewer links.
e. Players with greater links in the network will exhibit greater Peer
Resistance Efficacy compared to players with fewer links.
7. The knowledge and attitudes of alters in both advice and friendship networks
will influence the knowledge and attitudes respectively of the self.
a. For the advice network:
i. Players whose alters in the network possess greater knowledge
will exhibit greater Knowledge than players whose alters
possess lower knowledge.
ii. Players whose alters in the network possess greater attitudes
will exhibit greater Attitudes than players whose alters possess
lower attitudes.
iii. Players whose alters in the network possess greater self-
efficacy will exhibit lower Self-Efficacy than players whose
alters possess lower self-efficacy.
b. For the friendship network:
i. Players whose alters in the network possess greater knowledge
will exhibit greater Knowledge than players whose alters
possess lower knowledge.
ii. Players whose alters in the network possess greater attitudes
will exhibit greater Attitudes than players whose alters possess
lower attitudes.
iii. Players whose alters in the network possess greater self-
efficacy will exhibit lower Self-Efficacy than players whose
alters possess lower self-efficacy.
61
Chapter 3
Methodology
The survey data were collected in collaboration with the Instituto Peruano de
Paternidad Responsable (INPPARES). The intervention was conducted in early 2005
in the neighborhood of San Juan de Lurigancho, a suburb of Lima.
Study Site
It is estimated that 70% of HIV/AIDS infections are concentrated in the
urban and semi-urban coastal settlements of Lima and Callao (UNAIDS, 2005). San
Juan de Lurigancho is the most populous of Lima’s districts, and therefore Peru’s,
with a population estimated to be exceeding a million people. Municipal estimates in
2000 indicate a population of 831,634 inhabitants (sanjuandelurigancho.com), more
than half of whom live in shanty-town conditions within new squatter settlements
(Baffigo et al, 2001). The terrain is hilly, arid and inhospitable, with high population
density, and much of the economy depends upon self-employment, with limited
opportunities available in the formal sector. Domestic conditions can be inadequate,
with limited availability of running water, inadequate hygienic sanitation, and
sometimes non-existent transportation infrastructure. Unsurprisingly, the area is
crime-ridden, with slums mushrooming, and home to the two largest prison facilities
in Peru. According to the Comisión de la Verdad y Reconciliación Oficina de
62
Comunicaciones e Impact, more than 80,000 residents of San Juan de Lurigancho
were displaced due to political violence between 1980 and 2000.
Given this backdrop, the expectation for delivery of health services can
hardly be high. However, local health centers run by the government and non-
governmental health agencies exist, though often the community has to play a
significant part in the delivery of these services. One morning I arrived at a school to
administer the pre-intervention survey only to find that all the students had been sent
to parade the local neighborhood with banners informing the populace how to
minimize the spread of dengue fever. At another time, some students were
summarily summoned by the municipal health official to the local health center. It is
such conditions that provide the principal rationale for choosing this area to conduct
the intervention and study.
Study Design
The study was planned as a health communication intervention in
conjunction with the outreach centers of INPPARES in the district of San Juan de
Lurigancho. The intention was to replicate the health interventions implemented by
INPPARES in the normal course of operations. Thus, the design would be to involve
only those activities that were conducted within the INPPARES framework.
Secondly, the design was to preclude all those activities that had low generalizability,
and could not be replicated on a wider scale.
63
The study design consisted of a quasi-experimental design of a pre- and post-
intervention health program with a control group with panels chosen from four
neighborhood schools. The schools were chosen randomly from a list of schools that
formerly participated in health interventions run by INPPARES. Municipal records
indicate that there are 650 schools in San Juan de Lurigancho. The four
neighborhood schools that participated in the health intervention were Colegio
Amistád Perú – Japón (APJ), Colegio Daniel Alomias Robles (DAR), Colegio
Independencia Americana (IA), and Colegio Nicolás Corpenico (NC). The classes
chosen to participate in the study were first selected by the school administration
based on age eligibility criterion, and then from this pool were randomly selected by
the researchers (See Table 1).
Table 1. School Attended by Participants
School Frequency Percent
Colegio Daniel Alomias Robles 66 30
Colegio Independencia Americana 52 23.6
Colegio Nicolás Corpenico 50 22.7
Colegio Amistád Perú – Japón 52 23.6

Communication Materials
The materials used in this study were created by INPPARES, in conjunction
with a team of youth volunteers. INPPARES develops games providing information
64
about sexual and reproductive health, including the transmission of STDs, proper
usage of contraceptives, family planning, and other related information. The games
chosen for this study were designed, created and distributed by INPPARES.
Prior to beginning the study, it was decided to compare the effectiveness of
traditional game-based health interventions with more modern computer-based
multimedia material. Youth volunteers, guided by INPPARES personnel, develop
and implement a variety of outreach activities, including workshops, film and theatre
viewing, personal counseling, and school-based programs. As part of this program, a
number of multimedia materials have been created, two of which were chosen for
this study. The two games chosen by me were based on their similarity of content,
but dissimilarity of delivery. The multimedia computer-based game was called
Planeta Riesgo X and the traditional board game was called Inteliyes!!.
Once the four schools had been selected, the choice of game distribution was
chosen randomly. INPPARES staff preferred to place the computer-based game in
schools with better computing facilities.
Participants at two schools, Colegio Independencia Americana, and Colegio
Nicolás Corpenico, were exposed to a multimedia CD, Planeta Riesgo X, which aim
to provide such information (See Figure 5). In the CD, two cybernetic youth search
for new adventures, in the course of which they encounter situations dealing with
sexual and reproductive health. Based on their responses, players are guided to
65
specific health information. They are also free to explore various sections of the CD
to seek information in a more dynamic manner.
Modeled performance was designed into the interactive game with players
choosing screen avatars based on gender. These avatars were then exposed to critical
sexual relationship-relevant situations, and the players were expected to make
decisions. After a round, the players were given the right answers and allowed to
continue to information sections. Here players gained information to fill gaps in their
sexual and reproductive knowledge.
Figure 5: Co-playing of Computer Multimedia Game

66
Participants at the other two schools, Colegio Amistád Perú – Japón and
Colegio Daniel Alomias Robles, were exposed to a board game, Inteliyes!! (See
Figure 6). In this competitive board game, four players play to advance their tokens
based on rolling dice. At each roll of the dice, a player is presented with the
opportunity to answer questions from a card that has four distinct areas of knowledge
about sexual and reproductive health—A, B, C, and D. Depending on the location of
the player’s token on the board, the competing players ask him one of these four
questions. The cards are then discarded, and play continues to the next player. The
game ends when one player advances all of his or her tokens to the winning area.
Figure 6: Co-playing of Traditional Board Game

67
Procedures
A pre-intervention survey was administered to all sections in the participating
schools (See Figure 7).
Figure 7: Filling out of Pre-Intervention Survey

Prior to administering the survey, the study coordinators guided the group
through a practice exercise to familiarize the respondents with the use of the scales.
A practice rating scale, based on Bandura’s (2006) Guide for constructing self-
efficacy scales, was administered to familiarize students with the efficacy rating
form. Two approximately equal class sections were selected at each school. After
68
two weeks, at each school, both sections either played the CD game or the board
game. Since the study design divided the schools into control and test schools,
approximately half the subjects played on the computer with the CD game preloaded,
while the other half played the board game (See Figure 8). The post-survey was
administered immediately after the participants played either the CD game or the
board game. The pre- and post-intervention surveys contained identical questions,
except for a change in the order of questions to eliminate risks to testing validity.
Questions about the game-playing experience were added to the post-intervention
survey. Additionally, the baseline survey contained questions about a respondent’s
friendship and advice networks, while the co-playing networks of respondents were
recorded via observation during the health intervention. None of the subjects were
compensated for participating in the study.
69
Figure 8: Co-playing of Computer Multimedia Game

Survey Sample
Two hundred and forty-nine students, aged 15 -18 years old, of both genders
participated in the study. 49.3% of the sample was female, while 50.7% was male.
The median age was 16, which comprised 37.6% of the sample. The largest group
was aged 15 years old, comprising 45.1% of the sample. Those 17 years old were
14.6% of the total, while those above 17 years of age comprised a mere 2.8% of the
70
sample. Tables 2 and 3 present the frequency data for age and gender of the
participants.
Table 2. Gender of Participants
Gender Frequency Percent
Male 108 49.3
Female 111 50.7
Table 3. Ages of Participants
Age (Years) Frequency Percent
15 96 45.1
16 80 37.6
17 31 14.6
18 4 1.9
19 2 .9

The gender links in the social network suggest that there is no strict gender
bias, as both males and females connect with each other. This is illustrated in the
social network sociogram from the Amistád Perú Japón school (See Figure 9).
71
Figure 9. APJ School Advice Network (Gender)
Female
Male

There were no subject enrollment restrictions. 220 students completed the
baseline survey, of which 180 participated in the implementation by playing the
game. Two hundred and nine students, of which 29 had not completed the baseline
survey, participated in the intervention, after which they were administered the post-
intervention survey.
72
Instrumentation
The questionnaire solicited responses from participants on their media system
access for health information, their knowledge and attitudes regarding sexual and
reproductive health issues, and their perceived self-efficacy. Participants also
identified individuals in their social networks they believed to be their friends and
the individuals with whom they most discussed health issues. The questionnaire was
designed to be non-specific for sexuality, allowing for answers by both heterosexual
and homosexual respondents. All questions were close-ended, using a variety of
Likert-type scales. The questionnaire was designed by me in English, and then
translated by the INPPARES staff to Spanish. A reverse Spanish-to-English language
translation was generated to check for errors. A copy of the survey instrument is
included in the appendix. Details of the confirmatory factor analysis are provided in
the Results chapter.
Media System Access: A series of questions assessed the respondent’s
sources of information and communication resources, probing frequency of media
usage, ease of access, trust, and frequency of obtaining health information. The
information and communication resources tracked specifically included the Internet,
radio, newspapers, magazines, cinema, brochures, friends and family, fellow-
students in the class, doctors and health workers, and health organizations.
The main usage frequency question asked was, ‘In the past 7 days, how often
did you access … ’, and used a 7-point scale. The ease of access question, ‘How easy
73
or difficult is it for you to access …’ used a 10-point Likert-type scale that ranged
from ‘Very Difficult’ to ‘Very Easy’. A series of questions about sexual and
reproductive health were led by the usage question, ‘How often have you obtained
health information in the past 30 days from … ’, which used a 4-point Likert-type
scale.
This was followed by trust in communication resources, which was, ‘How
much do you trust the information about health from … ’. Media coverage of certain
health issues, such as family planning, HIV/AIDS, sexuality and contraception were
probed with the question, ‘In the past 30 days, how much have you heard from the
media about … ’. Additionally, a specific question related to the issue of media
portrayals of sexuality was addressed by the question, ‘How much sexual content do
you think there is on …’. These 3 questions utilized a 4-point Likert-type scale,
where the choices ranged from ‘Not at all’, A little’, ‘Some, to ‘A lot.’
Knowledge of family planning methods: The knowledge questions measured
the respondent’s awareness of issues involving sexual and reproductive health. The
questions covered two broad areas; knowledge of family planning methods, and
HIV/AIDS transmission. The first series of questions established awareness of a
variety of family planning methods, such as the condom, sterilization, the pill, IUDs,
injections, implants, the rhythm method, withdrawal, and traditional methods. The
items were dichotomous yes and no answers. The latter eight items were summed
together to create a family planning knowledge factor named Knowledge. Since the
74
knowledge of condoms was as high as 96% for the sample, the item was deleted
from the factor, and treated individually as a variable, Condoms.
Attitudinal Measures: The attitudinal questions measured the respondent’s
feelings surrounding the issues involving sexual and reproductive health. Several
indicators captured respondent’s attitudes towards condom usage. The respondents
were asked to respond on a Likert-type scale from 1 to 7, where 1 meant ‘strongly
disagree’ and 7 meant ‘strongly agree’. Six questions addressed the use of condoms
at reducing the chances of contracting the virus: ‘I will try and use condoms if I have
sex in the future,’ ‘I am able to use condoms to prevent HIV/AIDS,’ ‘Using condoms
prevents HIV/AIDS and sexually transmitted diseases,’ ‘Using birth control methods
to prevent pregnancy is good,’ and ‘Condoms help to plan the family and protects
from AIDS.’ An additional item addressed the role of social networks in dealing with
protection from HIV/AIDS: ‘My friends think it is a good idea to protect against
HIV/AIDS.’ The resulting summed 6-items measured Attitudes.
Barriers towards practice of safe sexual and reproductive health habits:
Exploratory factor reduction and inter-item reliability analysis suggested that the
following items, that might prevent people from adopting safe sexual habits, could
be combined to form a single factor. These statements required agreement on a 7-
point Likert-type scale with whether ‘family planning methods cost too much’, were
‘inconvenient’ or ‘hard to get’, whether it was ‘embarrassing to use family planning
services’, and the ‘difficulty in discussing pregnancies within the community’. These
75
five items were summed together in order to create a variable that measured
hindrances to family planning called Barriers.
Efficacy Measures: Schwarzer and Renner (2006) suggest, based on a
longitudinal study in Germany, that ‘health-specific self-efficacy is significantly
related to corresponding health behaviors’ (p. 14). They studied the relationship between
self-efficacy measures of preventive nutrition, physical exercise, and alcohol resistance
with corresponding behaviors to find that there are significant correlations.
The efficacy questions were designed to measure two distinct domains of
self-efficacy—self-regulatory efficacy, and social self-efficacy. These scales are
based on scales developed by Bandura (1990, Choi, Fuqua, & Griffin, 2001; Miller,
Coombs, & Fuqua, 1999), titled Multidimensional Scales of Perceived Self-Efficacy
(MSPSE). Respondents were asked to mark their degree of confidence by recording
a number from 0 to 100. A practice exercise was scheduled to familiarize the
respondents with this type of scale prior to administering the questionnaire.
The questions about self-regulatory self efficacy try to understand the extent
to which respondents can resist peer pressure to smoke cigarettes, drink beer, wine,
and liquor, smoke marijuana, and engage in sexual intercourse. The summative 4-
item scale for the measure of self-regulatory self-efficacy was named Peer-
resistance efficacy.
The social self-efficacy measure addressed four questions: whether
respondents could ‘make and keep friends of the opposite sex’ (Note that this one
76
item that did not capture homosexual relationships) , ‘carry on conversations with
others’, ‘work well in a group’, and ‘get a friend to help me when I have social
problems’. The summative 4-item scale for this measure of self-efficacy was labeled
Social efficacy.
Interpersonal Dialogue: Another measure to understand the impact of
respondents’ social networks included a 3-item-scale labeled Talk. The respondents
were asked to respond on a scale from 1 to 7, where 1 meant ‘strongly disagree’ and
7 meant ‘strongly agree’, how much they agreed whether they could talk to their
sexual partner(s) about family planning methods, use condoms, and prevent
HIV/AIDS.
Playing Measure: This 5-item scale was designed to measure the
respondent’s attitude towards the game they had played. The respondents were asked
to respond on a Likert-type scale from 1 to 7 (1 = ‘strongly disagree,’ 7 = ‘strongly
agree’). The questions ranged from ease of use and access, to usefulness, trust and
recommendation of the game to others. The specific questions asked were ‘I trust the
information contained in the game’, ‘I find it easy to get access to information by
playing the game,’ ‘I find it easy to use the game according to my needs,’ ‘I find the
game to be useful for health information,’ and ‘I will recommend the use of the game
to my friends.’. These five items were summed to create a game-playing measure
labeled Game.
77
Data Analysis
The data were analyzed using different analytical approaches listed below,
and each hypothesis was analyzed separately. The first hypothesis was tested using a
paired sample t-test to test for differences in the knowledge, attitude, and efficacy
variables between the pre- and post-intervention samples. The second hypothesis
used the paired sample t-test to determine whether the interactive game sample had
greater impact on key health indicators than that of the sample that played the
traditional board game. The third hypothesis was analyzed using independent
samples t-tests to check whether different levels of efficacy would produce greater
health knowledge and attitudes. The fourth hypothesis used a similar procedure to
test the same hypothesis for different levels of ease of game play.
Social network analysis was used in order to determine the levels of
knowledge, attitudes, and self-efficacy of alters in individuals’ networks. These were
then averaged for each individual. Further, social network measures such as degree
centrality, density, and links were determined. Hypotheses 5, 6, and 7, were tested
using these social network measures, using independent samples t-tests.
Multiple regression analyses were run with the attitude variables as the
independent variables. The regression analysis was carried out separately for the pre-
intervention sample and the post-intervention sample. The dependent variables for
the pre-intervention sample were the knowledge and attitude variables, Knowledge
78
and Barriers, the discussion variable, Talk, and the efficacy variables, Social, Peer-
resistance. For the post-intervention sample, in addition to these variables, the
independent factor for ease of game-play, Game, was also regressed on the Attitude
variable. In both cases, the regressions were controlled for the effects of age, school
and gender.
79
Chapter 4
Results
The purpose of this study was to examine the impact of mediated and
network influences on improving health knowledge and attitudes. The results of the
statistical analyses, including descriptive statistics, reliability, exploratory factor
analysis, confirmatory factor analysis, correlations, t-tests, regression, and social
network analysis, are presented in this chapter. First, the chapter outlines the
frequency distributions, means and standard deviation for the sample. This is
followed by factor loadings for the latent variables. Cronbach’s α (alpha) for internal
consistency was used for the reliability analyses. Spearman’s rank-order correlation
coefficient r
s
(rho) was used in preliminary analyses to test the strength of
relationships between the latent variables. Both paired-sample and independent
sample t-tests were conducted in order to test hypotheses. The visual heuristic of the
sociogram was used to illustrate findings from the statistical analyses.
Descriptive Statistics
The descriptive statistics for the seven composite scales used in the analysis
are presented in Table 4. Further, the item means were computed by dividing the sale
means by the number of items, and the standard deviations associated with them are
also reported. It should be noted that the Barriers scale is viewed negatively, i.e. the
higher the value, the greater the barriers towards usage of family planning methods.
80
The results show that the item means of the Attitudes scale (M = 6.02, SD = .22) is
high, while the Barriers scale (M = 3.08, SD = .10) is low. This suggests that
respondents generally hold positive attitudes, while the perceived barriers are low,
although not negligible. In terms of awareness of methods of family planning,
knowledge of condoms (M = 0.96, SD = .20) is already quite high. However,
awareness of other methods represented by the composite scale of Knowledge (M =
0.66, SD = .01) is much lower. Finally, respondents’ self-efficacy in seeking social
resources, the Social Efficacy scale, (M = 75.95, SD = 22.44) showed the highest
item mean score compared to the scales measuring their ability to resist peer pressure,
the Peer Resistance Efficacy scale, (M = 70.48, SD = 1.97). This implies that
respondents seek out social support, but are not as good at resisting the pressure put
on them by their friends and colleagues.
Table 4. Descriptive Statistics
Variables Scale Scale Item
# Mean SD Mean Variance
Knowledge of Family Planning
Methods
0-1 8 5.30 2.38 .66 .01
Barriers towards Family Planning 1-7 5 15.42 6.27 3.08 .10
Attitudes towards Condom Usage 1-7 6 36.13 5.95 6.02 .22
Dialogue with Sexual partner 1-7 3 17.20 3.75 5.73 .05
Efficacy: Social efficacy 0-100 4 303.83 74.53 75.95 22.44
Efficacy: Peer resistance 0-100 4 281.95 139.48 70.48 1.97
Ease of Game Play 1-7 5 31.26 4.05 6.25 .05

81
Reliability and Validity
The overall reliabilities of the Knowledge scale (Cronbach’s α = .80),
Barriers scale (Cronbach’s α = .67), and Attitudes scale (Cronbach’s α = .68) were
within the acceptable limits. Table 5 also shows the alphas of the scales when one of
the items was deleted from the scale.
82
Table 5. Reliability and Item Total Correlation Indices of the Knowledge and
Attitude Scales

Item
Corrected
Item-
Total
Correlation

α if item
deleted
Knowledge of Family Planning Methods (α = .80)
– Sterilization .470 .792
– Pill .515 .786
– IUD .472 .791
– Injections .508 .786
– Implant .617 .768
– Rhythm method .586 .773
– Withdrawal .537 .781
– Traditional methods .458 .794

Barriers towards Family Planning (α = .67)
– Family planning methods cost too much. .403 .641
– Family planning methods are inconvenient .362 .659
– Family planning methods are hard to get .536 .582
– It is embarrassing to get family planning services .466 .613
– It is better not to talk about pregnancy in my
community
.404 .643

Attitudes towards Condom Usage (α = .68)
Condoms help to plan the family and protects from
AIDS
.472 .627
My friends think it is a good idea to protect against
HIV/AIDS
.411 .653
Using condoms prevents HIV/AIDS and sexually
transmitted diseases
.319 .691
I will try and use condoms if I have sex in the
future
.514 .612
Using birth control methods to prevent pregnancy is
good
.395 .654
I am able to use condoms to prevent HIV/AIDS .443 .638

83
Exploratory factor analysis, using principal component analysis with varimax
rotation, was conducted for the three knowledge and attitude scales described above.
The principal component analysis is reported in Table 6. The factor loadings for the
Knowledge scale and Barriers scale were found to load on the predicted scales, and
had factor loadings over the suggested .40. The factor loading for the Attitudes scale,
however, produced loading on two factors, with the item ‘Using condoms prevents
HIV/AIDS and sexually transmitted diseases’ loading on the second factor. Even
though the alpha for the Attitudes scale would increase if the ‘Using condoms
prevents HIV/AIDS and sexually transmitted diseases’ item were to be dropped,
from a theoretical point of view this item was important, and thus was retained for
the analysis.
84
Table 6. Factor matrix of Knowledge and Attitudes using Principal Components
Analysis
Item Factor Loadings
Knowledge of Family Planning Methods
Sterilization .606
Pill .650
IUD .605
Injections .644
Implant .740
Rhythm method .713
Withdrawal .668
Traditional methods .591

Barriers towards Family Planning
Family planning methods cost too much. .637
Family planning methods are inconvenient .579
Family planning methods are hard to get .760
It is embarrassing to get family planning services .698
It is better not to talk about pregnancy in my community .631

Attitudes towards Condom Usage
Condoms help to plan the family and protects from AIDS .659 .213
My friends think it is a good idea to protect against HIV/AIDS .646 -.402
Using condoms prevents HIV/AIDS and sexually transmitted
diseases
.473 .808
I will try and use condoms if I have sex in the future .704 .177
Using birth control methods to prevent pregnancy is good .625 -.309
I am able to use condoms to prevent HIV/AIDS .676 -.288

The reliabilities for covariates and mediating factors are reported in Table 7
and Table 8. The alphas reported are within the acceptable range, and the loadings
from the principal components analysis all load over .40. The efficacy scales were
very reliable, with all items loading on the expected factors. The Cronbach’s alpha
range from .71 to .93 for the Social Efficacy, and Peer Resistance Efficacy scales
85
respectively. The alphas for the final two measures were on the lower side, Talk (.59)
and Game Play (.69), but were still acceptable.
Table 7. Reliability and Item Total Correlation Indices of the other Scales
Item Corrected
Item-Total
Correlation
α if item
deleted
Sexual partner dialogue (α = .59)
– I can talk to my sexual partner about
preventing HIV/AIDS
.432 .449
– I can talk to my sexual partner about using
condoms
.428 .449
– I can talk about family planning methods with
my sexual partner(s)
.346 .573

Social efficacy (α = .71)
– Make and keep friends of the opposite sex .579 .613
– Carry on conversations with others .568 .620
– Work well in a group .400 .715
– Get a friend to help me when I have social
problems
.497 .664

Self Efficacy: Resist peer pressure (α = .93)
– Resist peer pressure to smoke cigarettes .859 .904
– Resist peer pressure to drink beer, wine, or
liquor
.814 .919
– Resist peer pressure to smoke marijuana .846 .910
– Resist peer pressure to have sexual
intercourse
.848 .907

Ease of Game Play (α = .69)
– I trust the information contained in the game .490 .628
– I find it easy to get access to information by
playing the game.
.433 .657
– I find it easy to use the game according to my
needs.
.446 .647
– I find the game to be useful for health
information.
.538 .616
– I will recommend the the game to my friends. .370 .677
86
Table 8. Factor matrix of other factors using Principal Components Analysis
Item Loadings
Dialogue with Sexual partner
I can talk to my sexual partner about preventing HIV/AIDS .773
I can talk to my sexual partner about using condoms
.773
I can talk about family planning methods with my sexual
partner(s)
.681

Social efficacy
Make and keep friends of the opposite sex .796
Carry on conversations with others .785
Work well in a group .628
Get a friend to help me when I have social problems .728

Self Efficacy: Resist peer pressure
Resist peer pressure to smoke cigarettes .926
Resist peer pressure to drink beer, wine, or liquor .898
Resist peer pressure to smoke marijuana .913
Resist peer pressure to have sexual intercourse .913

Ease of Game Play
I trust the information contained in the game .707
I find it easy to get access to information by playing the game. .652
I find it easy to use the game according to my needs. .661
I find the game to be useful for health information. .751
I will recommend the use of the game to my friends. .600

Sources of Health Information
The objective of this section of the data analysis was to understand the
sources of health information for students. The importance of media sources and
social influences on their health knowledge and attitudes is elaborated in further
sections. Here, however, the relative importance of these sources of health
information are explored to determine the most influential, the most frequently
87
accessed, and the most trusted sources of information. The frequencies were
analyzed to see, for example, whether a media source such as television is more
frequently accessed for health information than a social resource such as health
organizations, and if so, whether it is trusted more as well. Based on the literature
review, it is expected that peers such as classmates, would not be an important
source of health information.
In terms of the most important ways cited for obtaining medical and health
information, (N = 617), health organizations were mentioned most frequently (N =
120), followed by traditional media such as television (N = 113), radio (N = 108),
and newspapers (N = 62). Social resources come next in importance, with friends and
family (N = 56) marginally more important than doctors or health care workers (N =
46). Interestingly, classmates in school (N = 24) are the least important source of
medical and health information. The detailed results can be found in Table 9, and are
depicted in Figure 10.
88
Table 9. Most Important Ways of Obtaining Health Information from Media and
Social Resources
Sources N Percent
Television 113 17
Radio 108 16
Newspapers 62 9
Magazines 45 7
Brochures 43 6
Friends Or Family 56 8
Classmates 24 4
Doctors/Health-care workers 46 7
Health Organization 120 18

Figure 10. Most Important Ways of Obtaining Health Information from Media and
Social Resources

0
20
40
60
80
100
120
140
Health Organizations
Television
Radio
Newspapers
Magazines
Brochures
Friends/Family
Doctors/Healthcare
Classmates
Number accessing Media

89

A series of independent samples t-tests confirmed that there are no
differences between males and females in terms of the importance placed on
different media and the social resources from which to obtain medical and health
information (see Table 10).
Table 10. Most Important Ways of Obtaining Health Information from Media and
Social Resources for differences between Genders
Sources M (Male) M (Female) df t-value p
Television 2.70 2.58 216 1.10 .27
Internet 2.82 2.93 216 -.33 .74
Radio 2.70 2.63 214 .64 .52
Newspapers 2.55 2.61 211 -.49 .62
Magazines 2.20 2.24 213 -.29 .76
Cinema 1.29 1.33 213 -.43 .66
Brochures 2.44 2.65 215 -1.50 .13
Friends Or Family 2.82 2.86 214 -.37 .70
Classmates 2.03 2.10 213 -.62 .53
Doctors/Health-care workers 3.31 3.27 213 .31 .75
Health Organization 3.38 3.42 214 -.31 .75

One might therefore expect the frequency of obtaining health information
from these various sources to vary as a function of the importance attached to the
sources. Table 11 presents the descriptive statistics. The most frequently accessed
source of health information is television (M = 3.05, SD = .99) followed by friends
and family (M = 2.81, SD = 1.13). Traditional media such as radio (M = 2.59, SD =
90
1.13), and newspapers (M = 2.66, SD = 1.10) follow. Less frequently accessed
sources of medical and health information are health organizations (M = 2.50, SD =
1.14) and doctors or health-care workers (M = 2.40, SD = 1.12). Similar to their
importance as a health information resource, classmates at school (M = 2.14, SD =
1.05) are one of the least frequently accessed source of medical and health
information.
Table 11. Frequency of Obtaining Health Information from Media and Social
Resources
Sources Mean SD
Television 3.05 0.99
Internet 2.23 1.13
Radio 2.59 1.13
Newspapers 2.66 1.10
Magazines 1.91 1.00
Cinema 1.30 .72
Brochures 2.30 1.11
Friends Or Family 2.81 1.00
Classmates 2.14 1.05
Doctors/Health-care workers 2.40 1.12
Health Organization 2.50 1.14

A series of independent samples t-tests in Table 12 confirmed that there are
no differences between genders in terms of frequency of obtaining health
information from media and social resources.
91
Table 12. Independent Samples T-test for Frequency of Accessing Health
Information from Media and Social Resources for differences between Genders
Sources M (Male) M (Female) df t-value p
Television 3.18 2.95 216 1.73 .08
Internet 2.27 2.19 214 .47 .63
Radio 2.64 2.54 213 .64 .51
Newspapers 2.72 2.59 210 .84 .39
Magazines 1.83 1.98 210 -1.06 .28
Cinema 1.26 1.33 213 -.72 .46
Brochures 2.24 2.36 207 -.83 .40
Friends Or Family 2.69 2.93 210 -1.80 .07
Classmates 2.10 2.16 211 -.45 .64
Doctors/Health-care workers 2.51 2.29 211 1.45 .14
Health Organization 2.62 2.37 216 1.60 .11

Table 13 displays the mean levels of trust that respondents place in various
sources of medical and health information. Students place their highest level of trust
in socially obtained health information. Health organizations (M = 3.40, SD = 0.99)
and doctors or health-care workers (M = 3.29, SD = 1.04) are the most trustworthy,
followed by friends and family (M = 2.84, SD = .84). Traditional media are less
trustworthy than social resources, as indicated by the results of radio (M = 2.66, SD
= .84), television (M = 2.63, SD = 0.83), and newspapers (M = 2.58, SD = .88).
Classmates, mirroring their importance and frequency of access, are much less
trusted (M = 2.07, SD = .85).

92
Table 13. Trust Health Information from Media and Social Resources
Sources Mean SD
Television 2.63 0.83
Internet 2.72 1.00
Radio 2.66 0.84
Newspapers 2.58 .88
Magazines 2.21 .99
Cinema 1.31 .63
Brochures 2.54 1.01
Friends Or Family 2.84 .84
Classmates 2.07 .85
Doctors/Health-care workers 3.29 1.04
Health Organization 3.40 .99

Independent samples t-tests in Table 14 indicate that both genders are similar
in term of trust in health information from a variety of sources.
93
Table 14. Independent Samples T-test for Trust in Health Information from Media
and Social Resources for differences between Genders
Sources M (Male) M (Female) df t-value p
Television 2.70 2.58 216 1.10 .27
Internet 2.82 2.93 216 -.33 .74
Radio 2.70 2.63 214 .64 .52
Newspapers 2.55 2.61 211 -.49 .62
Magazines 2.20 2.24 213 -.29 .76
Cinema 1.29 1.33 213 -.43 .66
Brochures 2.44 2.65 215 -1.50 .13
Friends Or Family 2.82 2.86 214 -.37 .70
Classmates 2.03 2.10 213 -.62 .53
Doctors/Health-care workers 3.31 3.27 213 .31 .75
Health Organization 3.38 3.42 214 -.31 .75

This analysis suggests that traditional media are more frequently accessed for
medical and health information than social resources such as friends and family, and
doctors. However, in terms of trustworthiness, the social resources get greater
emphasis than media sources. The results are displayed in Figure 11.
94
Figure 11. Frequency of Obtaining, and Trust in, Health Information from Media and
Social Resources

Television
Radio
Newspapers
Internet
Magazines
Cinema
Friends/Family
Classmates
Doctors/Healthcare
Frequency / Trust
Frequency
Trust

Correlations between the three key sources of health information, television,
friends and family and doctors, and the knowledge and attitude variables are
presented in Table 15. The results indicate that frequency of access for television
does not have a significant relationship with the knowledge and attitude variables.
However, the social resources have significant relationships with knowledge; friends
and family (.16, p < .05) and doctors and health-care workers (.20, p < .01). Those
who access doctors and health-care workers more have lower barriers towards family
planning, as can be seen in the negative correlation (-.16, p < .05).
95
Table 15. Correlation Matrix for Frequency of, and Trust in, Sources for Health
Information and Knowledge Attitude, and Efficacy variables
Frequency of Access Level of Trust
Television Friends
and
Family
Doctors/
healthcare
workers
Television Friends
and
Family
Doctors/
healthcare
workers
Knowledge .04 .16* .20** .22** .23* .28**
Barriers .00 -.02 -.16* -.03 -.06 -.18**
Attitudes .00 .00 .08 .15* .10 .09
* p < .05, ** p < .01

The results for trustworthiness show that trust in information from television
has significant correlations with the knowledge (.22, p < .01) and attitude variables
(.15, p < .05). The trustworthiness in social resources is significantly related to the
Knowledge and Barriers variables, but is not significantly correlated with Attitudes.
The unreliability of cinema is noteworthy, which is the least trusted source of
health information (M = 1.31, SD = .63). As a consequence, cinema is also the least
frequented source of medical and health information (M = 1.30, SD = .72).
Preliminary Analysis
The correlations among the variables of knowledge and attitudes, and the
efficacy variables are presented in Table 16 for the pre-intervention data and in Table
17 for the post-intervention data. The significant correlation within the group of
knowledge and attitudinal composite scales that was maintained both pre- and post-
96
intervention was the negative relationship between Knowledge and Barriers (pre = -
.31, p < .01; post = -.35, p < .01). Interestingly, the correlation between Knowledge
and Attitudes (.16, p < .01), in the pre-intervention sample is not significant in the
post-sample.
Table 16. Correlations for the Pre-Intervention Sample
Efficacy
Knowledge Barriers Attitudes Social Peer-
Pressure
Barriers -.31**
Attitudes .16*
Social .20** -.18** .22**
Peer-
Pressure
.25** .37** .34***
Group .23** -.15* .31** .59** .39**
* p < .05, ** p < .01

Table 17. Correlations for the Post-Intervention Sample
Efficacy
Knowledge Barriers Attitudes Social Peer-
Pressure
Barriers -.35**
Attitudes
Social .27** .23**
Peer-
Pressure
-.15* .14* .16*
Group .22** -.17* .24** .60** .30**
* p < .05, ** p < .01
97
The two efficacy scales positively correlate to each other in both pre- and
post-intervention samples. The Barriers variable has a negative correlation with
Social efficacy (-.18, p < .01) in the pre-intervention sample, and with Peer
Resistance efficacy (-.15, p < .05) in the post-intervention sample.
Summary of Hypothesis Testing
Hypothesis 1 stated that there would be a positive impact of playing the game
on key indicators of health knowledge and attitudes, and efficacy measures.
Hypothesis 1 was supported. For the entire sample, all the knowledge and attitudinal
variables were significant in the predicted direction, and the results are presented in
Table 18. With the confidence level at 95%, the paired-sample t-test indicated that
there was a significant difference in the Knowledge scale, t (177) = 2.42, p < .01.
The mean difference for the knowledge score was .03 with a standard mean error
of .01. The 95% confidence interval ranged from .00 to .06. For the Barriers variable,
the paired-sample t-test indicated that there was a significant difference, t (178) = -
3.70, p < .01. The mean difference for the barrier score was -.33 with a standard
mean error of -.09. The 95% confidence interval ranged from -.51 to -.15. The
paired-sample t-test indicated that there was a significant difference in the Attitudes
scale, t (178) = 4.27, p < .01. The mean difference for the Knowledge score was .37
with a standard mean error of .08. The 95% confidence interval ranged from .19
to .54. The sub-hypotheses for the first hypothesis for the efficacy variables were
also all supported.
98
Table 18. Mean Differences and Standard Error between Pre and Post Samples
Knowledge Barriers Attitudes Social Peer-
Resistance
Condoms
Mean
Difference
.03* -.33** .37** 4.80** 7.42** .03*
Standard
Error
.01 .09 .08 1.16 2.39 .01
* p < .05, ** p < .01

With the confidence level at 95%, the paired-sample t-tests indicated that
there was a significant difference between the pre- and post-intervention samples for
the efficacy variables. For the Social efficacy scale, t (178) = 4.11, p < .01. The
mean difference for the Knowledge score was 4.80 with a standard mean error of
1.16. The 95% confidence interval ranged from 2.50 to 7.11. For the Peer Resistance
efficacy measure, the paired-sample t-test indicated that there was a significant
difference, t (178) = 3.10, p < .01. The mean difference for the Barrier score was
7.42 with a standard mean error of 2.39. The 95% confidence interval ranged from
2.70 to 12.14.
Hypothesis 2 stated that the interactive health game would more effective
than the traditional game when measured by key indicators of health knowledge and
attitudes, and efficacy measures. The change in these key variables pre- and post-
intervention were compared between the two game conditions. The paired-sample t-
99
tests for hypothesis two were not significant. The results can be found in Table 19.
We find that there is no significant difference in the two types of interventions.
Table 19. Means Differences and Standard Error between Interactive and Health
game for the Pre-Intervention Sample
∆Knowledge ∆Barriers ∆Attitudes ∆Social ∆Peer-
Resistance
Mean
Difference
.01 .09 -.15 .1.85 1.99
Standard Error .03 .18 .17 2.36 4.85
* p < .05, ** p < .01

Hypothesis 3 stated that respondents with higher efficacy would have greater
health knowledge and attitudes than low efficacy players. This hypothesis was tested
for the two efficacy variables, Social, and Peer-resistance. Independent samples t-
tests were conducted with the efficacy variables as the grouping variables, divided
into two groups by the median value. The results are presented in Table 20 and Table
21. The results for the pre-intervention group suggest that for all efficacy variables,
Knowledge and Attitudes of the high efficacy groups is significantly greater than
those with lower efficacy. However, with the pre-intervention group, the hypothesis
that the perceived Barriers of the high efficacy group will be less than that of the low
efficacy groups, is only significant for the Peer-resistance efficacy variable.
100
Table 20. Means Differences and Standard Error between High and Low Social
Efficacy Groups for the Pre-Intervention Sample
Knowledge Barriers Attitudes
Mean Difference .10** -.25 .46**
Standard Error .03 .16 .15
* p < .05, ** p < .01

Table 21. Means Differences and Standard Error between High and Low Peer
Resistance Efficacy Groups for the Pre-Intervention Sample
Knowledge Barriers Attitudes
Mean Difference .14** -.47** .55**
Standard Error .03 .16 .15
* p < .05, ** p < .01

To verify that the intervention had the same impact on both the high and low
efficacy groups for all the efficacy variables, a series of independent samples tests
were conducted. The results are presented in Table 22 and Table 23. We conclude
that the changes in scores for all knowledge and attitudinal variables were similar for
both high and low efficacy groups, i.e., the intervention had a similar impact on all
respondents.
Table 22. Means Differences and Standard Error between High and Low Social
Efficacy Groups for the Pre-Intervention Sample
∆Knowledge ∆Barriers ∆Attitudes
Mean Difference -.03 -.14 -.22
Standard Error .03 .18 .17
* p < .05, ** p < .01
101

Table 23. Means Differences and Standard Error between High and Low Peer
Resistance Efficacy Groups for the Pre-Intervention Sample
∆Knowledge ∆Barriers ∆Attitudes
Mean Difference -.03 -.11 -.23
Standard Error .03 .18 .17
* p < .05, ** p < .01

Hypothesis 4 stated that respondents with greater ease of game-play will have
greater health knowledge and attitudes than those with lower ease of game-play. For
this hypothesis the Game variable was tested for the post-intervention group, the
results of which are shown in Table 24. Players who found it easier to play the game
had significantly better Knowledge and Attitudes compared to those who did not find
the game as easy to play. The hypothesis was not supported for the Barriers variable.
With the confidence level at 95%, the independent-samples t-test indicated that there
was a significant difference in the Knowledge scale, t (206) = 2.78, p < .01. The
mean difference for the knowledge score was .10 with a standard mean error of .03.
The 95% confidence interval ranged from .03 to .17. For the Attitudes variable, the
independent-samples t-test indicated that there was a significant difference, t
(178.09) = 4.49, p < .01. The mean difference for the attitudes score was .59 with a
standard mean error of .13. The 95% confidence interval ranged from.33 to .85. For
the efficacy variables, the independent-samples t-test indicated that there was a
significant difference in both the Social and Peer-resistance efficacy variables.
102
Table 24. Means Differences and Standard Error between High and Low Ease of
Game Play Groups for the Post-Intervention Sample
Knowledge Barriers Attitudes Social
Efficacy
Peer-
resistance
efficacy
Mean
Difference
.10** -.32 .59** 7.90** 9.22*
Standard
Error
.03 .18 .13 2.27 3.7
* p < .05, ** p < .01. Equal variance not assumed.

Hypothesis 5 stated that respondents with more central positions, as defined
by centrality measures within social network analysis would have greater health
knowledge, attitudes, and efficacy than those on the peripheries. There are three
networks of interest; two unobserved links obtained via respondent self-reports, the
health advice and friendship networks, and the observed co-playing network. The
hypothesis was only tested with degree centrality for the first two observed groups
because they have largely connected networks. The observed co-playing network
consists of multiple small and unconnected networks of players, and hence the
centrality measures are quite meaningless. The results are presented in Table 25 and
Table 26.
103
Table 25. Means Differences and Standard Error between High and Low Advice
Network Degree Groups for the Pre-Intervention Sample
Knowledge Barriers Attitudes Social Peer-Resistance
Mean Difference -.01 .07 .26 8.40** 3.79
Standard Error .04 .18 .15 2.52 4.87
* p < .05, ** p < .01

Table 26. Means Differences and Standard Error between High and Low Friendship
Network Degree Groups for the Pre-Intervention Sample
Knowledge Barriers Attitudes Social Peer-Resistance
Mean Difference .01 -.04 .30 6.35* 9.35
Standard Error .04 .17 .15 2.57 4.75
* p < .05, ** p < .01

For the advice network it was found that the only significant variable was
Social efficacy. Respondents with high centrality had significantly higher Social
efficacy than those with low degree centrality. For degree centrality of the advice
network, the independent-samples t-test indicated that there was a significant
difference, t (192) = 3.33, p < .01. The mean difference was 8.40 with a standard
mean error of 2.52. The 95% confidence interval ranged from 3.43 to 13.38. For the
friendship network it was similarly found that the only significant variable was
Social efficacy in the case of degree centrality. The independent-samples t-test
indicated that there was a significant difference, t (218) = 2.46, p < .01. The mean
104
difference was 6.35 with a standard mean error of 2.57. The 95% confidence interval
ranged from 1.27 to 11.44.
The heuristic of visual representation of the social network suggests that
more central adolescents had higher social efficacy. Figure 12 and Figure 13
illustrate this phenomenon with respective advice and friendship network data from
the Independencia Americana school. The darker circles, representing high social
efficacy, are located in more central positions than the light colored circles,
representing low social efficacy.
Hypothesis 6 stated that players with greater links in the network would
exhibit greater health knowledge and attitudes than players with less links. This
hypothesis was tested for the pre-intervention group (See Table 27 and Table 28). It
was found that, for both advice and friendship networks, those players who had
greater personal network links had significantly better Social Efficacy as compared to
those who had less links in their immediate networks.
Table 27. Means Differences and Standard Error between High and Low Personal
Links of the Advice Network for the Pre-Intervention Sample
Knowledge Barriers Attitudes Social
Efficacy
Peer- resistance
efficacy
Mean
Difference
.00 .02 .26 6.19* -1.10
Standard
Error
.04 .18 .15 2.56 4.89
* p < .05, ** p < .01. Equal variance not assumed.

105
Table 28. Means Differences and Standard Error between High and Low Personal
Links of the Friends Network for the Pre-Intervention Sample
Knowledge Barriers Attitudes Social
Efficacy
Peer- resistance
efficacy
Mean
Difference
-.03 .11 .38* 7.09** 1.36
Standard
Error
.04 .17 .16 2.66 4.97
* p < .05, ** p < .01. Equal variance not assumed.

For the advice network, the paired-sample t-test indicated that there was a
significant difference in the Social Efficacy scale, t (192) = 2.42, p < .05. The mean
difference for the Social Efficacy score was 6.19 with a standard mean error of 2.56.
The 95% confidence interval ranged from 1.14 to 11.24. For the friendship network,
the independent-sample t-test indicated that there was a significant difference in the
Social Efficacy scale, t (218) = 2.65, p < .01. The mean difference for the Social
Efficacy score was 7.09 with a standard mean error of 2.66. The 95% confidence
interval ranged from 1.83 to 12.35. Visual representation of the social networks
confirms that those nodes with a greater number of immediate links tend to have
higher social efficacy (See Figure 12 and Figure 13).
106
Figure 12. Independencia Americana School Advice Network (Social Efficacy)

High
Low

107
Figure 13. Independencia Americana School Friendship Network (Social Efficacy)

High
Low

For the friendship network only, those with greater links had higher attitudes
as well. The independent-sample t-test indicated that there was a significant
difference in the Attitudes measure, t (218) = 2.38, p < .05. The mean difference for
the Attitudes score was .38 with a standard mean error of .16. The 95% confidence
interval ranged from .06 to .71. Analysis of the social network in Figure 14
108
complements this finding. The social network from Nicolás Copernico school
demonstrates that the nodes with greater attitudes, represented by the dark color, can
be found to have greater links. Conversely, those nodes with relatively lower
attitudes, represented by the light color, can be found to be on the margins of the
network.
Figure 14. Nicolas Copernico School Friendship Network (Attitudes)

High
Low

109
Hypothesis 7 stated that the knowledge and attitudes of alters in the network
would be associated with the knowledge and attitudes of the individuals. The
correlations are shown in Table 29.
Table 29. Correlations Between Network Alters and Individuals for the Pre-
Intervention Sample
Efficacy
Knowledge Attitudes Social Peer-Pressure
Advice Network .56** .13 .15 .40**
Friendship Network .45** .10 .15 .61**
* p < .05, ** p < .01

The hypothesis that knowledge of immediate alters in a personal network will
have an impact on self knowledge was significant for the advice network. The
independent-sample t-test indicated that there was a significant difference in the
Knowledge measure, t (150) = 5.68, p < .01. The mean difference for the Knowledge
score was .23 with a standard mean error of .04. The 95% confidence interval ranged
from .15 to .31. (See Table 30). The visual heuristic of the social network sociogram
in Figure 15 suggests that there is a grouping according to knowledge level. Data
from Daniel Alomias Robles school confirms that the knowledge level of one’s
immediate alters correlated with the individuals’ own knowledge level.
110
Table 30. Means Differences and Standard Error between High and Low Knowledge
and Attitudes of Alters for the Pre-Intervention Sample
Mean Difference Standard Error
Advice Network
Knowledge .23** .04
Barriers .08 .20
Attitudes .21 .17
Friends Network
Knowledge .14 .09
Barriers .59 .37
Attitudes .34 .41

The hypothesis that attitudes of immediate alters in a personal network will
have an impact on self attitudes was not supported for either advice or friendship
networks.
111
Figure 15. Daniel Alomias Robles School Advice Network (Knowledge)

High
Low

Summary of Multiple Regression Analysis
Multiple regression analyses were conducted to evaluate how well the
independent variables predicted attitudes towards condom usage, the Attitude
variable. The pre-intervention multiple regression indicated that the linear
combination of the independent variables were significantly related to the attitude
variable, F (7, 97) = 6.36, p < .001. The sample multiple correlation coefficient
was .56, indicating that approximately 31% of the variance observed in the Attitude
112
variable could be accounted for by the linear combination of the independent
measures.
Table 31 displays the standardized coefficient Betas, and the zero-order and
partial correlations for the predictor variables. The Knowledge, Talk, and the efficacy
variable, Peer-resistance, have positive and significant betas. Their partial
correlations are .28, .22, and .41 respectively. Based on these, it is tempting to
suggest that the squared sum of these predictors account for 29% of the 31%
explained variance. It is, however, difficult to reach this conclusion due to the inter-
relatedness of these measures.
Table 31. Standardized Coefficients, Bivariate and Partial Correlations of Predictors
with Attitudes Factor for the Pre-Intervention Sample
Predictors Beta Zero-Order Correlation Partial Correlation
Gender -.08 -.10 -.09
Age -.02 -.06 -.02
School 2 .05* -.03 .19
Knowledge .30** .33** .28**
Barriers .05 -.05 .05
Talk .20* .24* .22*
Peer-resistance .40** .40** .41**
* p < .05, ** p < .01

The multiple regression for the post-intervention indicated that the linear
combination of the independent variables were significantly related to the attitude
variable, F (12, 147) = 7.15, p < .001. The sample multiple correlation coefficient
113
was .60, suggesting that approximately 37% of the variance observed in the Attitude
variable could be accounted for by the linear combination of the independent
measures.
Table 32 displays the standardized coefficient Betas, and the zero-order and
partial correlations for the predictor variables. For the post-intervention, the Attitudes
variable for the pre-intervention, the Peer-resistance efficacy variable for the post-
intervention, and the intervention itself, in the form of the Play variable, have
positive and significant betas. Their partial correlations are .41, .17 and .21
respectively. An R
2
change analysis suggests that the additional variables did not
predict significantly over and above these variable, R
2
change = .04, F(5, 154) = 1.42,
p = .22. These predictors account for 32% of the explained variance of 37%. Thus,
knowledge, discussion, and barriers impact on the variance of the attitudes in the
post-intervention is already captured by the attitudes measure pre-intervention, and
that the additional change is due to playing the health game, and the peer-resistance
efficacy.
114
Table 32. Standardized Coefficients, Bivariate and Partial Correlations of Predictors
with Attitudes Factor for the Post-Intervention Sample
Predictors Beta Zero-Order Correlation Partial Correlation
Gender -.17** -.13 -.21
Age -.01 -.03 -.01
School 2 -.05 -.10 -.05
School 3 .01 .08 .01
School 4 .14 .08 .14
Knowledge -.10 .20 -.09
Barriers -.08 -.14 -.09
Talk .02 .29** .02
Social .06 .28** .06
Peer-resistance .15* .28** .17*
Attitudes_Pre .40** .44** .41**
Game .20** .32** .21*
* p < .05, ** p < .01

115
Chapter 5
Discussion
The purpose of this dissertation was to understand the underlying
relationships and influences in interactive health interventions. Diffusion of
innovations research and social cognitive theory provided the framework to establish
relationships. This study, however, was the first to utilize social network analysis as
a methodology to include social influences in a mixed-influences model. Since
vulnerable teenagers in developing countries face high-risk sexual and reproductive
health situations, the use of innovative and exciting interactive games was studied in
addition to traditional board games.
This chapter presents a summary of the findings, including the theoretical and
methodological significance of the study, and outlines the implications arising from
them. In conclusion, criticisms, limitations, and suggestions for future research are
discussed.
Theoretical and Methodological Significance
The uniqueness of this dissertation is that it contributes to a better
understanding of processes governing the adoption of beneficial health attitudes. It
does so by testing hypotheses derived from a model of influence incorporating
intrapersonal, interpersonal, and media influences. We find that health attitudes of
students in the study were influenced not only by individuals’ own knowledge and
116
self-efficacy determinants, but also by their interpersonal discussions. The results
indicate that the socio-structural influences in health interventions play an important
part. This is especially true in the target audience for this study, i.e., adolescents;
more so because of the sensitive nature of the content, viz., sexual and reproductive
health.
The results of the data analysis suggest that the proposed general model (See
Figure 4) may contribute to better explanation of variance than the general model
described in the theory chapter (See Figure 2). Critical extensions are discussed here.
Previous findings (Bandura, 2003; Rosenthal, Moore, & Flynn, 1991) found
that perceived self-efficacy is related significantly and positively to safe sex practices.
This research extends this notion to an individuals’ social network, especially that of
the peer group. The notion that the pre-existing knowledge and attitudes of friends
affects an individuals’ own is not new. However, this research provides evidence that
the type, number of links, and position of the individual within the relevant network
have the potential to affect their personal efficacy, knowledge and attitudes.
The barriers that individuals face in deciding to engage in safe sexual
practices did get reduced by playing the health games. However, unlike attitudes, the
barriers measured did not decrease more for those who found the game easier to play
than those who find the game difficult to play. It is possible that the act of co-playing
was relatively more responsible for the lowering of individuals’ barriers. The role of
117
the social network is further supported by the fact that those who have greater peer
resistance self-efficacy had lower barriers.
It would seem that the definition of the dependent variables within the theory
is critical when understanding psychosocial influences. The attitudes variable was
more affected by the psychological profile of respondents, while the barriers variable
is influenced by social factors. Both measure attitudes, yet face different influences.
It is tempting to conclude that positive attitudes are less susceptible to external-
influence than negative attitudes, but this is an argument that requires greater
attention.
The concept of self-efficacy is approached from a unique social perspective,
rather than adopting the traditional methods of measuring collective efficacy. Other
researchers have focused on self-efficacy as a mediating influence on the relationship
between knowledge and attitudes (usually studied as a behavioral variable). The key
extensions to that model include two dimensions for self-efficacy which capture
social aspects.
Here, self-efficacy is divided into the two theoretically different domains.
The first notion is that of social efficacy, or the ability to function effectively socially.
The second domain is that of being able to resist peer pressure exerted by one’s
immediate social group. It has been long theorized that health information too is
118
derived from dual sources, the media and the social network. The dual measures of
self-efficacy can be used to understand the influences of the media and social group.
The methodological contribution that this study makes is that the influence of
the social network is gauged by accurately determining the knowledge and attitudinal
levels of alters. It is suggested that specific relationships will determine the extent to
which variance in the observed dependent variables is explained. In this instance, we
find that the health advice network had greater explanatory power for knowledge
than the friendship network. This is not a surprising result, given the low importance
that classmates are given in terms of being a source of health information.
Summary of Major Findings and Implications
The first research question asked the question whether health games led to
improvement in health knowledge and attitudes. The findings of this dissertation
suggest that both traditional games and recent interactive computer-based games lead
to significant improvements in the health knowledge and attitudes of players.
Equally importantly, these games, through their co-playing nature, also positively
influence social and personal aspects of self-efficacy. Playing of these games led to
increases in all relevant knowledge, attitudinal, and efficacy variables, except for
barriers to engaging in productive behaviors.
This dissertation examines the effectiveness of multimedia games, both
interactive computer-based and traditional board-based, in improving health
119
knowledge, efficacy, and attitudes. The second research question investigated
whether interactive multimedia games were more effective than traditional games in
meeting sexual and reproductive health objectives for teenage audiences. The
findings suggest that both these strategies are equally effective at meeting health
goals. One might expect that disadvantaged youth would find it difficult to use a
computer-based intervention effectively due to their lack of familiarity with
technology. However, even in the slums of Lima, we find that this is not the case.
Although not actively incorporated in the research design, it is conjectured that co-
playing allows those with lesser technological familiarity to reap the benefits from
their more tech-savvy compatriots. We do find that those players who are more at
ease with the games are also significantly more likely to exhibit greater health
knowledge and attitudes after the intervention. Clearly there is a case to be made for
increased access to computers, and technological training to supplement individual
skills.
The third research question aimed to explore the relationship between
efficacy variables and key health indicators such as knowledge and attitudes. We
find that health knowledge and attitudes are higher for both those with high social
efficacy, as well as those who have high personal efficacy in resisting social
pressures. In addition, those with high personal peer-resistance efficacy have lower
negative attitudes as well. This suggests that the ability to counter social pressures
120
manifests itself in dampening barriers to engaging in productive behavior. The
ability to engage socially has no influence on one’s barrier level.
The fourth research question wished to determine the role of interpersonal
networks in affecting key health indicators such as knowledge and attitudes. Those
who are more central in the network, as determined by degree centrality, are more
likely to have better social efficacy. The related measure, number of links, is also
associated with greater social efficacy. This suggests that, unsurprisingly, that being
central or popular would make one have greater confidence in one’s ability to
connect socially. Clearly it is difficult to make a causal link. It may be that having
greater social efficacy leads individuals to make more connections, and be placed
more strategically at the center of their social networks. It is difficult to account for
why the knowledge and attitudinal measures are not significant for central figures.
However, given the influence of social efficacy on knowledge and attitudes, it is
suggested that there may be a mediating role of efficacy on knowledge and
attitudinal variables. Thus, the influence of the social network is captured via this
process.
The final research question investigated whether the type of communication
link would influence health knowledge and attitudes. The health information analysis
suggested that despite social resources being more important than media sources, it is
the latter that are accessed more than social resources. Particularly low as a source of
121
health information for these teenage respondents are classmates at school. This
would explain why the knowledge of the immediate friends’ network is not an
influence on one’s own knowledge. However, the knowledge of people one goes to
for health advice is a significant contributor to one’s own knowledge. It seems that
the type of social network under examination will determine whether there is an
influence, and the strength of the influence.
Practically, this study guides practitioners on not just the what and why, but
the how to implement ICT interventions in health. This information can be a
valuable resource in design and implementation of games for health educators,
educational institutions, and researchers. In an age of increasing penetration of
technology into hitherto unconnected parts of the globe, health organizations now
have the ability to use exciting and innovative new ICTs to attract youth audiences.
We find that the use of characters and modeled situations in multimedia/audiovisual
media in conjunction with interpersonal discussions can be an effective tool in
disseminating health knowledge and attitudes.
However, despite the importance of interpersonal communication, there is
hesitancy in seeking these out. As mentioned previously, cultural and social barriers
exist within any social system. Specifically, youth do not access peer resources as
much as they do traditional media and health organizations in seeking medical
information. There is a need to resolve this missing link in addition to enhancing the
122
health system connection. While games can be one strategy for youth-directed
interventions, institutions should encourage youth to seek out discussions with
friends, family, and health resources. Multimedia games form one aspect of a multi-
pronged strategy; clearly social support needs to be bolstered.
Multimedia games enable social health resources to be more approachable;
and technological innovation delivers credibility and interest of the youth. Despite
lack of familiarity with ICTs, the lack of technological skills is not an impediment to
effectiveness. While the impoverished circumstances of youth do not allow them to
own ICTs within their households, the burgeoning of Internet telecenters and gaming
centers allows students shared usage. Increasingly, even schools in rural areas and
urban slums are implementing technologically-based education. The private sector is
involved in creating computer centers within schools. It is advances such as these
that offer health organizations the opportunity to administer innovative interventions
at the school-level.
Limitations
The primary limitation of the study was that the field-based testing situation,
as opposed to a controlled laboratory environment, led to compromises in research
design. First, the requirements of filling testing quotas was dependent on the
cooperation of the schools and the interest of the students. The difficulties in
minimizing loss of subjects in the post-group, and in one schools’ case, to administer
123
the post-group at all, led to missing subjects. The incompleteness of the dataset led to
compromises in the testing methodologies due to low sample sizes. One direct
consequence of this was the inability to distinguish differences in specific features of
the interactive computer-based game versus the traditional board game.
The second limitation of the study design concerns the measures utilized.
Due to cultural constraints concerning the subject matter, and the age of the
respondent pool, it was not possible to get behavioral measures. This constrained the
ability to make comparisons with other health studies, most of which use behavior as
their dependent variable. Further, the small effect sizes help explain only a portion of
variance; other omitted variables, such as behavior, could possibly have increased
the explanatory power. Given the specific nature of the intervention, it would have
been best to establish and test domain specific efficacy scales prior to data-collection.
However the constraint of limited time in the field did not permit this to happen.
Finally, this dissertation makes certain assumptions about the social network
influence that are do not replicate real life phenomena. The influence of alters is
averaged out across all the immediate links in an individuals’ network. The reason to
make this assumption is due to complexity of statistical testing. To have bi-
directional influence from multiple alters would make the research design too
complex, though it would be a more accurate representation of reality. This is
certainly a direction for future research.
124
Suggestions for Future Research
There are complexities in adopting a multi-level, and multi-theoretical,
approach to answering questions about the effectiveness of ICTs in meeting
development goals. Many questions thus remain unexplored, and unanswered. The
following replications and extensions to this dissertation are proposed.
There is a need to construct reliable domain specific scales for specific health
topics. Many of the self-efficacy questions did not pertain specifically to sexual and
reproductive health topics for adolescents. For example, the peer-resistance self-
efficacy scale asked questions about alcohol usage and smoking, which are not
particularly related to the topic of this dissertation. As a consequence, while these
variables were correlated with the dependent variable, they did not remain significant
in the multivariate analysis. Ideally, a structural equation modeling approach should
have been employed to determine the combined effects suggested in the model.
However, due to the high chi-squares and less than adequate model fit, this approach
was discarded. This is certainly a requirement for future research.
The second direction proposed for future research is to explore the impact of
co-playing in laboratory controlled experimental conditions. The current study, being
conducted in the field, had to bow to the expediencies of research partners,
institutions that provided respondents, and time constraints. It would be extremely
useful to explore the bi-directionality of influence that occurs in social networks, and
125
to design research that could measure this. The current study has initiated the process.
Hopefully, this interesting phenomenon is understood at greater depth in the future.
126
REFERENCES
Abrahamson, E., and Rosenkopf, L. (1997). Social network effects on the extent of
innovation diffusion: A computer simulation. Organization Science, 8, 289-
309.
Abrami, P., Cholmsky, P. and Gordon, R. (2001). Statistical Analysis for the Social
Sciences. Boston: Allyn & Bacon.
Agha, S. (2003). The impact of a mass media campaign on personal risk perception,
perceived self-efficacy and on other behavioral predictors. AIDS Care;15,
749-762.
Amir, M., Roziner, I., Knoll, A. Neufeld, M.Y. (1999). Self-efficacy and social
support as mediators in the relation between disease severity and quality of
life in patients with epilepsy. Epilepsia, 40 (2), 216–224.
Anderson, C.A., and Bushman, B. J. (2001). Effects of violent video games on
aggressive behavior, aggressive cognition, aggressive affect, physiological
arousal, and prosocial behavior: A meta-analytic review of the scientific
literature. Psychological science 12 (5), 353-359.
Anderson, C.A. and Dill, K.E. (2000) Video games and aggressive thoughts, feelings,
and behavior in the laboratory and in life. Journal of Personality and Social
Psychology, 78 (4), 772-790.
Baffigo, V.; Albinagorta, J.; Nauca, L.; Rojas, P; Alegre, R.; Hubbard, B.; and
Sarisky, J. (2001) Community Environmental Health Assessment in Peru’s
Desert Hills and Rainforest. American Journal of Public Health, 91,10.
Ball-Rokeach, S.J. (1998). A theory of media power and media use: Different stories,
questions, and ways of thinking. Mass Communication and Society, 1, 5-40.
Bandura, A. (1977). Social Learning Theory. Englewood Cliffs, NJ: Prentice-Hall.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive
theory. Prentice-Hall: New Jersey.
Bandura, A. (1990). Multidimensional Scales of Perceived Self-Efficacy. Stanford,
CA: Stanford University.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman.
Bandura, A. (1999). A social cognitive theory of personality. In L. Pervin and O.
John (Eds.), Handbook of personality (2nd ed). New York: Guilford
Publications.
Bandura, A. (2001). Social cognitive theory: An agentic perspective. Annual Review
of Psychology, 52, 1-26.
127
Bandura, A. (2002). Social cognitive theory of mass communication. In Bryant, J.
and Zillmann, D. (Eds.), Media effects. Advances in theory and research.
(2nd ed.), 121-153.
Bandura, A. (2004). Health promotion by social cognitive means. Health Education
& Behavior, 31; 143-154.
Bandura, A. (2006). Adolescent development from an agentic perspective. In Pajares,
F. and Urdan, T. (Eds.) Self-efficacy beliefs of adolescents. Information Age
Publishing: Greenwich, Connecticut.
Bandura, A. (2006). Guide for constructing self-efficacy scales. In Pajares, F. and
Urdan, T. (Eds.) Self-efficacy beliefs of adolescents. Information Age
Publishing: Greenwich, Connecticut.
Bandura, A., Barbaranelli, C., Caprara, G. V., and Pastorelli, C. (1996). Multifaceted
impact of self-efficacy beliefs on academic functioning. Child Development,
67, 1206-1222.
Bandura, A., Pastorelli, C., Barbaranelli, C., and Caprara, G. V. (1999). Self-efficacy
pathways to childhood depression. Journal of Personality and Social
Psychology, 76, 258-269.
Baym, N. K. (2001). Interpersonal life online. In S. Livingston and L. Lievrouw
(Eds.), The Handbook of New Media. London: Sage Ltd.
Bell, D. (1999). The Coming of Post-Industrial Society (Special Anniversary
Edition). New York: Basic Books.
Bergsjø, P. (2002). Condoms in the age of AIDS. Acta Obstetricia et Gynecologica
Scandinavica, 81(1), 1 – 4.
Bradley, R. H., and Corwyn, R.F. (2001). Home environment and behavioral
development during early adolescence: The mediating and moderating roles
of self-efficacy beliefs. Merrill-Palmer Quarterly, 47 (2), 165–187.
Bridges.org (2001). Spanning the digital divide: Understanding and tackling the
issues. Downloaded August 16, 2003 from www.bridges.org/spanning.
Brodie, M., Foehr, U., Rideout, V., Baer, N., Miller, C., Flournoy, R., and Altman, D.
(2001). Communicating health information through the entertainment media:
A study of the television drama ER lends support to the notion that
Americans pick up information while being entertained. Health Affairs, 20,
192-199.
Brown, M.M. (2001) ‘Can ICTs Address the Needs of the Poor?’ A Commentary
from UNDP. Downloaded August 16, 2003 from http://www.undp.org/.
Brown, S. J., Lieberman, D. A., Gemeny, B. A., Fan, Y. C., Wilson, D. M., and Pasta,
D. J. (1997). Educational video game for juvenile diabetes: Results of a
controlled trial. Medical Informatics, 22, 77-89.
128
Bull, S. S., McFarlane, M., and King, D. (2001). Barriers to STD/HIV prevention on
the Internet. Health Education Research, 16(6), 661-670.
Buller, D. B., Woodall, W. G., Hall, J. R., Borland, R., Ax, B., Brown, M., and Hines,
J. M. (2000). A web-based smoking cessation and prevention program for
children aged 12 to 15. In R. E. Rice and C. K. Atkin (Eds.), Public
Communication Campaigns (3rd ed.) (pp. 357-372). Newbury Park, CA:
Sage Publications.
Burt, R. S. (1987). Social contagion and innovation: Cohesion versus structural
equivalence. American Journal of Sociology, 92, 1287-1335.
Castells, M. (2000). The information age: economy society and culture, vol.1: The
rise of the network society. Oxford: Blackwell.
Castells, M. (2001a). Information technology and global capitalism. In W. G. Hutton,
Anthony (Ed.), Living with Global Capitalism. London: Vintage.
Castells, M. (2001b). The Internet galaxy. Oxford and New York: Oxford University
Press.
Cecchini, S., and Scott, C. (2003). Can Information and Communications
Technology Applications Contribute to Poverty Reduction? Lessons from
Rural India. Information Technology for Development, 10, pp. 73–84.
Central Intelligence Agency (2005). World Factbook. Office of Public Affairs, CIA:
Washington, DC.
Chib, A. (2006). Wireless technologies in disaster-preparedness and response:
Learnings from Asia post-Tsunami. Presentation from International
Communication Association conference. Dresden, Germany.
Chib, A. (2005). Modeling Efficacy Processes: Distinguishing Individual and
Collective Attitudes towards HIV/AIDS Prevention in Namibia. Presentation
from National Communication Association conference. Boston,
Massachusetts.
Chirinos, J.L., Salazar V.C., and Brindis, C.D. (2000). A profile of sexually active
male adolescent high school students in Lima, Peru. Cad Saúde Pública 2000;
16:733-46.
Choi, N; Fuqua, D. R.; and Griffin, B. W. (2001). Exploratory analysis of the
structure of scores from the multidimensional scales of perceived self-
efficacy. Educational and Psychological Measurement, 61, 3, 475-489.
Chowdhury, N. (2000) ‘Information and Communications Technologies and IFPRI’s
Mandate: A Conceptual Framework.’ Downloaded Sept. 18, 2000 from
http://www.ifpri.org.
129
Chowdhury, N. (2000a) ‘Poverty Alleviation and Information/Communications
