A Bayesian multivariate probit analysis of Korean firms' information system adoption

Publication Date27 Sep 2011
AuthorJae Young Choi,Yeonbae Kim,Yungman Jun,Yunhee Kim
SubjectEconomics,Information & knowledge management,Management science & operations
A Bayesian multivariate probit
analysis of Korean firms’
information system adoption
Jae Young Choi
Center for Growth Engine Industries,
Korea Institute for Industrial Economics and Trade (KIET), Seoul, Korea
Yeonbae Kim
Technology Management, Economics and Policy Program,
Seoul National University, Seoul, Korea
Yungman Jun
Korea Communications Commission, Seoul, Korea, and
Yunhee Kim
Technology Management, Economics and Policy Program,
Seoul National University, Seoul, Korea
Purpose – The purpose of this paper is to reveal the core determinants and adoption patterns of the
major enterprise information systems.
Design/methodology/approach – This study incorporated the core representative and meaningful
explanatory variables in the major previous literatures and analyzes the core determinants of
businesses’ adoption of the essential information systems and the substitutionary patterns among
them, using a Bayesian multivariate probit model, which is based on McFadden’s random utility
model and capable of handling multiple response data.
Findings – It was found that not only factors from the classical technological diffusion viewpoint but
also factors such as organizational tools and strategic behaviors play an important role in firms’
adoption of information systems. Specifically, epidemic effect generally outweighs size effect, and
putting more effort into the intensity of information strategy planning is more influential than the
hiring of a professional chief information officer. On the other hand, such variables as age of the firm,
labor intensity, and number of PCs per person generally have no significant impacts. Finally,
a relatively strong complementary relationship exists between enterprise resource planning and
customer relationship management adoption, and between e-buy and groupware adoption.
Originality/value – The results presented in this paper have important implications for firms on a
minimal budget that want to maximize their productivity through the adoption of information
systems. They also provide important information for government policymakers whose job it is to
design strategies for the successful deployment of information systems.
Keywords Korea, Customerrelationship management, Electronicresources, Information systems,
Information systemadoption, Bayesian multivariateprobit, ERP, CRM, e-Buy, Groupware
Paper type Research paper
1. Introduction
The growth of the information and communications technology (ICT) industry, and the
diffusion of such technology, constitutes an important worldwide trend that continues
The current issue and full text archive of this journal is available at
Korean firms’
system adoption
Received 25 March 2011
Revised 3 June 2011
Accepted 13 June 2011
Industrial Management & Data
Vol. 111 No. 9, 2011
pp. 1465-1480
qEmerald Group Publishing Limited
DOI 10.1108/02635571111182791
to transform economic and management systems. In particular, studies have shown
that organizations, in their adoption of ICT systems, may gain efficiency in production
operations and improve the coordination of economic activities among their business
partners by reducing transaction and coordination costs (Amit and Zott, 2001;
Cavusgil, 2002; Globerman et al., 2001).
Even though there are some existing researches on the determinants of firms’
information system adoption (Astebro, 2002; Gretton et al., 2004; Hollenstein and
¨rter, 2004), most such studies do not take into account a firm’s adoption of more
than one information system (multiple adoption behavior). Moreover, many previous
empirical studies of information system adoption have difficulty in analyzing firms’
behavior on the adoption of information systems due to small sample sizes.
The study uses data from a sample of 438 firms from all areas of business, including
manufacturing, collected by the Korean National Information Society Agency (NIA)
in 2005 and employs a Bayesian multivariate probit model to investigate determinants
affecting companies’ adoption of information systems and determine whether
substitutionary patterns exist among such information systems.
The study focuses on adoption behavior with regard to four specific types of
enterprise information system which can be considered to be representative of the
major categories of enterprise information system used by companies: enterprise
resource planning (ERP), customer relationship management (CRM), e-buy (EBUY),
and groupware (GW). Most of the existing empirical research uses the firm’s stock of
computers or automation tools as a dependent variable. From this, the adoptions of
four types of information system are used as dependent variable in our model.
Specifically, the latter three of the above-mentioned systems represent three main
applications of ICT in the company: CRM is a demand management system; EBUY is a
supply chain management system; and GW is a support chain management system.
(For a fuller description of these sorts of applications, see Section 2.) The fourth sys tem,
ERP, plays the role of integrating all the data and organizational processes of the three
other categories into a unified system. On the basis of this framework, one can analyze
the adoption behavior of firms with regard to these four main information technologies.
The study’s methodology – Bayesian multivariate probit analysis enables one to
model a firm’sadoption of multiple information systems and analyze the interdependence
of one choice on another – that is, patterns of substitution (or complementarity)
among the technologies. In fact, recent literatures in management science started
to regard correlations between activities as indexes which explain the existence of
complementarity (Cassiman and Veugelers, 2006; Bocquet et al., 2007).
The data used in this study are quite reliable in terms of source and size. In the past,
empirical studies of determinants of information system adoption have had difficulty
deriving strongly reliable results due to small sample sizes. Our sample consists of
438 Korean firms representative of the whole of the business sector. The data were
collected through a stratified sampling method from a 2005 survey conducted by the
Korean NIA, Korea’s preeminent top ICT policy and technical support government
The results of this study will surely lend insight to the policy makers designing
corporate and policy strategies for the successful deployment of information systems
in Korea. Figure 1 shows a summarized schematic diagram of the research concept,
with the estimation model identified.

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT