WHY DOES SOCIAL CONTEXT MATTER?
Integrating Innovative Technologies with Best Practice Models for Public and
Behavioral Health Promotion
Luba Botcheva and Shobana Raghupathy
Sociometrics Corporation, Los Altos, California, U.S.A.
{lbotcheva, shobana}@socio.com
Keywords: Technology, Innovation, Acceptance, Integration, Adoption, Behaviour, Change, Health, Social context.
Abstract: In this paper we will present a framework that is intended to guide synthesis of different theoretical
perspectives for the purpose of developing strategies for integrating IT use in diverse social settings. First,
we will briefly review existing theoretical models grounded in behavioural science; and present our
company’s approach for development of products using technology innovation that take in account the
individual, organizational and contextual community characteristics. Secondly, we will illustrate this
approach with three case study examples in the fields of public/behavioral health and education. Finally, we
will conclude with theoretical and practical considerations that can be used by IT developers to maximize
adoption and implementation of innovative technologies.
1 INTRODUCTION
It is universally acknowledged that information and
communication technologies (ICTs or IT) hold huge
potential for enhancing the effectiveness of services
in different sectors. Their ability to reach new
populations, improve communications, and
transform service delivery can increase the
effectiveness and social impact of private, public
and non-profit initiatives. Yet, the positive effects of
innovative IT will only be fully realized if, and
when, they are widely spread and used. It is a well-
accepted fact that the existence of technology does
not guarantee its utilization. Attempts to promote the
adoption and diffusion of innovative IT have often
failed due to a lack of understanding of the factors
that affect acceptance and use of technology by
individuals and organizations. A notable example is
the recent failure of the One Laptop Per Child
(OLPC) project, which aimed to distribute millions
of $100 laptops to disadvantaged school children but
failed to anticipate the social and institutional
problems that could arise in trying to diffuse
technology in the developing country context
(Kraemer, Dedrick, & Sharma, 2009).
As the OLPC case demonstrates, investigating
social context is vital to understanding the
acceptance and use of technologies. Researchers
have often addressed the issue of why individuals
and organizations who would benefit from
technological systems do not use them but
traditionally most of the research has focused on
technological factors and has rarely been applicable
to different sectors and social contexts (Al-Gahtani,
2008). With the rapid utilization of IT in different
spheres of life and across geographical and
economic dimensions, best practice models have
shifted focus to the potential adopter and the
organization or community into which the
technology will be integrated. An adopter based,
instrumentalist approach incorporating both macro-
and micro-level perspectives now appears to be the
most widely used to promote the adoption and
diffusion of innovative ITs.
However, a gap exists between these best
practice models and IT adoption strategies. In
particular, the non-profit and public sectors as a
whole lag behind the private sector in the adoption
of technologies. There has been little scholarly
research into the IT adoption in the non-profit and
public sectors. This paper discusses ways that
technology acceptance models can be utilized to
develop multi-level approaches for facilitating IT
adoption in diverse social settings (educational,
health and community-based organizations) with
special emphasis of the contextual characteristics
that determine the success of this process.
151
Botcheva L. and Raghupathy S.
WHY DOES SOCIAL CONTEXT MATTER?Integrating Innovative Technologies with Best Practice Models for Public and Behavioral Health Promotion.
DOI: 10.5220/0004459701510157
In Proceedings of the First International Symposium on Business Modeling and Software Design (BMSD 2011), pages 151-157
ISBN: 978-989-8425-68-3
Copyright
c
2011 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
2 BEST PRACTICE IN
TECHNOLOGY ADOPTION
2.1 Existing Theoretical Models
A rich body of literature has emerged that employs
behavioural science theories to model factors
affecting the acceptance and use of technology at
both the individual level and the organizational
level. Prominent examples include the Technology
acceptance model (Davis, Bagozzi, & Warshaw,
1989), Theory of planned behaviour (Ajzen, 1985),
Unified theory of acceptance and use of technology
(Venkatesh, Morris, Gordon B. Davis, & Davis,
2003), Diffusion of innovation (Rogers, 1995), and
the Technology, organization, and environment
framework (Tornatzky and Fleischer 1990). Some of
these models focus on technological and individual
factors influencing acceptance but as models have
become more sophisticated and better validated,
there has been an increasing acknowledgement of
the centrality of environmental and contextual
constructs. Key constucts in these models that relate
to contextual factors include compatability (with
existing technology, work practices, beliefs and
values), social influence, professional environment,
organizational structure. In addition, many of the
models discuss individual factors that are inevitably
related to broader environmental and social contexts,
such as attitudes, beliefs and subjective norms. A
comparison of theories of change and contextual
constructs defined in models of technology
acceptance is presented in the Appendix.
Drawing from these theories we present a revised
multi-contextual model, adapted from the extended
technology acceptance model (Dadayan & Ferro,
2005) to incorporate broad social influences such as
culture, community context, and ideologies that are
critical for technology adoption in real life contexts
(see Figure 1).
Individual context refers to the characteristics of
individual end-users and their attitudes to
technology;
Technological context refers to the characteristics
of the technology such as functionality and user-
friendliness; and
Implementation context refers to the user’s
environment, including organizational factors
(climate, support, readiness); broad social influences
(community, culture, ideologies); and technology
compatibility.
In the next sections drawing on the extensive
experience of our company in developing innovative
products for public and behavioural health
promotion, we will outline the methodology that we
use to examine these different characteristics for
developing technology products that are adequately
integrated within real-life contexts.
Figure 1: Revised Multi-contextual Model of Technology Acceptance.
BMSD 2011 - First International Symposium on Business Modeling and Software Design
152
2.2 Our Approach: The Intersection
Between Research, Technology
and User Needs
Established in 1983 and selected as an exemplary
small business in 2007 by the National Academy of
Sciences, Sociometrics’ primary mission is to
develop and disseminate behavioural and social
science research-based resources for a variety of
audiences in order to: 1) promote healthy
behaviours; and 2) prevent or reduce behaviours that
put an individual’s health
and well-being at risk.
During the last ten years, Sociometrics’ staff has
developed engaging IT public and behavioural
health promotion products and has also accrued
significant experience developing and disseminating
research-based materials tailored for diverse target
audiences through large scale websites, digitalized
effective program materials, data libraries, and
evaluation and training e-tools (see full description
at www.socio.com).
All products are developed based on a thorough
examination of user needs and preferences with a
special attention to the contexts in which they will
be implemented to assure their acceptance and
relevance.
3 CASE STUDIES
3.1 e-Learning Products for Early
Intervention Professionals
In this case example we present the process of deve-
loping interactive program tutorials tailored to the
different learning styles of early intervention
professionals working in diverse settings (child care
centres, hospitals, and community-based centres).
The project was part of a Sociometrics initiative
funded by NIH aimed at assembling in one place—
for public dissemination, distribution, and
replication—treatment programs in the area of early
childhood intervention. One of the important goals
of the project was to design technology assisted
professional tutorial materials to assist early
childhood professionals to implement programs with
fidelity in their professional settings. Following the
principles of Dabbagh and Bannan-Ritland (2005)
for on-line learning design, we first identified users’
characteristics and learning styles and then explored
their learning and professional context. Using
interviews with potential users, observations of their
professional context and reviews of relevant
literature we were able to outline contextual
characteristics that determined the design and
technological modality of the professional tutorials
(see Table 1).
3.1.1 Individual Context
The analysis of the individual context showed that
technology based products are not used routinely by
potential users, which determined relatively high
technology anxiety; attitudes to technology vary
among professionals with medical and
administrative personnel being more positive. In
response to this context we decided to create
learning tools that are easy to use by people with no
previous experience with technology.
Table 1: Context analysis for developing early childhood intervention professional tools.
Context
Early Childhood Intervention Contextual
Characteristics
Solutions
Individual Context
Technology anxiety
Technology attitudes
Technology anxiety relatively high
Attitudes vary among professionals
but in most cases technology based
products are not used routinely
Create learning tools that are easy
to use by people with no previous
experience with technology
Technological Context
Performance expectancy
Effort Expectancy
Performance expectancy is high;
users will not use these tools if they do not
believe that they will improve their direct
work
Based on the high work load and
relative low level of technology skills, the
effort expectancy is for easy use
Constructivist approach– learning
by doing, e.g., cognitive apprenticeships,
situated learning, problem-based learning,
efficient learning (minimizes tangential
activity)
Practical (product of instruction is
useful for everyday activity)
Implementation context
Compatibility with existing
technical systems
Organizational support
Professional culture
Technical system limited to basic
software
Technology support often is scarce
Resistance to change; conservative
organizational climates
System that is compatible with
basic software
Materials that will not require a lot
of additional support
Adaptable to diverse learners’
styles and contexts
WHY DOES SOCIAL CONTEXT MATTER? - Integrating Innovative Technologies with Best Practice Models for Public
and Behavioral Health Promotion
153
3.1.2 Technological Context
The technological context analysis showed that
performance expectancy is high; users will not use
these tools if they do not believe that they will
improve their direct work; based on the high work
load and relative low level of technology skills the
effort expectancy is for easy use. To match these
expectations we selected a constructivist approach
for the pedagogical design of the products that is
efficient and practical.
3.1.3 Implementation Context
We analysed several aspects of the implementation
context: 1) compatibility with existing technological
systems: most of the technical systems used by our
users were limited to basic software, which meant
that the professional tools should utilize basic
software that is widely available; 2) Professional
culture; most of the professionals worked in the non-
profit sector, either at health or educational
environments that have been shown to be relatively
conservative and resistant to change (Botcheva et
al.2003): thus the use of our tools should not require
a lot of changes in the routine practices; they should
be tailored to different professional environments
and specific learner needs allowing customisation;
3) Organisational support of implementation:
technology support is scarce in most of the
organisations; thus the tools should be easy to
maintain with minimum technical support.
This multi-level analysis led us to the decision to
create e-learning materials using Adobe interactive
PDFs that will include hands on examples and
resources tailored to the different learning styles and
experience of the individual learner. Interactive
PDFs fit seamlessly into the complex pattern of
diverse learners’ needs, constraint and resources. On
the one hand, they are: completely stable; typically
get past firewalls; require no special software/system
other than Adobe Reader; printable, and easy to
maintain. On the other hand, they are: highly
dynamic; allow audio, video, automations;
indexing/bookmarking and easy linking. These
characteristics helped us to create interactive and
engaging e-learning tools that fit the context of and
user characteristics of early childhood professionals.
3.2 Using the Individual, Technological
and Implementation Context to
Design e-Tools for Data Collection
Public schools in the United States are required to
annually collect and report data on drug use and
other high-risk behaviours from elementary, middle
and high school children. All schools receiving
federal and state funding are expected to collect
baseline data for establishing incidence or
prevalence of data on truancy rates, drug and
violence related suspensions and expulsions, drug
incidence, and prevalence rates, and for
demonstrating simple percentage changes in
outcomes for end of the year performance reports.
It is often difficult, however, for schools to
engage in periodic data collection efforts in the light
of budget constraints and time constraints (Mantell,
Vittis, & Auerbach, 1997; Sedivy, 2000). Teachers
are expected to take on responsibilities other than
teaching even at a time when there are increasing
pressures on them to raise students’ academic
achievement levels. Thus, collection and monitoring
of data on substance use or other health concerns are
perceived as consuming valuable time (Hallfors,
Khatapoush, Kadushin, Watson, & Saxe, 2000). In
response to this need, Sociometrics designed a web-
based survey development and analysis tool that
would allow for swift, efficient and most
importantly, cost-effective data collection, analysis
and reporting. The online system would allow
students to login to a pre-programmed survey with
measures on drug use patterns, truancy and other
high risk behaviours; answer the survey questions;
then logout once he or she is finished. The survey
data would automatically be deposited in a secure
web server and can be accessed by the teacher for
analysis. Such an e-tool is cost effective as it
automates the survey creation and administration
process, and relieves the teacher of burdensome
tasks such as printing surveys, distributing them and
then entering and processing the data.
In designing such a tool, we first started with a
needs assessment that took account of the larger
operational context: specifically, we investigated the
wide range of constraints, limitations and facilitating
factors at the individual (teacher), technological
(school infrastructure), and implementation context
(school). We first identified the primary consumers
of the product. These included not just school
teachers and principals who were responsible for the
data collection and reporting, but also district and
state level supervisors at the State Department for
Education who were responsible for school funding
allocations and monitored school progress. Next, we
conducted numerous focus groups and interviews
with the target audience. The qualitative studies
yielded useful insights into current data collection
efforts in schools, and offered valuable design,
BMSD 2011 - First International Symposium on Business Modeling and Software Design
154
content and dissemination guidelines for our e-tool.
Some of these insights are outlined below.
3.2.1 Individual Context
Teachers began their data collection efforts by
selecting measures, creating paper and pencil
surveys, and administering the surveys in the
classroom during recess, after school hours or
whenever possible, during a health education class.
The data were coded and entered by hand and then
reported to the district. The schools had to provide
districts with end of the year “performance reports”
which reported simple changes in drug and violence
incidents over the school year as part of their
assessment. Numerous problems with the process
were reported, such as insufficient funding and time,
as well as lack of technical assistance. Those in the
poorly funded districts mentioned high student
movement and attrition, and problems in tracking
students. Teachers complained about the time and
effort involved in tracking such data and preparing
reports. They also complained about not receiving
any form of technical assistance from their districts
in terms of selecting appropriate measures. Many
teachers lacked the capacity to conduct basic
statistical analysis such as means, and percentage
changes in key behavioural indicators. It therefore
became clear to us that any online data collection
tool would need to pre-program measures and
indicators that were popular, reliable and validated.
We conducted a poll of the most popularly used
measures and indicators for capturing school
performance and developed pre-programmed
surveys incorporating such indicators. A basic
statistical tool was developed that would allow
teachers to derive summary statistics such as means
and frequencies (e.g. no: of student arrests on drug
related charges, percentage of boys and girls referred
to treatment services etc) without having download
data or use any external software such as SPSS.
(Teachers were also given the option to download
the data if they desired). Finally, it became clear that
because of time and capacity limitations, an
interface needed to be created that would allow
students to take the survey from multiple locations,
and over different points in time. This realization led
not only in interface design changes, but also the
format and structure in how data were to be stored.
3.2.2 Technological Context
One major concern that emerged was whether the
introduction of a new, online data collection system
would introduce a steep “learning curve” for
teachers. Another concern was related to the
investments that the school would be required to
undertake in order to adopt it. To emphasize the
utility and user-friendliness of the e-tool, we decided
to present our product concept in images and terms
that were already familiar to the groups. For
example, we used an existing online survey
system—SurveyMonkey—to illustrate our e-tool
and the precise manner in which it would different
and specially tailored for their school needs. Some
of the school teachers in the focus group were
already using online teacher evaluations: we
encouraged them to speak to the group about their
experiences (both positive and negative), and
highlighted how our product would attempt to
overcome the limitations and replicate the successes
of their experience. By the end of our needs
assessment, teachers and administrators were by and
large, receptive to the idea of online data collection
indicating that our strategies for reducing
“technology anxiety” and establishing “performance
expectancy” were successful.
3.2.3 Implementation Context
During the design and product development stage of
any school-based e-tool, it is absolutely essential to
ensure its compatibility with the school’s
technological infrastructure. A technology
“screener” was mailed out to the focus group
participants and interviewee’s participants in order
to assess the basic “minimum” technological
capacity that the e-tool would have to be compatible
with. Questions included: number of computers in
the school, Internet access, and bandwidth etc.
While almost all schools in our focus group had
Internet access, it became clear that lack of access to
sufficient computers (along with time constraints)
was yet another feature that necessitated group log-
ins from multiple locations (such as libraries,
computer labs and even homes). Privacy and
confidentiality of student data subsequently emerged
as a concern. As a first step, we designed a single
question interface with autoprogression; the screen
automatically gets refreshed once a question was
answered thereby minimizing the amount of time a
response was present on screen. The interface also
included separate login IDs for students and
administrators. Students could use their IDs to login
from any location that was convenient to them while
the administrator had sole control over the data
collected. Administrators using the online statistical
tool (described earlier) would be able to do so
without accessing individual student data. If the
WHY DOES SOCIAL CONTEXT MATTER? - Integrating Innovative Technologies with Best Practice Models for Public
and Behavioral Health Promotion
155
administrator did choose to download the data, the
downloaded data was made available without subject
identifiers in order to maintain confidentiality.
Besides confidentiality, another concern was
related to product pricing and affordability.
Participants identified “frontline” funding and
decision making agencies and offices at the state,
county and district level that could spearhead the use
of online data collection mechanisms in their
districts. We learned that pursuing business
opportunities with state and district agencies (rather
than individual schools) would allow costs to be
incurred by these agencies and would facilitate large
scale adoption of the technology at the ground level.
4 IMPLICATIONS FOR
DEVELOPERS
There are several theoretical and practical
implications for developers that stem from this
analysis.
First, the review of existing theoretical models of
technology acceptance highlight the importance of
developing multi-dimensional approaches that take
in account different social contexts to fully
understand the processes of technology integration
in real life contexts. Interdisciplinary teams
incorporating the knowledge and skills of
technology developers, social and behavioural
scientists will be best suited to solve this problem.
Second, the analysis of technology acceptance in
the non-profit sector highlights the critical
importance of broad social context, such as culture,
ideology, and community climate.
While in
industry, the transition from research and
development to the field primarily focuses on the
end user, in the non-profit sector, there is a range of
intermediary factors (agencies, policies) that
influence if and how the product reaches the end
user. Thus, conventional theories regarding
technology diffusion and adoption need to be
modified with regard to the non-profit and public
sectors.
Third future research and development effort
should focus on development of practical tools and
screeners that will facilitate the translation of
contextual characteristics into technical
requirements for development of products that can
be easily adopted and integrated in real life contexts.
ACKNOWLEDGEMENTS
We want to thank Katie Boswell and Eileen Moyles
for valuable input to this paper. The research
developments reported in this paper are funded by
National Institute on Deafness and Other
Communication Disorders and National Institute on
Drug Abuse.
REFERENCES
Ajzen, I. (1985). From Intentions to Action: A Theory of
Planned Behavior. Action-Control: From Cognition to
Behavior (pp. 11-39). Springer.
Dabbagh, N., & Bannan-Ritland, B. (2005). Online
learning: concepts, strategies, and application.
Pearson/Merrill/Prentice Hall.
Dadayan, L., & Ferro, E. (2005). When Technology Meets
the Mind: A Comparative Study of the Technology
Acceptance Model. In M. A. Wimmer, R.
Traunmüller, Å. Grönlund, & K. V. Andersen (Eds.),
Electronic Government (Vol. 3591, pp. 137-144).
Berlin, Heidelberg: Springer Berlin Heidelberg.
Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989).
User Acceptance of Computer Technology: A
Comparison of Two Theoretical Models. Management
Science, 35(8), 982-1003.
Hallfors, D., Khatapoush, S., Kadushin, C., Watson, K., &
Saxe, L. (2000). A Comparison of Paper vs.
Computer-Assisted Self Interview for School,
Alcohol, Tobacco, and Other Drug Surveys.
Evaluation and Program Planning, 23(2), 149-55.
Kraemer, K. L., Dedrick, J., & Sharma, P. (2009). One
laptop per child: vision vs. reality. Communications of
the ACM, 52, 66–73. doi:10.1145/1516046.1516063
Mantell, J. E., Vittis, A. T. D., & Auerbach, M. I. (1997).
Evaluating HIV prevention interventions. New York:
Plenum Press.
Rogers, E. M. (1995). Diffusion of innovations. Simon and
Schuster.
Sedivy, V. (2000). Is Your Program Ready to Evaluate Its
Effectiveness? A Guide to Program Assessment. Los
Altos, CA: Sociometrics Corporation.
Venkatesh, V., Morris, M. G., Gordon B. Davis, & Davis,
F. D. (2003). User Acceptance of Information
Technology: Toward a Unified View. MIS Quarterly,
27(3), 425-478.
BMSD 2011 - First International Symposium on Business Modeling and Software Design
156
APPENDIX
Table 2: Comparison of theories of change and key contextual constructs in models of technology acceptance.
Model and authors Theory of change Key constructs related to context
Technology, organization,
and environment (TOE)
framework (Tornatzky and
Fleischer 1990)
At the organizational level, three aspects influence
the process by which an enterprise adopts and
implements a technological innovation:
technological context, organizational context, and
environmental context.
Environmental context is the arena in which a
firm conducts its business—its industry,
market structure, competitive pressures,
technology support infrastructure and
government regulation.
Theory of planned behavior
(TPB) (Ajzen 1985, Ajzen
1991, Bajaj and Nidumolu
1998)
At the individual level, behavior is influenced
solely by behavioral intention and behavioral
intention in turn is influenced by attitudes toward
behavior, by subjective norms and by perceived
behavioral control.
Behavioral intention is influenced by
attitudes, subjective norms and perceived
control but these are not explicitly linked to
broader environmental context.
Diffusion of innovation
(DOI) (Rogers 1995)
At both individual and organizational level,
innovations are communicated through certain
channels over time and within a particular social
system. Diffusion through an organization is
related to individual (leader) characteristics,
internal characteristics of organizational structure,
and external characteristics of the organization.
The external characteristics element of the
model refers to system openness.
Technology acceptance
model (TAM) (Davis 1986,
Davis 1989, Davis et al.
1989)
At the individual level, “perceived usefulness”
(outcome expectation) and “perceived ease of use”
(self-efficacy) influence decisions about how and
when individuals will use a new technology, with
intention to use serving as a mediator of actual use.
External variables may be antecedents or
moderators of perceived usefulness and
perceived ease of use. However, there is an
assumption that when someone forms an
intention to act, that they will be free to act
without limitation.
Multi-contextual technology
acceptance framework (Hu
et al. 1999, Chau and Hu
2002)
At the individual level, technology acceptance
behavior is influenced by factors pertaining to the
individual context, the technological context, and
the implementation context.
“Implementation context” refers to the user’s
professional environment.
Unified theory of acceptance
and use of technology
(UTAUT) (Venkatesh et al.
2003)
At the individual level, technology use is directly
determined by performance expectancy, effort
expectancy, social influence, and facilitating
conditions. The impact of these factors is
moderated by gender, age, experience, and
voluntariness of use.
Social influence refers to the degree to which
an individual perceives that others believe he
or she should use a particular technology.
Facilitating conditions refer to the degree to
which an individual believes that an
organizational and technical infrastructure
exists to support the use of a particular
technology.
Extended technology
acceptance model (Dadayan
and Ferro, 2005)
At the individual level, technology acceptance is
influenced by not only technological factors but
also by the individual context and the
implementation context.
The implementation context includes three
determinants — compatibility, social
influence, and organizational facilitation.
Not-for-profit internet
technology adoption model
(O’Hanlon and Chang 2007)
At the organizational level, technology adoption is
influenced by technical capacity, compatibility
(with the organization’s work practices, beliefs and
values), support (of staff and donors), and
organizational characteristics.
Organisational practices, beliefs and values
are critical for the adoption process.
WHY DOES SOCIAL CONTEXT MATTER? - Integrating Innovative Technologies with Best Practice Models for Public
and Behavioral Health Promotion
157