Technologies.’ Dec. 2000. Towards a Motif for the United Nations ICT Task
Force. Downloaded from http://www.eb2000.org/
Cody, M. J., Fernandes, S. and Wilkin, H. (2004). Entertainment-education programs
of the BBC and the BBC World Service Trust. In Singhal, A., Cody, M.J.,
Rogers, E.M., and Sabido, M. (Eds.), Entertainment-Education and Social
Change: History, Research, and Practice (pp. 243-260). Mahwah, NJ:
Lawrence Erlbaum Associates.
Collins, R.L., Elliott, M.N., Berry, S.H., Kanouse, D.E., and Hunter, S.B (2003).
Entertainment television as a healthy sex educator: The impact of condom-
efficacy information in an episode of Friends. Pediatrics, 112 (5), 1115-1120.
Comisión de la Verdad y Reconciliación Oficina de Comunicaciones e Impacto
Publico Press release 71.
Dock, A., and Helwid, J. (Eds.). (1999). Interactive radio instruction: Impact,
sustainability, and future directions (Education and Technology Technical
Notes Series; Volume 4, Number 1). Washington, DC: World Bank HDDEG.
The World Bank HD Network.
Donner, J. (2005). Research Approaches to Mobile Use in the Developing World: A
Review of the Literature. ICMCAM conference proceedings. (pp. 1-20).
Donner, J.(2004) The use of mobile phones by microentrepreneurs in Kigali,
Rwanda: Changes to social and business networks. ARNIC workshop
proceedings. (pp. 1-24).
Duncombe R. and R. Heeks (1999) ‘Information, ICTs and Small Enterprise:
Findings from Botswana’, IDPM Manchester Working Paper No. 7.
Downloaded August 16, 2003 from http://idpm.man.ac.uk/.
Durkin, K. and Barber, B. (2002). Not so doomed: Computer game play and positive
adolescent development. Applied Developmental Psychology, 23, 373-392.
Elder, J. P. (2001). Behavior change and public health in the developing world.
Thousand Oaks, CA: Sage.
Flanagin, A., and Metzger, M. (2001). Internet use in the contemporary media
environment. Human Communication Research, 27(1), 153-181
Freeman, L. (1979). Centrality in social networks: Conceptual clarification. Social
Networks, 1, 215-239.
Galperin, H. and Bar, F. (2006) The Microtelco opportunity: evidence from Latin
America. International Development Research Centre. Downloaded August
16, 2003 from http://www-rcf.usc.edu/.
Gee, J. P. (2003). What video games have to teach us about learning and literacy.
Palgrave Macmillan: New York.
130
Gerster, R. and Zimmerman, S. (2003). Information and Communication
Technologies (ICTs) for Poverty Reduction? Berne, Switzerland: SDC.
Granovetter, M. (1978). Threshold models of collective behavior. American Journal
of Sociology, 83, 1420-1443.
Hardee, K., Agarwal, K., Luke, N., Wilson, E., Pendzich, M., Farrell, M., and Cross.
H. (1999). Reproductive health policies and programs in eight countries:
Progress since Cairo. International Family Planning Perspectives, 25, 2-9.
Hawkins, R. J. (2002) “Ten Lessons for ICT and Education in the Developing
World.” In CID (Center for International Development) (2002) The Global
Information Technology Report 2001-2002: Readiness for the Networked
World. Oxford: OUP.
Haythornewaite, C. and Wellman, B. (1998). Work friendship, and media use for
information exchange in a networked organization. Journal of the American
Society for Information Science, 49 (2), 1101-1114.
Haythornewaite, C. (2001). Introduction to the Internet in everyday life. American
Behavioral Scientist, 45, 363-382.
Howard, P. E. N., Raine, L., and Jones, S. (2001). Days and nights on the Internet:
The impact of a diffusing technology. American Behavioral Scientist, 45 (3),
382-404.
Hutton, W, and Giddens, A. (2000). On the edge: Living in global capitalism.
Vintage: UK.
International Telecommunications Union (2003), World Telecommunication
Development Report. Access Indicators for the Information Society.
Jemmott J. B., Jemmott, L.S., Fong, G.T., and McCaffree, K. (1999). Reducing HIV
risk-associated sexual behavior among African American adolescents:
Testing the generality of intervention effects. American Journal of
Community Psychology, 27 (2), 161-187.
Jung, J.-Y., Qiu, L. J., and Kim, Y.-C. (2001). Internet Connectedness and
Inequality: Beyond the "Divide". Communication Research, 28(4), 507-535.
Kane, T. T., Gueye, M., Speizer, I., Pacque-Margolis, S., and Baron, D. (1998). The
impact of a family planning multimedia campaign in Bamako, Mali. Studies
in Family Planning, 3, 309-323.
Kasen, S., Vaughan, R.D., and Walter, H.J. (1992). Self-efficacy for AIDS
preventive behaviors among tenth grade students. Health Education
Quarterly;19(2),187-202.
Katz, E. (1957) The two-step flow of communication: An up-to-date report on an
hypothesis. Public Opinion Quarterly, 21, 61-78.
131
Katz, E., and Lazarsfeld, P. F. (1955). Personal influence: The part played by people
in the flow of mass communication. New York: Free Press.
Katz, J. E., and Rice, R. E. (2002). Social consequences of the Internet use: Access,
involvement, and interaction. Cambridge, Massachusetts: The MIT Press.
Keeble, L. (2003). Why create? A critical review of a community informatics project.
Journal of Computer Mediated Communication, 8 (3).
Kenny, C. (2001). Information and communication technologies and poverty.
TechKnowLogia. (pp. 7-11).
Kenny, C. (2002) Information and Communication Technologies for Direct Poverty
Alleviation: Costs and Benefits. Development Policy Review 20 (2), 141–157.
Kerlinger, F. N., and Lee, H.B. (2000). Foundations of Behavioral Research.
Orlando: Harcourt.
Klapper, J.T. (1960). The effects of mass communication. New York: Free Press.
Knoke, D., and Kuklinski, J.H. (1986). Network Analysis. Beverly Hills, CA: Sage
Publications.
Koku, E., Nazer, N., and Wellman, B. (2001). Netting scholars: online and offline.
American Behavioral Scientist, 44(10), 1752-1774.
Kusunoki, L., Gunaira, J., Navarro, C., Velásquez, C. (2005). Report on monitoring
the declaration of commitment on HIV/AIDS. United Nations General
Assembly Special Sessions on HIV/AIDS Monitoring of the Declaration of
Commitment on HIV/AIDS. Peru.
Lenhart, A., Horrigan, J., Rainie, L., Allen, K., Boyce, A., Madden, M., and O’Grady,
E. (2003). The ever-shifting internet population: A new look at internet
access and the digital divide. Downloaded August 16, 2003 from
http://www.pewinternet.org/.
Li, F., Mcauley, E., Fisher, K.J., Harmer, P., Chaumeton N., and Wilson, N.L. (2002).
Self-efficacy as a mediator between fear of falling and functional ability in
the elderly. Journal of Aging and Health, 14; 452-466.
Lieberman, D. A. (2001). Using interactive media in communication campaigns for
children and adolescents. In Rice, R.E. and Atkin, C.K. (Eds.). Public
communication campaigns. Thousand Oaks, Ca.: Sage.
Lievrouw and Livingstone (2002). The Handbook of New Media: Social Shaping
and Consequences of ICTs. London: Sage.
Linderoth, J., Lindström, B., and Alexandersson, M. (2004). Learning with computer
games. In Goldstein, J., Buckingham, D. and, Brougere, G. (Eds.) Toys,
games and media. Erlbaum: New Jersey.
132
Loli, A.; Aramburú, C. and Paxman, J. M. (1987). Sexuality in Peru. International
Family Planning Perspectives, Special Number:17-27.
Mansell, R. (1999). New media competition and access: The scarcity-abundance
dialectic. New Media and Society, 2 (2), 155-182.
Mansell, R. and U. When (1998) Knowledge Societies: Information Technology for
Sustainable Development. Prepared for the United Nations Commission on
Science and Technology for Development. Oxford University Press, Oxford
McGuire, W. J. (2001). Theoretical foundations of campaigns. In R. Rice and C.
Atkin, (Eds.), Public Communication Campaigns (pp. 22-48). London: Sage.
McKee, N., Manoncourt, E., Yoon, C., and Carnegie, R. (2000). Involving people,
evolving behavior. Penang: United Nations Children's Fund.
Miller, J. W.; Coombs, W. T.; and Fuqua, D. R. (1999). An examination of
psychometric properties of Bandura’s multidimensional scales of perceived
self-efficacy. Measurement and Evaluation in Counseling and Development,
31, 4, 186-197.
Monge, P.R. and Contractor, N.S. (2003). Theories of communication networks.
New York: Oxford University Press.
Mossberger, K., Tolbert, C. J., and Stansbury, M. (2003). Virtual inequality: Beyond
the digital divide. Washington D.C.: Georgetown University Press.
Mosteller F. Innovation and evaluation. Science. 1981;211:881-886.
Mundorf, N. and Laird, K. R. (2002). Social and psychological effects of information
technologies and other interactive media. In Bryant, J and Zillmann. D. (Eds.),
Media Effects: Advances in Theory and Research, (pp. 583-602). Hillsdale,
NJ: Erlbaum.
National Telecommunications and Information Administration (2002). A nation
online: How Americans are expanding their use of the Internet. Downloaded
August 19, 2003 from http://www.ntia.doc.gov/ntiahome/.
Nie, N.H., and Erbring, L. (2000). Internet and society: A preliminary report.
Stanford Institute for the Quantitative Study of Society, Palo Alto, CA.
Downloaded August 16, 2003 from http://www.stanford.edu/group/siqss/.
Norris, P. (2001). Digital divide: Civic engagement, information poverty and the
internet worldwide. Cambridge ; New York: Cambridge University Press.
Novak, T. P., and Hoffman, D.L. (1998) Bridging the Racial Divide on the Internet.
Science, 280, (5362), 390-391.
OECD, Digital Opportunities for Poverty Reduction: Addressing the international
digital divide)
133
O'Leary, A. (1992). Self-efficacy and health: Behavioral and stress-physiological
mediation. Cognitive Therapy and Research, 16 (2), 229-245.
Organization for Economic Co-operation and Development. (2001). Understanding
the digital divide. Downloaded August 16, 2004 from http://www.oecd.org/.
Ott, J., Greening, L., Palardy, N., Holderby, A., and DeBell, W.K. (2000). Self-
efficacy as a mediator variable for adolescents' adherence to treatment for
insulin-dependent diabetes mellitus. Children's Health Care, 29 (1), 47-63.
Paisley, W. J. (2000). Public Communication Campaigns: The American Experience.
In R. E. Rice and C. K. Atkin, Public Communication Campaigns (3rd ed.)
(pp. 3-21). Newbury Park, CA, Sage Publications.
Pajares, F. Hartley, J., and Valiante, G. (2001). Response format in writing self-
efficacy assessment: Greater discrimination increases prediction.
Measurement and Evaluation in Counseling and Development, 33, 214-221.
Papa, M. J., Singhal, A., Law, S., Pant, S., Sood, S., Roger, E. M., and Shefner-
Rogers, C. L. (2000). Entertainment-education and social change: An
analysis of parasocial interaction, social learning, collective efficacy, and
paradoxical communication. Journal of Communication, 50, 31-55.
Piotrow, P., and de Fossard, E (2004). Entertainment-Education as a Public Health
Intervention. In Singhal, A., Cody, M.J., Rogers, E.M., and Sabido, M. (Eds.),
Entertainment-Education and Social Change: History, Research, and Practice
(pp. 39-60). Mahwah, NJ: Lawrence Erlbaum Associates.
Piotrow, P.T., and Kincaid, D. L. (2001). Strategic communication for international
health programs. In Rice, R.E. and Atkin, C.K. (Eds.) Public communication
campaigns. Thousand Oaks, Ca.: Sage.
Prahalad, C.K. (2005). The Fortune at the Bottom of the Pyramid. NJ: Wharton.
Prahalad, C.K. and Hammond, A. (2002). Serving the world’s poor, profitably.
Harvard Business Review, 80 (9), 48-57
Prochaska, J.O., DiClemente, C.C., and Norcross, J.C. (1992). In search of how
people change applications to addictive behaviors. American Psychologist,
47 (9), 1102-1114.
Proenza, F. J. (2002). E-ForAll: A poverty reduction strategy for the information age.
Rantanen, T. (2001). The old and the new: Communication technology and
globalization in Russia. New Media and Society, 3, 85-105.
Rice R. and Atkin, C. (2001). Public Communication Campaigns. London: Sage.
Rimal, R. (2003). Intergenerational transmission of health: The role of intrapersonal,
interpersonal, and communicative factors. Health Education and Behavior, 30
(10), 10-28.
134
Rizzo, S., and McLaughlin, M. (2006, May 9, 2006). Addressing PTSD,
PsychoTherapy, & Stroke rehabilitation with Games & Game Technologies.
Paper presented at the Games for Health, University of Southern California.
Robertson, T.S. (1971). Innovative behavior and communication. New York: Holt,
Rinehart & Winston.
Rogers, E. (2004). Delivering Health Messages Through the Internet to Hard-To-
Reach U.S. Audiences in the Southwest. In Singhal, A., Cody, M.J., Rogers,
E.M., and Sabido, M. (Eds.), Entertainment-Education and Social Change:
History, Research, and Practice. Mahwah, NJ: Lawrence Erlbaum Associates.
Rogers, E. M. (2003). Diffusion of Innovations (5th Edition). New York: Free Press.
Rogers, E. M., and Kincaid, D. L. (1981). Communication Networks. Toward a New
Paradigm for Research. New York: The Free Press, A Division of Macmillan
Publishing Co., Inc.
Rogers, E. M., and Shoemaker, F. (1971). Communication of innovations: A cross-
cultural approach. New York: The Free Press.
Rogers, E. M., Vaughan, P. W., Swalehe, R. M. A., Rao, N., Svenkrud, P., and Sood,
S. (1999). Effects of an entertainment-education soap opera on family
planning behavior in Tanzania. Studies in Family Planning, 30 (3), 193-211.
Rosenthal, D., Moore, S., and Flynn, I. (1991). Adolescent self-efficacy, self-esteem
and sexual risk-taking. Journal of Community and Applied Social
Psychology, 1, 77-88.
Schiller, D. (1999). Digital Capitalism: Networking the global market system.
Massachusetts Institute of Technology.
Schunk, D.H., and Meece, J.L. (2006). Self-efficacy development in adolescence. In
Pajares, F. and Urdan, T. (Eds.) Self-efficacy beliefs of adolescents.
Information Age Publishing: Greenwich, Connecticut.
Schwarzer, R. and Luszczynska, A. (2006). Self-efficacy, adolescents’ risk-taking
behaviors, and health. In Pajares, F. and Urdan, T. (Eds.) Self-efficacy beliefs
of adolescents. Information Age Publishing: Greenwich, Connecticut.
Schwarzer, R., and Renner, B. (On-line). Health-Specific self-efficacy scales.
Downloaded June 16, 2005 from http://userpage.fu-berlin.de/~health.
Scott, J. (2000). Social Network Analysis (2nd ed.). Thousand Oaks, CA: SAGE
Publications.
Servaes, J. (1999). Communication for development: One world, multiple cultures.
Cresskill, NJ: Hampton Press, Inc.
135
J. Servaes, T. L. Jacobson and S. A. White (Eds.) (1996). Participatory
Communication for Social Change (pp. 150-161). Thousand Oaks, CA: Sage
Publications.
Servon, L.J. (2002). Bridging the digital Divide: Technology, community, and public
policy. Massachusetts: Blackwell Publishing.
Sherry, J. L. (2001). The effects of violent video games on aggression: A meta-
analysis. Human Communication Research, 27 (3), 409-431.
Sherry, J. L. (2007). Violent video games and aggression: Why can’t we find effects?
In Preiss, R. W., Gayle, B. M., Burrell, N., Allen, M., and Bryant, J. (Eds.)
Mass Media Effects Research. Erlbaum: New Jersey.
Singh, S., and Wulf, D. (1991). Sexual activity, union and child-bearing among
adolescent women in the Americas.. International Family Planning
Perspectives, 17 (4), 137-144.
Singhal, A., and Rogers, E. (2001). The Entertainment- Education Strategy in
Communication Campaigns by. In Rice, R.E. and Atkin, C.K. (Eds.) Public
Communication Campaigns. Thousand Oaks, Ca.: Sage. (pp. 343-356)
Singhal, A. and Rogers, E. M.. (2002). A theoretical agenda for Entertainment-
Education. Communication theory, 12, 117-135.
Singhal, A. and Rogers, E. M. (1990). The entertainment-education strategy in
communication campaigns. In R. Rice and C. Atkin, (Eds.), Public
Communication Campaigns (pp. 343-356). London: Sage.
Singhal, A., and Rogers, E. M. (2004). The Status of Entertainment-Education
Worldwide. In Singhal, A., Cody, M. J., Rogers, E. M., and Sabido, M. (Eds.),
Entertainment-Education and Social Change: History, Research, and Practice
(pp. 3-20). Mahwah, NJ: Lawrence Erlbaum Associates.
Singhal, A., Cody, M.J., Rogers, E.M., and Sabido, M. (Eds.), Entertainment-
Education and Social Change: History, Research, and Practice (pp. 351-376).
Mahwah, NJ: Lawrence Erlbaum Associates.
Singhal, A., Sharma, D., Papa, M. J., and Witte. K. (2004). Air cover and ground
mobilization: Integrating entertainment-education broadcasts with
community listening and service delivery in India. In Singhal, A., Cody, M.J.,
Rogers, E.M., and Sabido, M. (Eds.), Entertainment-Education and Social
Change: History, Research, and Practice (pp. 351-376). Mahwah, NJ:
Lawrence Erlbaum Associates.
Sharf, B. F., Freimuth, V. S., Greenspon, P., and Plotnick, C. (1996). Confronting
cancer on thirty something: Audience response to health content on
entertainment television. Journal of Health Communication,1,157–172.
136
Sherry, J. L. (2002). Media saturation and entertainment-education. Communication
Theory, 12, 206-224.
Skuse, Andrew (2001). Information Communication Technologies, Poverty and
Empowerment, Social Development Department, Dissemination Note NO. 3,
DFID.
Sood, S. Menard, T., and Witte, K. (2004). The Theory Behind Entertainment-
Education. In Singhal, A., Cody, M. J., Rogers, E. M., and Sabido, M. (Eds.),
Entertainment-Education and Social Change: History, Research, and Practice
(pp. 117-148). Mahwah, NJ: Lawrence Erlbaum Associates.
Soul City. (2001). Social Change: The Soul City experience. Johannesburg, South
Africa: Soul City, the Institute for Health and Development Communication.
Storey, J.D., and Jacobson, T.L. (2004). Entertainment-education and participation:
Applying Habermas to a case study of a population program in Nepal. In
Singhal, A., Cody, M.J., Rogers, E.M., and Sabido, M. (Eds.), Entertainment-
Education and Social Change (pp. 417-434). Mahwah, NJ: Lawrence
Erlbaum Associates.
Thomas, R., Cahill, J., and Santini, L. (1997). Using an interactive computer game to
increase skill and self-efficacy regarding safer sex negotiation: Field test of
results. Health Education & Behavior, 24.
Tichenor, P.J., Donohue, G.A., and Olien, C.N. (1970) Mass media flow and
differential growth in knowledge. Public Opinion Quarterly, 34 (2), 159-70.
Trujillo, M. (2001). The global digital divide: exploring the relation between core
national computing and national capacity and progress in human
development over the last decade. Unpublished doctoral dissertation.
UNAIDS. (2005) Global summary of the HIV/AIDS epidemic, 2005. New York.
United Nations (2003). Human development report. United Nations Development
Program (UNDP). New York.
United Nations (2005). The Millennium Development Goals Report 2005. New York.
United Nations Development Programme (2001). Global Report on Human
Development. Geneva. Downloaded August 16, 2003 from
hdr.undp.org/reports/.
US Internet Council. (1999). State of the Internet: USIC's report on use & threats in
1999. Downloaded August 16, 2003 from www.usinternetcouncil.org.
Valente, T. W. (1996). Network models of the diffusion of innovations. Hampton
Press, Inc. New Jersey.
Valente, T. W. (2002). Evaluating health promotion programs. Oxford University
Press: New York, NY.
137
Valente, T. W. (2005). Network models and methods for studying diffusion of
innovations. In Carrington, P., Wassermann, S. and Scott, J. (Eds.), Models
and methods in social network analysis. Cambridge: Cambridge University
Press. (pp. 98-116).
Valente, T. W., Kim, Y. M., Lettenmaier, C., Glass, W. and Dibba, Y. (1994). Radio
and the promotion of family planning in The Gambia. International Family
Perspectives Planning, 20, 96-100.
Valente, T. W., and Saba, W. P. (1998). Mass media and interpersonal influence in a
reproductive health communication campaign in Bolivia. Communication
Research, 25, (1), 96-124.
Vaughan, P. W., and Rogers, E. M (2000). A staged model of communication
effects: Evidence from an entertainment-education radio soap opera in
Tanzania. Journal of Health Communication , 5, 203-227.
Vaughan, P.W., Regis, A., and St. Catherine, E. (2000). Effects of an entertainment-
education radio soap opera on family planning and HIV prevention in St.
Lucia. International Family Planning Perspectives, 26 (4), 148-157.
Vorderer, P. (2000) Interactive Entertainment and Beyond. In D. Zillmann and P.
Vorderer (Eds.), Media Entertainment: The psychology of its appeal (pp.
235-248). Mahwah, NJ: Lawrence Erlbaum Associates.
Wasserman, S., and Faust, K. (1994). Social network analysis: Methods and
applications. New York: Cambridge University Press.
Wellman, B. (2001). Physical place and cyber place: The rise of networked
individualism. International journal for Urban and Regional Research, 25 (2),
227-252.
Wellman, B., and Berkowitz, S. D. (1988) Social structures: A network approach.
Cambridge: Cambridge University Press.
Wellman, B., Haase, A.Q., Witte, J. and Hampton, K. (2001). Does the Internet
increase, decrease, or supplement social capital? Social networks,
participation, and community commitment. American Behavioral Scientist,
45, 436-455.
WHO (2005). The World Health Report 2005. Geneva.
Wireless Internet Institute (W2i) (2003). The wireless Internet opportunity for
developing countries. Boston: The Wireless Internet Institute, World Bank
infoDev & the United Nations ICT Task Force.
Witte, K. (1996). Notes from the field: Does publishing in academic journals make a
difference? Journal of Health Communication, 1, 221-226.
Wolf, L., Castro, C., Navarro, J., and Garcia, N. (2002). Television for secondary
education: Experience of Mexico and Brazil. In W. Haddad (Ed.),
138
Technologies for education: Potentials, parameters, and prospects (pp. 144-
152). Paris: UNESCO.
Yeager, B. A. C.; Huttly1, S. R. A.; Diaz, J.; Bartolini, R.; Marin, M., and Lanata, C.
F. (2002) An intervention for the promotion of hygienic feces disposal
behaviors in a shanty town of Lima, Peru . Health Education Research, 17 (6),
761-773.
139
APPENDICES
140
APPENDIX A
English-language Questionnaire
General instructions: The objective of this questionnaire is to evaluate the
effectiveness of the program in which you are participating. Your answers will be
confidential, so feel free to honestly answer your true beliefs.
Mark with a (X) inside the parentheses, the option that comes closest to your
personal beliefs, feelings or opinions on the topic, in the most sincere way that you
can. Pay a lot of attention to the questions, and state your answer clearly before
proceeding to the next question.

1. Please indicate your number from the roster. ______________

2. Gender: ( ) Male ( ) Female

3. Using the following scale, mark your age?

( ) 14 years old ( ) 15 years old ( ) 16 years old
( ) 17 years ( ) 18 years ( ) 19 years

4. On a typical weekday, Monday through Friday, how many hours do you
watch television each day?
________Hours [Choose hours between 0 to 24]

5. On a typical weekend day, how many total hours do you watch television each
day?
________Hours [Choose hours between 0 to 24]

6. In the past seven days, on how many days did you…

0 1 2 3 4 5 6 7
Read a newspaper?………………………………..

Watch the national news on television?…………..

Watch the international news on television?……..

Listen to radio—telenovellas, talk shows or news?.

Use the Internet for email and chat?………………

Use the Internet, other than email and chat?………

141
7. In general, how easy or difficult is it for you to access the following resources?
VERY VERY
DIFFICULT EASY
(a) Computer………1 2 3 4 5 6 7 8 9 10
(b) Internet……….. 1 2 3 4 5 6 7 8 9 10

8. Thinking about all of the different ways of communicating and getting
information, what are the three most important ways that you get medical
and health information for yourself or for your family? (Circle three only)

Television…………………………………… 1
Internet………………………………………. 2
Radio………………………………………… 3
Newspapers………………………………….. 4
Magazines…………………………………… 5
Cinema………………………………………. 6
Brochures ……………….…………………... 7
Friends or family……………………………. 8
People at work………………………………. 9
Doctors or other health care workers……….. 10
Health organizations………………………… 11
Please specify ______________
______________
Other sources of health information………… 14
Please specify ______________
______________

142
9. You may have different sources for health information How often have you
obtained health information in the past 30 days from the following
sources? Would you say…

Not at all Less than
once per
week
Once per
week
A few
times a
week
Television
Internet
Radio
Newspapers
Magazines
Cinema
Brochures
Friends or family
People at work
Doctors/health workers
Health organizations

10. Thinking about the past 30 days, how much have you heard from the media
about (including television, radio, newspapers, magazines and the
Internet ):

Not at all A little Some A lot
Family planning
HIV or AIDS
Sexuality
Contraception

143
11. How much do you trust the information about health from each of the
following:

Not at all A little Some A lot
Television
Internet
Radio
Newspapers
Magazines
Cinema
Brochures, pamphlets/flyers
Friends or family
People at work
Doctors/health care workers
Health organizations

12. How much sexual content do you think each of the following has?

Not at all A little Some A lot
Television
Cinema
Internet
Radio
Newspapers
Magazines
Video Games

13. At the moment I have a romantic relationship, and am part of a stable
couple?

Yes No Don’t Know

144
14. The following is a list of statements that people can make about their sexual
habits. Using the scale provided, indicate the extent to which you agree or
disagree with the following statements. (Circle one option for each statement).
STRONGLY STRONGLY
DISAGREE AGREE
I find it difficult to get contraceptives
1 2 3 4 5 6 7
I find it difficut to use contraceptives
1 2 3 4 5 6 7
My friends think it is a good idea to protect against
unwanted pregnancies
1 2 3 4 5 6 7
When having sex, I tend to avoid thoughts about
HIV/AIDS
1 2 3 4 5 6 7
I can talk to my sexual partner about preventing HIV/AIDS
1 2 3 4 5 6 7
I can talk to my sexual partner about using condoms
1 2 3 4 5 6 7
I am able to use condoms to prevent HIV/AIDS
1 2 3 4 5 6 7
I can talk about family planning methods with my sexual
partner(s)
1 2 3 4 5 6 7
I have multiple sexual partners 1 2 3 4 5 6 7
It is difficult to tell my sexual partner to that we should
protect ourselves with a condom or another birth-control
method during sex
1 2 3 4 5 6 7
I feel shame when explaining to my sexual partner how I
would like to have a sexual relationship
1 2 3 4 5 6 7

15. Have you heard of the following methods of family planning. Please check
all those you know of:
Sterilization
Pill
IUD
Injections
Implant
Condom
Rhythm method
Withdrawal
Traditional methods

16. An HIV test sees whether you have the virus that causes AIDS. Have you
heard of the HIV test?

Yes No Don’t Know

17. Have you ever had a test for HIV (the test for AIDS)?
145

Yes No Don’t Know
18. Is there anything a person can do to avoid getting the HIV/AIDS virus?

Yes No Don’t Know

19. What can a person do to avoid getting HIV/AIDS. (Check all that apply).

Abstain from sex

Avoid anal sex

Use condoms during sex

Avoid sharing razor blades

Avoid sex with prostitutes

Avoid sex with multiple partners

Shared needles usage with drugs

Check origin of blood in transfusions

Avoid sex with homosexuals

Only have oral sex with infected people

Avoid kissing infected people

Go away when an infected person sneezes or
coughs

Take care to avoid mosquito bites

Sleep in a different bed from an infected
person

Avoid oral sex

Have sex with single partner

None of the above

I don’t know

20. In general, how many close friends do you have? ("Close friends" include
relatives or non-relatives that you feel at ease with, can talk to about private
matters, and can call on for help.)

Number of close friends ________________

21. Please list the 5 people you talk to the most. Please choose their numbers
from the roster.

1. _______ 2. _______ 3. _______ 4. _______ 5. _______

22. Please list the 5 people you consider your closest friends. Please choose their
numbers from the roster.
1. _______ 2. _______ 3. _______ 4. _______ 5. _______

23. Please list the 5 people you ask for advice, or would consider asking, about
issues concerning family planning, sexuality, contraception and HIV/AIDS.
Please choose their numbers from the roster.
1. _______ 2. _______ 3. _______ 4. _______ 5. _______

146
24. The following is a list of statements that people can make about sexuality.
Using the scale provided, indicate the extent to which you agree or disagree with
the following statements. Circle one option for each statement.

STRONGLY STRONGLY
DISAGREE AGREE

Birth control methods are effective in preventing pregnancy
1 2 3 4 5 6 7

Using birth control methods to prevent pregnancy is good
1 2 3 4 5 6 7

I will try and use condoms if I have sex in the future
1 2 3 4 5 6 7

To prevent pregnancies, I intend to be celibate
1 2 3 4 5 6 7

I am at risk for getting HIV/AIDS
1 2 3 4 5 6 7

The possibility of getting AIDS frightens me
1 2 3 4 5 6 7

Using condoms prevents HIV/AIDS and sexually transmitted
diseases
1 2 3 4 5 6 7

My friends think it is a good idea to protect against
HIV/AIDS
1 2 3 4 5 6 7

Condoms help to plan the family and protects from AIDS
1 2 3 4 5 6 7

Birth-control methods are bad because they cause organ
dysfunctions sometimes cancer
1 2 3 4 5 6 7

Birth-control methods have different effects in each person
1 2 3 4 5 6 7

Sexually transmitted diseases and AIDS only happens to
homosexuals and sexual workers (prostitutes).
1 2 3 4 5 6 7

A woman can suggest to her partner to use condoms in a
sexual relationship
1 2 3 4 5 6 7

I believe that my friends have already had sexual
relationships
1 2 3 4 5 6 7

In the last month, I have shared my knowledge and opinions
about sexuality with others
1 2 3 4 5 6 7

147
25. This questionnaire is designed to help us get a better understanding of the
kinds of things that are difficult for students. Please rate how certain you are
that you can do each of the things described below by writing the appropriate
number. Rate your degree of confidence by recording a number from 0 to 100 using
the scale given below:

Cannot Moderately Highly Certain
do at all can do can do
0 10 20 30 40 50 60 70 80 90 100

(0-100)
Make and keep friends of the opposite sex ______
Carry on conversations with others ______
Work well in a group ______
Get a friend to help me when I have social problems ______

Resist peer pressure to smoke cigarettes ______
Resist peer pressure to drink beer, wine, or liquor ______
Resist peer pressure to smoke marijuana ______
Resist peer pressure to have sexual intercourse ______

The statements below describe situations that commonly arise in groups of friends.
For each situation please rate how certain your group of friends, working together as
a whole, can manage them effectively.

Support each other in times of stress ______
Help each other to achieve their personal goals ______
Build respect for each other's particular interests ______
Build trust in each other ______

26. In general, would you say that your health is …

(1) Excellent (2) Good (3) Fair (4) Poor (5) Very poor

27. In the past 12 months, how much of a hardship was it for you or your family
personally to pay health care costs?

(1) Not at all (2) A little difficult (3) Some (4) Very difficult

28. If you wanted to know about a health problem or disease, do you know of
any resource centers or organizations that you could go to for information?
Yes No Don’t Know
148
29. The following is a list of statements that people can make about their sexual
habits. Using the scale provided, indicate the extent to which you agree or
disagree with the following statements. Circle one option for each statement.
STRONGLY STRONGLY
DISAGREE AGREE
Family planning methods cost too much.
1 2 3 4 5 6 7
Family planning methods are inconvenient
1 2 3 4 5 6 7
Family planning methods are hard to get
1 2 3 4 5 6 7
It is embarrassing to get family planning services
1 2 3 4 5 6 7
My religion is opposed to family planning methods
1 2 3 4 5 6 7
I know where to get family planning methods
1 2 3 4 5 6 7
Health workers understand my concern about family
planning methods
1 2 3 4 5 6 7
I can talk about family planning methods with my friends
and family
1 2 3 4 5 6 7
My religion encourages us to protect ourselves from
HIV/AIDS
1 2 3 4 5 6 7
It is better not to talk about pregnancy in my community
1 2 3 4 5 6 7

30. The following is a list of statements that people can make about their use of
the game you just played. Using the scale provided, indicate the extent to which
you agree or disagree with the following statements. Circle one option for each
statement.
STRONGLY STRONGLY
DISAGREE AGREE
I find it easy to use the game according to my needs.
1 2 3 4 5 6 7
I find it easy to get access to information by playing the
game.
1 2 3 4 5 6 7
I find the game to be useful for health information.
1 2 3 4 5 6 7
I trust the information contained in the game
1 2 3 4 5 6 7
I will recommend the use of the game to my friends.
1 2 3 4 5 6 7

149
APPENDIX B
Spanish-language Questionnaire
Instrucciones generales: El objetivo de esta encuesta es evaluar la efectividad de los
materiales educativos que estás utilizando. Tus respuestas serán confidenciales, con
la finalidad que te sientas libre para contestar.

Marca con una (X) dentro de los paréntesis la opción que se asemeje más a tus
creencias, sentimientos u opiniones en cada uno de los temas. Presta mucha atención
a las preguntas y revisa la claridad de tus respuestas antes de proceder a la próxima
pregunta.

1. Por favor indica tu número de la lista. ______________

2. Género: ( ) Varón ( ) Mujer

3. Marca tu edad?

( ) 16 años ( ) 17 años ( ) 18 años ( ) más de 18 años

4. En un día típico, de lunes a viernes, ¿cuántas horas miras televisión cada día?

________Horas [escoge entre 0 a 24 horas]

5. ¿Cuántas horas miras la televisión cada fin de semana?

________Horas [escoge entre 0 a 24 horas]

6. En los últimos siete días, ¿cuántos días ...
0 1 2 3 4 5 6 7
Leíste un periódico?…………………………

Viste las noticias nacionales en la televisión?

Viste las noticias internacionales en la televisión?

Escuchaste radio o noticias?………....................

Usaste el correo electrónico y el chat?……

Usaste la internet, por otro motivo?..........

150
7. En general ¿qué tan fácil o difícil es para ti acceder a los siguientes recursos?.
MUY MUY
DIFÍCIL FÁCIL
(a) La computadora…1 2 3 4 5 6 7 8 9 10
(b) El internet…... 1 2 3 4 5 6 7 8 9 10

8. Piensa sobre las diferentes formas de comunicar y conseguir información,
¿cuáles son tus tres más importantes formas de conseguir información sobre
salud, para ti o para tu familia?

Televisión…………………………………… 1
Internet………………………………………. 2
Radio………………………………………… 3
Periódicos .………………………………….. 4
Revistas………………………..…………… 5
Cine…………………………………. ……. 6
Folletos……………….……………...……... 7
Amigos o familia……………………………. 8
Estudiantes de tu salón de clase ……………………9
Doctores u otro personal de salud ..........…… 10
Instituciones de salud……………..………… 11
Por favor especifica otra forma _________________

9. Tienes diferentes fuentes de información ¿con qué frecuencia obtuviste
información sobre salud en los últimos 30 días?

Nada Menos de una
vez por semana
Una vez por
semana
Unas veces
por semana
Televisión
Internet
Radio
Periódicos
Revistas
Cine
Folletos
Amigos o familia
Estudiantes de tu salón de
clase

Doctores u otro personal
de salud

Instituciones de salud

151
10. Piensa en los últimos 30 días, ¿cuánto escuchaste en los medios de
comunicación (televisión, radio, periódicos, revistas e internet) sobre:

Nada Un poco Algo Mucho
Planificación familiar
VIH o SIDA
Sexualidad
Anticoncepción

11. ¿Cuánto confías en la información sobre salud que cada uno de los
siguientes brinda:

Nada Un poco Algo Mucho
Televisión
Internet
Radio
Periódicos
Revistas
Cine
Folletos
Amigos o familia
Estudiantes de tu salón de clase
Doctores u otro personal de salud
Instituciones de salud

12. ¿Cuánto contenido sobre sexualidad piensas que tiene cada uno de los
siguientes?

Nada un poco Algunos mucho
La televisión
El cine
Internet
La radio
Los periódicos
Las revistas
Los juegos de videos

13. En este momento yo tengo una relación de pareja, enamorado (a) estable.

Sí No No se

152
14. La siguiente es una lista de declaraciones que las personas pueden hacer
sobre sus hábitos sexuales. Usando la escala, indica hasta que punto estás de
acuerdo o en desacuerdo con las siguientes declaraciones.

FUERTEMENTE FUERTEMENTE
EN DESACUERDO DE ACUERDO
Encuentro difícil conseguir métodos anticonceptivos 1 2 3 4 5 6 7

Encuentro difícil usar métodos anticonceptivos 1 2 3 4 5 6 7

Mis amigos piensan que es una idea buena protegerse contra los embarazos no
1 2 3 4 5 6 7
Cuando tengo relaciones sexuales, tiendo a evitar pensar en el VIH/SIDA
1 2 3 4 5 6 7
Puedo hablar con mi pareja sexual acerca de prevenir el SIDA 1 2 3 4 5 6 7
Puedo hablar con mi pareja sexual sobre usar 1 2 3 4 5 6 7

Estoy dispuesto a usar los condones para prevenir el SIDA 1 2 3 4 5 6 7

Puedo hablar sobre los métodos de planificación familiar con mi pareja sexual
1 2 3 4 5 6 7
Tengo muchas parejas sexuales 1 2 3 4 5 6 7

Es difícil decirle a mi pareja sexual que debemos protegernos con un condón u otro método
1 2 3 4 5 6 7
anticonceptivo durante las relaciones sexuales 1 2 3 4 5 6 7

Siento vergüenza al explicar a mi pareja sexual cómo me gustaría tener nuestras relaciones sexuales
1 2 3 4 5 6 7

15. ¿Has oído hablar de los siguientes métodos de planificación familiar? Por
favor verifica todos aquellos que conoces :
La esterilización
La píldora
T de Cobre
Los inyectables
Implantes hormonales (Norplant)
El condón
El método de ritmo
El coito interrumpido (eyacular fuera de la vagina)
Los métodos naturales

16. Una prueba de VIH se utiliza para verificar el virus que causa el SIDA.
¿Has oído hablar de la prueba de VIH?
Sí No No se

153
17. ¿Alguna vez te hiciste una prueba para saber si tenías el VIH (examen del
SIDA)?
Sí No No se
18. ¿Hay algo que una persona puede hacer para evitar contraer el virus de
VIH?
Sí No No se

19. ¿Qué puede hacer una persona para evitar contraer el VIH? (puedes marcar
varias).

Abstenerse de tener relaciones sexuales
Evitar relaciones anales
Usar condones durante las relaciones sexuales
Evitar compartir las hojas de la navaja de afeitar
Evitar relaciones sexuales con prostitutas
Evitar tener relaciones sexuales con muchas parejas
Compartir agujas por drogas
Chequear la sangre en las transfusiones
Evitar tener relaciones sexuales con homosexuales
Tener sexo oral con las personas infectadas
Sólo teniendo relaciones orales con gente infectada
Evitar besar a las personas infectadas
Evitar a una persona infectada cuando estornuda o tose
Tener cuidado para evitar la picadura de un mosquito
Dormir en una cama diferente de una persona infectada
Evitar el sexo oral
Tener relaciones sexuales solo con una pareja
Ninguna de las anteriores
No sé

20. En general, ¿cuántos amigos cercanos tienes? (Los amigos “íntimos”
incluyen a parientes o compañeros de clase, con los que te sientes a gusto,
puedes hablar sobre temas privados y puedes llamarlos para pedir ayuda.)

Número de amigos íntimos ________________

21. Por favor menciona a tus 5 amigos cercanos. Por favor escoge sus números
de la lista.
1. _______ 2. _______ 3. _______ 4. _______ 5. _______

22. Por favor liste a las 5 personas que usted considera a sus amigos más íntimos.
Por favor escoja sus números de la lista.
1. _______ 2. _______ 3. _______ 4. _______ 5. _______

154
23. Por favor menciona a las 5 personas a las que recurres para un consejo o
consideras puedes preguntarles sobre sexualidad, métodos anticonceptivos,
VIH/SIDA. Por favor escoge sus números de la lista.
1. _______ 2. _______ 3. _______ 4. _______ 5. _______

155
24. La siguiente es una lista de declaraciones que las personas puede hacer
sobre sexualidad. Usando la escala, indica hasta que punto está de acuerdo o en
desacuerdo con las siguientes declaraciones.

Seleccione una opción para cada declaración

FUERTEMENTE FUERTEMENTE
DESACUERDO DE ACUERDO
Los métodos anticonceptivos son efectivos para prevenir un embarazo
1 2 3 4 5 6 7
Es bueno usar métodos anticonceptivos para prevenir embarazos
1 2 3 4 5 6 7
Las campañas que promueven la anticoncepción son manipulantes y poco confiables
1 2 3 4 5 6 7
Intentaré usar los condones si tengo relaciones sexuales en el futuro
1 2 3 4 5 6 7
Para prevenir los embarazos, yo intento tener sólo una pareja
1 2 3 4 5 6 7
Estoy en riesgo de contraer el VIH/SIDA 1 2 3 4 5 6 7

Tener SIDA estropearía mi futuro 1 2 3 4 5 6 7

La posibilidad de contraer SIDA, me asusta 1 2 3 4 5 6 7

Los condones son eficaces para prevenir el VIH/SIDA
1 2 3 4 5 6 7
Usar los condones previene el VIH y las infecciones de transmisión sexual
1 2 3 4 5 6 7
Mis amigos piensan que es buena idea protegerse contra el VIH/SIDA
1 2 3 4 5 6 7
Los condones ayudan a planificar la familia y proteger del VIH/SIDA
1 2 3 4 5 6 7
Los métodos anticonceptivos a veces causan trastornos como el cáncer
1 2 3 4 5 6 7
Los métodos anticonceptivos tienen efectos diferentes en cada persona
1 2 3 4 5 6 7
Las infecciones de transmisión sexual y el SIDA sólo les sucede a los homosexuales y a las
prostitutas. 1 2 3 4 5 6 7

Una mujer puede sugerir a su pareja sexual usar los condones en una relación sexual
1 2 3 4 5 6 7
Yo creo que mis amigos ya tuvieron relaciones sexuales
1 2 3 4 5 6 7
En el último mes, he compartido mis conocimientos y opiniones sobre sexualidad con otros
1 2 3 4 5 6 7

156
25. Esta encuesta está diseñada para ayudar a tener un mejor entendimiento de
las cosas que son difíciles para los estudiantes. Por favor, tasa cuan cierto tu
estás de poder hacer cada una de las cosas descritas de abajo escribiendo el
número apropiado.

Evalúa tu grado de confianza señalando los números del 0 a 100 de la siguiente
escala :

No puedo hacer todo / Moderadamente puedo hacerlo / Muy Cierto, puede hacerlo
0 10 20 30 40 50 60 70 80 90 100

(0-100)
Hacer y conservar a los amigos del sexo opuesto ______
Hacer y conservar a los amigos del mismo sexo ______
Mantener las conversaciones con otros ______
Trabajar bien en grupo
Conseguir que un amigo me ayude cuando yo tenga problemas ______

Resistirse a la presión del grupo para hacer cosas en el colegio que pueden traerme
problemas ______
Resistirse a la presión del grupo para fumar cigarros ______
Resistirse a la presión del grupo para cerveza, vino o licor ______
Resistirse a la presión del grupo para fumar marihuana ______
Resistirse a la presión del grupo para tener relaciones sexuales ______

Las siguientes declaraciones describen situaciones que comúnmente se dan en los
grupos de amigos. Para cada situación evalúa que tan cierto tu grupo de amigos
trabaja junto como un todo y pueden manejarlo efectivamente.

En conjunto, en el grupo de amigos:
Se apoyan unos a otros en tiempos de tensión ______
Se ayudan unos a otros para logras sus metas personales ______
Construyen el respeto entre unos y otros de acuerdo a sus intereses particulares
Se construye la confianza de unos a otros ______
Aceptan las necesidades de cada miembro para su independencia ______

26. En general, podrías decir que tu salud es…

( ) Excelente ( ) Buena ( ) Medianamente buena ( ) Pobre ( ) Muy pobre

157
27. ¿Cuánto ha sido molestoso para ti o tu familia pagar por el cuidado de la
salud en los últimos 12 meses?

( ) De ninguna manera ( ) un poco difícil ( ) Algo ( ) Muy difícil

28. ¿Si quisieras saber sobre un problema de salud o enfermedad, conoces
algunos centros u organizaciones de salud a las que podrías ir a por
información?

Sí No No se

29. Lo siguiente es una lista de declaraciones que las personas pueden hacer
sobre sus hábitos sexuales. Usando la escala, indica hasta que punto estás de
acuerdo o en desacuerdo con las siguientes declaraciones.

FUERTEMENTE FUERTEMENTE
DISCREPE ESTÁ DE ACUERDO
Los métodos de la planificación familiar son demasiado costosos.
1 2 3 4 5 6 7
Los métodos de la planificación familiar son inconvenientes
1 2 3 4 5 6 7
Los métodos de la planificación familiar son difíciles de conseguir
1 2 3 4 5 6 7
Es vergonzoso conseguir servicios de planificación familiar
1 2 3 4 5 6 7
Mi religión se opone a los métodos anticonceptivos
1 2 3 4 5 6 7
Se dónde obtener métodos
1 2 3 4 5 6 7
Los profesionales de la salud entienden mi inquietud sobre los métodos
anticonceptivos. 1 2 3 4 5 6 7

Puedo hablar sobre los métodos anticonceptivos con mis amigos y mi familia
1 2 3 4 5 6 7
Mi religión promueve protegernos de VIH/SIDA
1 2 3 4 5 6 7
Es mejor no hablar sobre el embarazo en mi comunidad
1 2 3 4 5 6 7

158
30. La siguiente es una lista de declaraciones que las personas pueden hacer
sobre el uso del juego que jugaste. Usando la escala, indica hasta que punto
estás de acuerdo o en desacuerdo con las declaraciones siguientes. Marca con un
círculo cada una de ellas.
FUERTEMENTE FUERTEMENTE
DISCREPE ESTÁ DE ACUERDO
Encuentro fácil de usar el juego según mis necesidades. 1 2 3 4 5 6 7

Encuentro fácil de acceder a la información jugando el juego. 1 2 3 4 5 6 7

Encuentro útil este juego para la información de salud. 1 2 3 4 5 6 7

Confío en la información contenida en el juego 1 2 3 4 5 6 7

Recomendaré el uso del juego a mis amigos. 1 2 3 4 5 6 7

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Mobile self-tracking for health: validating predictors, effects, mediator, moderator, and social influence

PDF

Building personal wellness communities: meaningful play in the everyday life of a network society

PDF

Contagious: social norms about health in work group networks

PDF

Civic games with 'local fit': embedding with real‐world neighborhoods and place‐based networks

PDF

The formation and influence of online health social networks on social support, self-tracking behavior and weight loss outcomes

PDF

Technical base, interests, and power in the two-level game of international telecom standards setting: the political economy of China's initiatives

PDF

The coevolution of multimodal, multiplex, and multilevel organizational networks in development communities

PDF

Examining internal communication practices to improve teachers’ motivation to use ICTs in instruction: a case study of Peruvian secondary schools

PDF

Social network norms and HIV risk behaviors among homeless youth in Los Angeles, California

PDF

The social groups approach to quitting smoking: An examination of smoking cessation online and offline through the influence of social norms, social identification, social capital and social support

PDF

Building social Legoland through collaborative crowdsourcing: marginality, functional diversity, and team success

PDF

Communicating organizational knowledge in a sociomaterial network: the influences of communication load, legitimacy, and credibility on health care best-practice communication

PDF

Mapping out the transition toward information societies: social nature, growth, and policies

PDF

Social media and health: social support and social capital on pregnancy-related social networking sites

PDF

Crisis and stasis: understanding change in online communities

PDF

The good, the bad, and the longitudinal: testing dynamic prosocial and toxic behaviors in online commercial games

PDF

Game recognize game: performative archives and alternate reality games

PDF

Using virtual environments to unobtrusively measure real-life risk-taking: findings and implications for health communication interventions

PDF

The evolution of knowledge creation online: what is driving the dynamics of Wikipedia networks

PDF

Social support in Cambodia: the role of peer educators in behavior change

Asset Metadata

Creator Chib, Arul I. (author)

Core Title Network influences in health initiatives: multimedia games for youth in Peru

School Annenberg School for Communication

Degree Doctor of Philosophy

Degree Program Communication

Publication Date 08/07/2007

Defense Date 06/20/2007

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag diffusion,games,HIV,ICT,network,OAI-PMH Harvest,Peru

Place Name Peru (countries)

Language English

Advisor Cody, Michael J. (committee chair), Bar, Francois (committee member), Valente, Thomas W. (committee member)

Creator Email chib@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m753

Unique identifier UC1379133

Identifier etd-Chib-20070807 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-540844 (legacy record id),usctheses-m753 (legacy record id)

Legacy Identifier etd-Chib-20070807.pdf

Dmrecord 540844

Document Type Dissertation

Rights Chib, Arul I.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu