ejise volume 13 issue 1

7/31/2019 EJISE Volume 13 Issue 1

1/96

ISSN 1566-6379 1 Academic Conferences LtdReference this paper as:Al-Jaghoub, S, Al-Yaseen, H and Al-Hourani, M. (2010) Evaluation of Awareness and Acceptability of Using e-Government Services in Developing Countries: the Case of Jordan The Electronic Journal Information SystemsEvaluation Volume 13 Issue 1 2010, (pp1 - 8), available online at www.ejise.com

Evaluation of Awareness and Acceptability of Using e-Government Services in Developing Countries: the Case ofJordan

Saheer Al-Jaghoub, Hussein Al-Yaseen and Mouath Al-HouraniAl-Ahliyya Amman University, Amman, [email protected]@[email protected]

Abstract: Similar to other developing countries, Jordan started a national e-Government initiative aiming tostreamline government procedures and make information and government services available to business andcitizens online. This paper presents the results of a pilot study that aims to assess factors which could influencethe awareness and use of e-Government services in Jordan. It investigates issues such as: accessibility of e-government, citizen's attitude toward various privacy and security, the required services and costs. The data wascollected using quantitative and qualitative methods including a survey and interviews with e-Governmentofficials. The results of this preliminary study suggest that awareness of e-government did not reach the required

level. These findings are hoped to be useful for researchers, practitioners and policy makers.

Keywords: ICT, e-government, developing countries, Jordan, evaluation, mixed research

1. Introduction

Many governments in developed and developing countries are now developing, implementing andimproving their strategies to transform government services using information and communicationtechnologies (ICTs) (Borras, 2004). This transformation of services is referred to as eGovernment, e-Gov., digital government, online government, or transformational government (Gupta et al., 2008).

E-Government, which is the term used in this paper, can be described as the use of any type ofinformation and communication technology to improve services and operations provided to differentparties such as: citizens, businesses, and other government agencies (Grant and Chau, 2005;

Gronlund and Horan, 2005; Adeshara et al., 2004; Arif, 2008).

E-Government has been classified in terms of activities and delivering models into four categories: theGovernment to Citizens or Government to Customer (G2C); Government-to-Business (G2B);Government-to-Employees (G2E); Government-to-Government (G2G); and Citizen-to-Citizen (C2C)(Lee et al., 2005; Carter and Belanger, 2003). A number of studies have focused on many issuesrelated to e-Government such as: e-Government strategies (Beynon-Davies, 2004; Williams andBeynon-Davies, 2004); e-Government program challenges (Barc and Cordella, 2004); e-Governmenttechnical issues (Cottam et al., 2004); e-Government usability websites (Mosse and Whitley, 2004); e-Government adoption (Ciborra and Navarra, 2005; Elsheikh et al., 2008; Dwivedi and Williams., 2008;Mofleh et al., 2008b).

The government of Jordan has started implementing an e-Government program in 2002 which seeks

to improve service delivery and increase the involvement of citizens through the use of ICT. Thispaper presents the results of a research done as a pilot studyin one of the Jordanian universities as apreliminary evaluation of Jordanian citizens awareness and acceptability of the e-Governmentprogram. The data was collected using both a survey and quantitative methods (focus groups andinterviews). The "Levels of e-Government application model" (NAO, 2002) is used as a theoreticalframework to aid in the analysis.

The paper is structured as follows: the introduction is followed by presenting the research approach,then the theoretical framework is discussed, after which the e-Government initiative in Jordan isbriefly presented. Section 5 presents the results, followed by the discussion and conclusion.
mailto:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]


2/96

Electronic Journal Information Systems Evaluation Volume 13 Issue 1 2010 (1 8)

www.ejise.com 2 Academic Conferences Ltd

2. Research approach

Mixed research approach (quantitative and qualitative methods) was used to collect the data for thispaper, including a survey, focus group discussions and interviews. Using a survey research method isconsidered to be an appropriate research approach to address the citizens' awareness and usage ofe-Government services (Choudrie and Dwivedi, 2005). However, combining quantitative and

qualitative (Mixed Research method) sources of information and analytical methods can build on thestrength of each type of data collection and minimize the weaknesses of any single approach(Tashakkori and Teddlie, 2003). A multi-method approach can increase both the reliability and validityof evaluated data. A mixed research approach is argued to be the most appropriate technique wheninvestigating a complex and emerging phenomena such as e-Government (Creswell, 2003; Greenand Preston, 2005; Sammons et al., 2005; Mofleh et al., 2008a; Shareef et al., 2009).

In order to achieve the objectives of this paper in exploring the issues related to the awareness andacceptability of e-Government in Jordan, the following questions needed to be answered byparticipants and were included in the questionnaire:

What percentage of the Jordanian population uses computers and the Internet? How do peopleuse these tools (applications and Internet)?

Where do people feel comfortable using computers and the Internet? For what purposes do theyuse the Internet? Why do they NOT use the Internet?

What are peoples' attitudes toward the Internet and the costs of its use?

Would people use government services if they were available on the Internet? How much wouldthey pay for these services?

What are the privacy and security concerns with respect to e-Government applications?

What are the main barriers of using government services over the Internet?

As mentioned earlier, this is a pilot study for a more comprehensive research that aims to evaluatethe Jordanian citizens' awareness and acceptability of e-Government services. The data for thesurvey and the focus groups was collected at Al-Ahliyya Amman University (AAU). Four sources ofdata and information were identified:

General information on e-Government services (Government online, 2009)

Data from a survey: we administered a survey instrument to a sample of undergraduate studentsatAl-Ahliyya Amman University. Many studies in ICT adoption have used students as a sample(Moon and Kim, 2001) as they have experience using the Internet; differ from the demographicsof the population of citizens; and the majority of undergraduate students use and have easyaccess to Internet services. The questionnaire contains four stages: designing and testing thequestionnaire based on two focus groups and some interviews; followed by data collection andthen data analysis; and finally the preliminary findings of the questionnaire.

Focus group discussions: focus groups were organized both before and after the survey. The firstfocus group meetings were designed to take place prior to the survey, and their main goal was todiscuss with participants their attitudes, usage and awareness of the internet and e-Governmentservices in Jordan to find out the reasons for no or low levels of usage. The questionnaire was

also distributed to the first focus group participants. The second round of focus group meetingswas conceptualized as a forum for discussing the survey results and, on this basis, developingrecommendations.

In addition to the above, interviews were conducted with the Director of e-Government Projectsand the e-Government initiative manager at the Ministry of Information and CommunicationTechnology (MoICT) in Jordan.

The results and the conclusions from the interviews were integrated and synthesized with the result ofthe survey.

3. Theoretical framework of e-government project

Similar to any project, an e-Government project is a unique, complex, and one time effort, withspecified limitations (time, budget, resources and performance) designed to meet governmental andagencies goals or stakeholders (Citizens, Businesses, Employees, and Agencies) needs (Al-Yaseenet al., 2008).


3/96

Saheer Al-Jaghoub et al.

www.ejise.com 3 ISSN 1566-6379

The model used in this paper is based on the "Levels of e-Government application model" (NAO,2002). This model proposes that each e-Government project goes through different phases from theeasy phase (Basic site) which mainly involves developing websites, piloting a few applications, andputting these services online to the most advanced phase in which e-government reaches a high levelof maturity and becomes fully integrated into the governance framework and activities of each sector

(Joined-up e-Government) (Yang and Paul, 2005; Santos and Heeks, 2003). Figure 1 shows the mainlevels of the e-Government project, which are discussed briefly below (NAO, 2002).

Figure 1: Levels of e-government application model (NAO, 2002)

Stage 1: Basic site provision of basic information, usually electronic versions of printdocuments; no email contact is given.

Stage 2: Electronic Publishing provision of extensive downloadable, static information via ICTs;basic email contact is given but there is no online interaction.

Stage 3: Interactive Publishing provision of dynamic, customizable information via ICTs; with acapacity to interact (e.g. via email and online submissions).

Stage 4: Transaction ability to authenticate users and register their identities reliably in order to

undertake complete transactions online (e.g. submitting a tax return); users may be able tomanage their own 'accounts' (i.e. a set of personal interactions/transactions).

Stage 5: Joined-up e-Government interlinking of different government structures/processes toproduce electronic one-stop shops; users can access their own files/accounts and manage theirrelationships with government wholly via the Internet.

4. e-Government initiative in Jordan

Jordan is a developing country in the Middle East with a population of about 5,000,000 and limitednatural resources. The e-Government initiative in Jordan has been one of a number of ICT relatedinitiatives launched with the succession of King Abdullah II to the throne in 1999, the aim of which isto transform the country into a knowledge-based economy (Al-Jaghoub and Westrup, 2003). As theprojects manager in MoICT stated: "In 2000, the Government of Jordan launched a national e-

Government initiative; in 2003 the initiative has been chartered by His Majesty King Abdullah II as ane-Government program, aiming to make information and services available to citizens on theInternet". Jordan's national e-Government program aims to provide government and agencies

Basic Site

ElectronicPublishing

Interaction

Publishing

Transaction

AccountManagement

Joined-Up e-Government


4/96



services throughout various electronic channels such as internet, SMS gate, mail and others wherework is in progress to develop a number of electronic services. The e-Government program aims todeliver high-quality services to consumers, businesses and organizations; improve governmentperformance and efficiency; enhance Jordans competitiveness; ensure public sector transparencyand accountability; reduce costs and increase ease of interaction with government; promotedevelopment of Jordans ICT sector; develop skills within the public sector; boost e-commerce

activities; and improve information security.

The e-Government initiative has been launched in terms of activities and delivering models into fourcategories: the Government to Citizens or Government to Customer (G2C) section aims to providecomprehensive information about all the services provided by the Jordanian Ministries and otherJordanian government agencies. Citizens can benefit from services related to life including birth,school, work, retirement, health services and so on. Government to Business (G2B) aims to provideinvestors in Jordan with information about all the services provided by the Ministries and othergovernment agencies, including how to start a new business, running a business, investmentincentives, privatization program, and investment benefits at free zones and so on. Government toEmployees (G2E) aims to provide governmental employees in different government agencies inJordan with information about all the services provided by the Ministries and government agencies(Director of e-Government program in Jordan).

5. Preliminary results of awareness and usage of e-government services

The survey was distributed to a sample of 1200 students from different departments at AAU. Of the1200 questionnaires, 1200 completed responses were collected; giving a response rate of 100%. Theresponse level is due to the fact that the questionnaires were answered and collected with thepresence of the researcher(s) during lectures. This rate is considered to be above expectation giventhat the generally accepted average responses to non-incentive based questionnaires are around20%. We analyzed the data from the responses of the questionnaire using a combination of theparametric statistical methods, Descriptive Analysis and Factor Analysis (Pett, et al., 2003). Studentswere asked to select from the list the closest choice of many variables. Each of these variables weremeasured using a five point Likert scales (1 = not important and 5 = very important).

For technically interested readers we report that a factor analysis technique was employed in order to

identify possible categories. Factor analysis was performed in three steps (following Berthold andHand, 2003):

A matrix of correlation coefficients for all possible pairings of the variables was generated.

Factors were then extracted from the correlation matrix using principal factors analysis.

The factors were rotated to maximize the relationships between the variables and some of thefactors and minimize association with others using Varimax Kaiser Normalization, whichmaintained independence among the mathematical factors. The Eigenvalues determined whichfactors remained in the analysis. Following Kaisers criterion, factors with an Eigenvalue of lessthan 1 were excluded. A Screen plot provides a graphic image of the Eigenvalue for eachcomponent extracted

5.1 Computer and internet useA large majority -100%- of the sample currently uses computers and most of the computer users alsouse the Internet. This rate is reasonable given the characteristics of the sample which consists ofuniversity students who have access to computers and the internet, and tend to rely on it foreducational and other purposes. This level of access may not be applicable to the the majority ofJordanians.

5.2 Importance of place and purpose of using internet

Most of the people in this sample report using computers at home, at work or at the university. Of thepeople who use computers: 100% use it at home; 60.2% use it at University; 14.7% use it at work asmost of the students are unemployed-; while 31.8% use it at a public place. As noted, most computerusers are also Internet users. Home is main place 84.1% for connecting to the Internet, 13.5% of

Internet users at work; 40.8% of Internet users at University; and 30.2% connect to the Internet at apublic place.


5/96

Saheer Al-Jaghoub et al.


5.3 Attitudes toward the Internet, cost and other factors

The reasons for not using the Internet are varied. The main reason is associated with 'Not enoughtime'; and the least important reason is 'Don't use computers', as shown in Table 1 below:

Table 1: Reasons for not using the Internet

Reasons for not using the Internet %

Not enough time 17.9Phone bill too high 15.1

Internet charge too high 12.3For security reasons 7.4

Concerned about kids 2.6Don't use computers 0.0

The most frequent Internet use was browsing the Internet (100%), entertainment (86.2%); sendingand receiving emails (56.2%); getting information (36.4%); shopping over the Internet (16.7%) and forpaying bills online (1.9%).

5.4 Using government services over the Internet

When we asked the participants if they know what e-Government is in Jordan, (74.1%) answered

'Yes'; while when we asked the same question in a different way within the survey, we found thatmore than (75%) of the participants do not actually know about e-Government services or its Website.Moreover, the study found that more than (85%) of the participants never logged in to e-GovernmentWebsite or never got any information.

In order to investigate how people might feel about using government services on the Internet, weasked a series of questions about peoples most current use of different services by going togovernment departments, and then we asked if they would use these services if they becomeavailable on the Internet. The respondents were also asked how much they are willing to pay for theconvenience of using such services online.

The most frequently used service included checking traffic tickets, while the least service wasrenewing the family document. As Table 2 indicates, many of the actual users of the services are very

interested in having an Internet-based delivery system:

Table 2: e-Government services

e-Government used services %Information about checking traffic tickets 81.3

Information about the weather 51.1Renew passport 39.4Renew ID card 28.4

renew a drivers license 18.9Paying bills 17.2

Information about car tax 16.8Apply for job 10.1

Renew health card 8.1

Pay taxes 7.9Tax refund 7.3Income tax settlement 7.1

Tax situation 6.5Renew family document 1.6

The results are presented in Table 3. Using a factor analysis cut-off level of 0.5, four factors wereconsidered the main reasons of using e-Government services, which we described as: 'personalinformation', 'security and health information', 'tax information', and 'other information'.

As mentioned earlier in Table 1, -for the privacy and security issues- nearly 7.4% of the study sampleagreed that they were worried about privacy on the Internet.


6/96



Table 3: Factor analysis of e-government services

e-Government services

Factors

Personalinformation

Securityand healthinformation

Taxinformation

Otherinformation

Information about checking traffic

tickets

0.973

Information about the weather 0.869Renew ID card 0.865

renew a drivers license 0.776Paying bills 0.973

Information about car tax 0.973Apply for job 0.784

Renew health card 0.974Pay taxes 0.928Tax refund 0.874

Income tax settlement 0.842Tax situation 0.841

Renew family document 0.933Note: Only loadings greater than 0.50 are shown

6. Discussion and conclusion

A striking result of the above survey is the low level of awareness of the e-Government programwithin the sample. The fact that our sample consisted of students who are young, most of them havehigh income, have access to the Internet and use it for various reasons, implies that they should atleast know what e-Government is. Given that many other Jordanians will not have the same level ofincome, Internet access, and easiness of usage means that they will most likely have even lowerawareness and usage levels of e-Government services. According to the e-Government officials, thisis expected and there are awareness plans in place to address this issue which includes conductingworkshops, making visits to different agencies such as schools, universities, companies and mediapromotions. However, such plans face the challenge of the limited resources available which includebudget, qualified personnel, and the culture (Director of projects).

The demanded services as shown in table 3 based on factor analysis indicates that such services arerelated to the needs of the people. For example, in our sample the least demanded service was thefamily document renewal which is reasonable taken into consideration that students in most cases donot have such a document which is required once a person starts his own family. For otherJordanians, the importance of services may be ranked differently.

Results indicated that people are willing to pay a fee to use some of the Internet-based services. Withrespect to how much people might pay for the convenience of such services (if it is available online),responses varied depending on the nature of the services provided by government. On average,people are willing to pay more to the first factor which we described as 'personal information', thanthey are for the other factors, which we described as 'security and health information', ''taxinformation', and 'other information', respectively. Payment for services may be linked to the person's

income especially that even within the focus group discussion a number of participants expressedconcerns about cost of using services and using the internet in general. Therefore, for otherJordanians with lower income levels this concern about cost is most likely applicable.

A very important issue that was pointed out during focus group discussions is the concern aboutsecurity of giving information over the internet. Generally speaking, there is a lack of trust of usingonline transactions especially when it comes to important documents or payments. This may also berelated to the culture in general within Jordan where the internet still seems to be used mainly forentertainment. For example, only 16% of the sample did shopping online. Using e-Governmentservices within such a culture is still problematic and needs serious attention.

NAO (2002) proposed a model to present different levels of e-Government services development.This model has five components: 1) Basic site; 2) Electronic Publishing; 3) Interactive Publishing; 4)

Transaction; and 5) Joined-up e-Government. This research applied the model of the Jordanian e-Government program in order to examine the current situation of the e-Government project. E-Government project in Jordan launched nine years ago, and based on the different sources of data


7/96


8/96



Green, A. and Preston, J. (2005) Editorial: Speaking in Tongues Diversity in Mixed Methods Research,International Journal of Social Research Methodology, Vol. 8, No. 3, Pp. 167-171.

Gronlund, A. and Horan, T. A. (2005) Developing a generic framework for e-Government, Communications of theAssociation for Information Systems, Vol. 15, Pp. 713-729.

Lee, S. M., Tan, X., Timi, S. (2005) Current practices of leading e-Government countries, Communications of theAssociation for Information Systems, Vol. 48, No. 10, Pp. 99-104.

Mofleh, S., Wanous, M. and Strachan, P. (2008a) Developing countries and ICT initiatives: Lessons learnt from

Jordan's Experience, The Electronic Journal on Information Systems in Developing Countries, Vol. 34, No.5, Pp. 1-17.

Mofleh, S., Wanous, M. and Strachan, P. (2008b) The gap between citizens and e-Government projects: thecase for Jordan, Electronic Government, an International Journal (EG), Vol. 5, No. 3, Pp. 275-287.

Moon, J. M. and Kim, Y. G. (2001) Extending the TAM for a World-Wide-Web Context, Information andManagement, Vol. 28, Pp. 217-230.

Mosse, B. and Whitley, E.A. (2004) Assessing UK e-Government websites: classification and benchmarking. InProceedings of the 12

thEuropean Conference on Information Systems, Turku, Finland.

NAO (2002) Government on the Web II, National Audit Office, London.Pett, M. A., Lackey, N. R. and Sullivan, J. J. (2003) Making sense of factor analysis: the use of factor analysis for

instrument development in health care research, Sage Publications, London.Sammons, P., Siraj-Blatchford, I., Sylva, K., Melhuish, E., Taggart, B. and Elliot, K. (2005) Investigating the

Effects of Pre-School Provision: Using Mixed Methods in EPPE Research, International Journal of SocialResearch Methodology, Vol. 8, No. 3, Pp. 207-224.

Santos, R. and Heeks, R. (2003) ICTs and Intra-Governmental Structure at Local, Regional, and Central Levels:Updating Concentional Ideas, available at:http://www.sed.manchester.ac.uk/idpm/research/publications/wp/igovernment/short/igov_sp07.pdf

Shareef, M., Kumar, U., Kumar, V., Dwivedi, Y. (2009) Identifying critical factors for adoption of e-Government,Electronic Government, an International Journal (EG), Vol. 6, No. 1, Pp. 70-96.

Tashakkori, A., and Teddlie, C (2003) The past and the future of mixed methods research: from MethodologicalTriangulation to Mixed Methods Designs. In Handbook of Mixed Methods in Social and BehavioralResearch, Tashakkori and Teddlie (Eds.). Thousand Oaks, CA: Sage.

Walsham, G., Interpretive case studies in IS research: nature and method, European Journal of InformationSystems, (4),

1995, pp. 74-81.Wiiliams, M.D. and Beynon-Davies, P. (2004) Implementing e-Government in the UK: An analysis of local-level

strategies. In Proceedings of the Americas Conference on Information Systems, New York.Yang, J. and Paul, S. (2005) E-Government application at local level: issues and challenges: an empirical study,

International Journal of an Electronic government, Vol. 2, No. 1, Pp. 56-76.
http://www.sed.manchester.ac.uk/idpm/research/publications/wp/igovernment/short/igov_sp07.pdfhttp://www.sed.manchester.ac.uk/idpm/research/publications/wp/igovernment/short/igov_sp07.pdfhttp://www.sed.manchester.ac.uk/idpm/research/publications/wp/igovernment/short/igov_sp07.pdf


9/96

ISSN 1566-6379 9 Academic Conferences LtdReference this paper as:Al-Yaseen,H, Al-Jaghoub, S, Al-Shorbaji, M and Salim, M. (2010) Post-Implementation Evaluation of HealthCareInformation Systems in Developing Countries The Electronic Journal Information Systems Evaluation Volume 13Issue 1 2010, (pp9 - 16), available online at www.ejise.com

Post-Implementation Evaluation of HealthCare InformationSystems in Developing Countries

Hussein Al-Yaseen, Saheer Al-Jaghoub, Maher Al-Shorbaji and Maher SalimAl-Ahliyya Amman University, Amman, [email protected]@[email protected]@ammanu.edu.jo

Abstract: Information and Communication Technology (ICT) project managers require accurate and reliableevaluation to allocate and control project resources. In addition, many private hospitals indicate that a number oftheir projects have failed; and between one and two thirds of ICT projects exceed their budget and time. Further,about half of the expensive ICT projects at the end will be considered out of control and cancelled. Justifying ICTinvestments is a long standing problem, and managers for the past decades have expressed concerns about thevalue they are getting from their investments, and they have been searching for ways to evaluate and justifythese projects. Hence, evaluation of ICT is therefore becoming an important issue for both managers andpractitioners. This paper aims to investigate the current practice of both types of evaluation: Prior Operational

Use evaluation -POUe- and Operational Use evaluation -OUe- in Jordanian private hospitals to better understandwhat is required for the evaluation process and its associated benefits; secondly, to collect information about howhospitals carry out the evaluation process. In doing so, we attempts to answer specific questions, such as: Howprevalent is POUe and OUe? What criteria are being used in both types of evaluation? What are their mainbenefits and uses of each type of evaluation? Results suggest that most decision makers do not place muchimportance on OUe of their IT/IS. Most managers tend to think of it only as a formality rather than a properevaluation process. Without adopting a formal OUe the cost of future health informatics would seem likely to beless accurately estimated.

Keywords: healthcare information systems, health informatics, evaluation, developing countries, Jordan

1. Introduction

Many organizations in developed and developing countries -in both private and public sectors- turned

to Information Technology/Information Systems (IT/IS) to meet the increasing demands onorganizations to increase their efficiency and effectiveness (Jones and Hughes, 2000). This impliesthat investment in IT/IS is high, which has been a problematic issue for more than one decade. Inaddition, there is a contradictory evidence as to whether or not IT/IS expenditure has resulted increating economic value for the organizations (Willcocks and Lester, 1999; Eldabi et al., 2003; Irani etal., 2002; Al-Yaseen, et al., 2006; 2008). Investments in IT/IS are growing extensively in mostorganizations; managers worry about the fact that benefits from IT/IS investments may not be as highas expected as large amounts of money are invested in IT/IS and there is not enough return from thisexpenditure (Irani et al., 2002; Kumar, 1990; Remenyi et al., 2000; Al-Yaseen et al., 2007).

Organizations specialized in healthcare are no exception as they have also joined in buildinginformation systems that also require investments in IT/IS (Wetter, 2007). These systems are knownas healthcare information systems which focus on optimizing and using information to increase

efficiency and effectiveness in healthcare organizations (Reichertz, 2006; Hayrinen et al., 2007).

This paper investigates the evaluation process of health information systems implemented inJordanian hospitals and explores the main issues related to the evaluation process such as: howprevalent prior operational use and operational use evaluation; main uses and benefits of adoptingboth types of evaluation.

The paper is organized as follows. Section 1 is an introduction about the importance of ICTs inhealthcare information systems. Section 2 discusses healthcare systems evaluation and the need forjustification. The details of the research methodology used are presented in section 3. Section 4presents data analysis and preliminary findings of the adoption of IT/IS evaluation types in healthcaresector. Finally, the discussion and conclusions are presented in section 5.
mailto:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]


10/96



2. IT/IS evaluation and justification

Justifying expenditure on IT/IS is a long standing problem, and managers for the past few decadeshave expressed concerns about the value they are getting from IT/IS investments; moreover theyhave been searching for ways to evaluate and justify the use of IT/IS (Gunasekaran et al., 2006; Al-Yaseen, et al., 2007; 2008). Such a continuous increase in investment coupled with continuous need

for justification presents a challenge to the IT/IS community. Many organizations reported that theyare uncertain about how to measure the impact and the outcomes of their IT/IS investments (Bradfordand Florin, 2003; Eldabi et al., 2003; Farbey et al., 1993; Gunasekaran et al., 2001; Lin and Pervan,2003; Liu et al., 2003; Skok et al., 2001; Al-Yaseen et al., 2007).

3. IT/IS evaluation types

Evaluation can be defined as: 'to judge or determine the significance, worth, or quality' (Webster'sDictionary). Willcocks (1992) defined IT/IS evaluation as the 'process of establishing by quantitativeand/or qualitative techniques the worth of IT/IS projects to the organizations'. Or the process ofassessing the worth of something (Beynon-Davies et al., 2000). Evaluation can also be considered interms of the effectiveness of the IT system in situ what a system actually accomplishes in relation toits stated goals (Al-Yaseen et al., 2004, Eldabi et al., 2003). We take the stance that evaluation is aprocess that takes place at different points in time, or continuously, explicitly searching for(quantitatively or qualitatively) the impact of IT projects (Eldabi et al., 2003). The value of this latterdefinition is that it explicitly recognises the different stages in the full lifecycle of an IT/IS project inwhich evaluation is performed, and provides the opportunity to discriminate between two decidedlydifferent views of the evaluation process, each serving different aims. The evaluation process is afundamental and critical activity and needs to be thoroughly conducted in any phase of the system'slife cycle (Galal et al., 2000). Furthermore, it is increasingly acknowledge that evaluation ofinformation systems is recognized as a complex and challenging activity, and there is no agreementon an ideal way to evaluate or how to make the evaluation process better (Dabrowska and Cornford,2001).

Evaluation types can be classified into two types with regards to the development stage of the systemor the timing of evaluation (Eldabi et al., 2003). Type A is a Prior Operational Use Evaluation;sometimes referred to as ex-ante; formative, or Prior-Implementation Evaluation; or as we shall referto it, Prior Operational Use evaluation (POUe). POUe is a 'predictive evaluation performed toforecast the impact of the project. This type of evaluation is carried out prior the system becomes intooperational use through the development stages of IT/IS- to justify the investment. Type B ofevaluation is carried out when the system becomes into operational use; this form of evaluation drawson real rather than projected data, and can be used to justify adoption (Love and Irani, 2001; Irani,2002); estimate the direct cost of the system, estimate the tangible benefits of the system (Liu et al.,2003); ensure that the system meets requirements (Irani, 2002); measure the system effectivenessand efficiency (Poon and Wagner, 2001); measure the quality of programs and to estimate indirectcosts and other costs (Love and Irani, 2001); or to measure the quality of programmes (Eldabi et al.,2003). We shall refer to this type as Operational Use evaluation (OUe). Figure 1 shows these formsof evaluation with respect to the system's life cycle from a systems inception to the end of its usefullife.

Time

Systems life cycle

Opera

tionalU

se

Evalu

ation

Development Stage System into operational use

Prior Operational

Use Evaluation

Figure 1:IT/IS evaluation types in the systems life cycle


11/96

Hussein Al-Yaseen et al.


There are also a number of empirical studies such as those reviewed by (Ballantine et al., 1996) which examined ex-ante evaluation (POUe), yet only a few {for example Kumar (1990) and to someextent Beynon-Davies et al., 2004)} that have explored the ex-post evaluation (OUe).

4. Research methodology

In order to gather as much data as possible to understand the whole picture about prior operationaluse and operational use evaluation as practices within Jordanian private hospitals and to understandthe benefits and obstacles of these two types of evaluation; the following data sources were used: General information on healthcare systems in Jordan (Government websites such as: Ministry of

HealthMOH-, Private Hospitals AssociationPHA-, 2009)

Data from a survey: we administered a survey instrument to all private hospitals. Questionnairecontains four stages: designing and testing the questionnaire; followed by data collection and thendata analysis; and finally the preliminary findings of the questionnaire. The survey was sent to 60private hospitals in Jordan.

In doing so, the following questions need to be answered by practitioners who are involved with IT/ISevaluation in healthcare sector: How prevalent is prior operational use and operational use evaluation of healthcare systems in

private hospitals in Jordan? What criteria are being used in order to evaluation healthcare systems?

What are the main uses and benefits of adopting the two types of evaluation of healthcaresystems?

We analyzed the data from the responses of the questionnaire using a combination of the parametricstatistical methods, Descriptive Analysis and Factor Analysis (Pett, et al., 2003). Respondents wereasked to select from the list the closest choice of many variables. Each of these variables weremeasured using a five point Likert scales (1 = not important and 5 = very important).

For technically interested readers we report that a factor analysis technique was employed in order toidentify possible categories. Factor analysis was performed in three steps (following Berthold andHand, 2003):

A matrix of correlation coefficients for all possible pairings of the variables was generated. Factors were then extracted from the correlation matrix using principal factors analysis.

The factors were rotated to maximize the relationships between the variables and some of thefactors and minimize association with others using Varimax Kaiser Normalization, whichmaintained independence among the mathematical factors. The Eigenvalues determined whichfactors remained in the analysis. Following Kaisers criterion, factors with an Eigenvalue of lessthan 1 were excluded. A Screen plot provides a graphic image of the Eigenvalue for eachcomponent extracted.

5. Data analysis and preliminary findings

This section presents aggregated results from direct answers to the research questions mentioned

above. The basic issues considered here are: reasons for adopting either type of evaluations, criteriaused for evaluations, and uses and benefits of adopting the two types of evaluation of healthcaresystems.

Of the 60 questionnaires addressed to all private hospitals, 19 completed questionnaires werereturned for a total response rate of 31.6%. The average IT/IS costs for the private hospitals in theresearch sample is ($328,000); while within the sample, 15% had implemented systems that costmore than ($1,140,000). 26.3% of the respondent hospitals have adopted IT/IS as a response toproblem(s), while 73.7% of the respondent hospitals have adopted IT/IS systems searching for waysof improving their efficiency and effectiveness.

6. Reasons for adopting POUe in healthcare systems

The results are presented in Table 1. Using a factor analysis cut-off level of 0.5, four factors wereconsidered the main reasons of adopting Prior Operational Use evaluation, which we described assystem completion and justification, system costs, system benefits, and other reasons.


12/96



Table 1: Reasons for adopting Prior Operational Use evaluation factor analysis

Reasons

FactorsSystem

completionand

justificationSystemcosts

Systembenefits

Otherreason

System meets requirements 0.767System effectiveness 0.782

System usage 0.791System efficiency 0.786Justify adoption 0.750System security 0.742

System performance 0.772Quality and completeness of system

documentation 0.670Hardware performance 0.655

Quality of programs 0.666Operational costs 0.598

Training costs 0.619Maintenance costs 0.584

Upgrade costs 0.542Reduction in clerical salaries 0.649Reduction in other staff costs 0.630

Other expenses saved 0.652Direct costs 0.682

Indirect costs 0.676Other costs 0.686

Tangible benefits 0.525Intangible benefits 0.611

Other benefits 0.578Barriers of adopting the system 0.542

Note: Only loadings greater than 0.50 are shown

The first factor System completion and justification is highly correlated with seven variables, the

second factor system costs is highly correlated with seven variables, and the third factor systembenefits are highly correlated with eight factors, whilst the fourth factor other reasons is highlycorrelated with two variables 'Other benefits' and 'barriers for adopting the system' which was alsofound to be the least evaluated reason in practice, as shown in Table 1.

7. Reasons for adopting OUe in healthcare systems

The results are presented in Table 2. Employing a factor analysis cut-off level of 0.5, three factorswere considered. The most important reasons for adopting Operational Use evaluation were identifiedfrom a five-point Likert scale ranging from 1 (not important) to 5 (very important). The results arepresented in Table 2. Three factors were considered as the main reasons of adopting OperationalUse evaluation, which we described as: system costs, system benefits, and other reasons.

Table 2: Reasons for adopting Operational Use evaluation factor analysis

Variables

FactorsOther

reasonsSystembenefits

Systemcosts

Estimating of system life 0.751Justify system adoption 0.751

Risks 0.741Barriers 0.612

Tangible benefits 0.677Intangible benefits 0.678

Other benefits 0.682Direct costs 0.646

Indirect costs 0.519Other costs 0.522



13/96



The first factor other reasons is highly correlated with four variables, the second factor systemsbenefits' is highly correlated with three variables, and the third factor systems costs' is highlycorrelated with three variables 'direct costs', 'indirect costs', and 'other costs' which was also found tobe the least evaluated reason in practice, as shown in Table 2.

8. OUe criteria used in healthcare systems

The results are presented in Table 3. Employing a factor analysis cut-off level of 0.5, four factors wereconsidered. OUe criteria were identified from a five-point Likert scale ranging from 1 (not important) to5 (very important). The results are presented in Table 3. four factors were considered as the maincriteria of OUe, which we described as: system completion, system information, system impact,and other criteria.

Table 3: Operational Use evaluation criteria factor analysis

Criteria

FactorsSystem

completionSystem

informationSystemimpact

Othercriteria

Internal controls 0.873Project schedule compliance 0.769

System security and disaster protection 0.794

Hardware performance 0.765System performance versus specifications 0.676

System usage 0.873Quality and completeness of system documentation 0.773

Accuracy of information 0.984Timeliness and currency of information 0.974

Adequacy of information 0.979Appropriateness of information 0.874

Quality of programs 0.842User satisfaction and attitude towards systems 0.959

User friendliness of system-user interface 0.874System's impacts on users and their jobs 0.928

System's fit with the impact upon organization 0.849Net operating costs (savings of system) 0.633


The first factor system completion is highly correlated to seven criteria, the second factor systeminformation are highly correlated to five criteria, the third factor system impact is highly correlated tofour criteria, whilst other criteria is correlated to one criterion net operating costs, which was alsofound to be the least evaluated criteria in practice. For more information, see Table 4 which showsthe construct loadings for the reasons of adopting Operational Use evaluation.

9. Discussion and conclusions

All of the responding private hospitals have adopted a formal POUe, but only about a third (31.5%)currently perform a formal OUe IT/IS. This means that about two thirds (68.5%) of the privatehospitals do not gather any evidence to establish how successful their IT/IS were, therefore cannot

use such information from OUe to improve their evaluation techniques and outcomes and decreasedeviation. Results suggest that most decision makers do not place much importance on OUe of theirIT/IS. Most managers tend to think of it only as a formality rather than a proper evaluation process.For example, amongst the 6 hospitals who consider adopting OUe those hospitals that seriouslyperform it tend to gain considerable benefits, including the validation of their original POUe estimates.In some cases, OUe is adopted in order to move responsibility from developers to users and forformal closure. When performed, the most popular reasons for adopting OUe were related to formalaspects of signing off the project (based around traditional measures such as meeting requirements,and achieving agreed metrics for effectiveness, usage, efficiency, security, performance, etc.), andsystem costs. The two factors -systems benefits and adoption barriers were found to be lessimportant. On the other hand, amongst the 19 private hospitals, the most frequent reason for adoptingOUe was related to systems benefits (both tangible and intangible). Most of the sampled privatehospitals attach greater importance to the measurement of benefits rather than the measurement of

costs. The most important claimed use and benefit of adopting OUe was system cost (operationalcost, training cost, maintenance cost, upgrade cost, reduction in other staff cost, reduction in salaries,and other expenses saved).As for the criteria used, the most frequently cited criterion for OUe was


14/96


15/96



Hayrinen, K., Saranto, K., and Nykanen, P. (2007) Definition, structure, content, use and impacts of electronichealth records: A review of the research literature, International Journal of Medical Informatics, Vol. 77,No. 5, Pp. 291-304.

Irani, Z. (2002) Information systems evaluation: navigating through the problem domain, International Journal ofInformation and Management, Vol. 40, Pp. 11 24.

Irani, Z., Sharif, A., Love, P.E.D. and Kahraman, C. (2002) Applying concepts of fuzzy cognitive mapping tomodel: The IT/IS investment evaluation process, International Journal of Production Economics, Issue

75, pp 199 211.Jones, S. and Hughes, J. (2000) Understanding IS Evaluation as a Complex Social Process. Proceedings of the

2000 Americas Conference on Information Systems (AMCIS) (10-13 August, Long Beach, Ca.), Editor:Chung, H. M., Association for Information Systems, Atlanta, pp 1123 1127.

Kumar, K. (1990) Post Implementation Evaluation of Computer Information Systems: Current Practices.Communications of the Association for Computer Machinery (ACM), Vol. 33, No. 2, N.Y., (January), pp203 212.

Lin, C. and Pervan, G. (2003) The practice of IS/IT benefits management in large Australian organisations,International Journal of Information and Management, Issue 41, pp 13 24.

Liu, Y., Yu, F., Su, S.Y.W. and Lam, H. (2003) A Cost-Benefit Evaluation Server for decision support in e-business, Journal of Decision Support Systems, Issue 36, pp 81 97.

Love, P. E. D. and Irani, Z. (2001) Evaluation of IT costs in construction, Journal of Automation in Construction,Vol. 10, Pp. 649 658.

Ministry of Health (MoH) Jordan,www.moh.gov.jo/MOH/En/home.php, (Accessed March, 2009).

Pett, M. A., Lackey, N. R. and Sullivan, J. J. (2003) Making sense of factor analysis: the use of factor analysis forinstrument development in health care research, Sage Publications, London.Poon, P. and Wagner, C. (2001) Critical success factors revisited: success and failure cases of information

systems for senior executives, Journal of Decision Support Systems, Vol. 30, Pp. 393 418.Private Hospitals Association (PHA) Jordan,www.pha-jo.com, (Accessed March, 2009).Reichertz, P. L. (2006) Hospital information systems Past, present, future, International Journal of Medical

Informatics, Vol. 75, No. 3-4, Pp. 282-299.Remenyi, D., Money, A., Sherwood-Smith, M. and Irani, Z. (2000), The effective management and management

of IT costs and benefits, Butterworth-Heinemann Ltd., London.Skok, W., Kophamel, A. and Richardson, I. (2001) Diagnosing information systems success: importance-

performance maps in the health club industry, International Journal of Information and Management,Issue 38, pp 409 419.

Wetter, T. (2007) To decay is system: The challenges of keeping a health information system alive, InternationalJournal of Medical Informatics, Vol. 76, Pp. 252-260.

Willcocks, L. (1992) Evaluating information technology investments, research findings and reappraisal, Journal ofInformation Systems, Vol.2, pp 243 268.

Willcocks, L. and Lester, S. (1999) Beyond the IT Productivity Paradox, Wiley, Chichester.
http://www.moh.gov.jo/MOH/En/home.phphttp://www.moh.gov.jo/MOH/En/home.phphttp://www.moh.gov.jo/MOH/En/home.phphttp://www.pha-jo.com/http://www.pha-jo.com/http://www.pha-jo.com/http://www.pha-jo.com/http://www.moh.gov.jo/MOH/En/home.php


16/96




17/96

ISSN 1566-6379 17 Academic Conferences LtdReference this paper as:Carton, F and Adam, F. (2010) Towards a Model for Determining the Scope of ICT Integration in the Enterprise:the Case of Enterprise Resource Planning (ERP) Systems The Electronic Journal Information SystemsEvaluation Volume 13 Issue 1 2010, (pp17 - 26), available online at www.ejise.com

Towards a Model for Determining the Scope of ICTIntegration in the Enterprise: the Case of EnterpriseResource Planning (ERP) Systems

Fergal Carton and Frederic AdamUniversity College Cork, [email protected]@afis.ucc.ie

Abstract: The question of integration of information systems (IS) into the planning and execution of operationalactivities has been the focus for researchers from different constituencies. Organisational theorists recognise theneed for integrating mechanisms for co-ordinating the actions of sub-units within an organisation. Centralisationhas been seen as a defensive reaction by organisations when placed under increasing external control , and alsoas a way to improve the efficiency of information processing, at least for routine tasks. In the meantime,researchers have been sceptical about the ability for structured information systems to deal with the complexity ofthe information flows within the organisation. Frameworks have also been identifying characteristics of the tasksthemselves that have a bearing on the amount of information processing required. The real world is complex andmoving, thus managers require flexibility in their interpretation of the mixed signals arising from this complexity.However, managers are working in environments where highly integrated information systems blur the distinctionbetween what is real and what is virtual. There is a need for an integration approach allowing organisations toquestion which areas of activity are worth integrating, and conversely which areas are better left under localcontrol. Where integrated, managers require processes for the maintenance of data integrity (people, tools,procedures). Based on field work involving two multi-national manufacturing companies, this paper proposes aframework for ERP integration, which describes the evolution of functionality gaps as an ongoing and inevitableprocess that requires management.

Keywords: ERP, enterprise, integration, framework, complexity

1. Introduction

In examining the extent to which information systems (IS) can be integrated into the planning andexecution of an organisations activities, the researcher may draw from rich seams of research from

different constituencies. The question of how best to control and co-ordinate the organisation to meetthe needs of its customers, while optimising the use of available resources, has tantalisedorganisational theorists from the earliest days of industrial and economic growth. Key to the conceptof control is information, such that goals may be communicated downwards and performance may bemeasured and communicated upwards. It has been acknowledged by the earliest students ofadministrative systems that the major cost of implementing control systems of any sort is theinvestment in human resources to design and maintain the system. Today, advances in data capture,communication and dissemination technologies have allowed designers to integrate the bureaucraticoverhead of administration of task monitoring into the activities themselves. However, this verysynergy between task and control can have negative side effects in terms of flexibility that areundermining the gains in efficiency. Indeed, the ubiquity of technology in all areas of the enterprisehas displaced the question of efficiency towards the administrative aspect of the task and away fromthe task itself. In this section the two strands of research (organisational and informational) are drawntogether to tease out the dimensions of a framework that might permit researchers to situate theadoption of integrated control systems within the broader context of the organisation and its aims.

1.1 Integration is a design attribute for organisational rationality

To Mintzberg (1989), organisation means collective action in the pursuit of a common mission.Organisation theory is the body of research that addresses itself to the problem of how to organise(Pugh 1997). How organisations should be controlled, by whom, and to what ends, are thefundamental issues for the private sector (Mintzberg 1989). An organisation is instrumentallyrational ifthe job gets done, economicallyrational if it gets done at the least cost (Thompson 1967). Thompson(1967) uses the analogy of the ad-hoc organisation of the community in the face of a natural disasterto emphasise that instrumental rationality can be achieved in the face of necessity, but that economicrationality requires more co-ordination and advance knowledge of what resources are required and

when. Interestingly, research into how managers actually work suggests a constant mode of crisis
mailto:[email protected]:[email protected]:[email protected]:[email protected]


18/96



management (Mintzberg 1989), denoting the inability to work to a co-ordinated plan because of thecontingencies of the moment. It would appear that despite best efforts to control and co-ordinate,managers have to deal with many sorts of unplanned events thrown at them in quick succession andrequiring immediate attention.

Organisational attributes are choices about the optimal design of the organisation in view of its aims.

Some structural variables considered in organisational theory are specialisation (functional and role),standardisation, formalisation, and centralisation (Inkson et al. 1968). An increasingly importantorganisational attribute is the incorporation of information systems into ways of working such that taskexecution and visibility of that performance are instantaneous. The challenge of studying suchorganisational attributes is that there are underlying interdependencies between these attributes andcontextual factors which are difficult to conceptualise. Formalisation of procedures is a step towardsstandardisation, which itself is a step towards centralisation. Furthermore, business context, cultureand managerial capability are additional softer attributes of the organisation which will confoundgeneralisations based on any one set of structural variables. In his work on organisationalconfigurations, Henry Mintzberg (1989) suggests that it is the combination of basic attributes of anorganisation that define its culture, rather than a focus on any one single attribute.

One of the essential dichotomies in the planning and management of routine activities is the trade-off

between control and flexibility. With Enterprise Resource Planning (ERP) systems, theinterdependence between control and other structuring attributes (such as formalisation,centralisation, specialisation) is configured in the software. As researchers had forewarned, thecomplexity of the control system made it necessary to invest in specialised skills to design andmaintain the system (Child 1973). Without the ability to conceptualise integration as an organisationalattribute among others, organisations risk becoming slaves to their self-imposed technostructure.

1.2 The use of information technologies for control

Since the earliest days of the application of information technologies to automate manual tasks, thequestion of integration has intrigued practitioners and academics alike. The potential for thetechnology to make information flow was understood, but its use in planning and monitoring ofperformance entailed a trust in the parameters and algorithms in the machine that had hitherto not

existed. As far back as 1958, Leavitt & Whisler suggested that information technologies wouldundermine the decision making role of the middle manager, consigning them instead to jobs that werehighly structured. The authors argued that the spread of information technology would be rapidbecause it would make centralisation much easier, making top executives less dependent onsubordinates. In addition, the faster processing of information would shorten the feedback loop thattests the accuracy of original observations and decisions.

There is a relationship between the need to process information and uncertainty deriving from theexternal business environment. Organisations require integrating mechanisms (Galbraith inPennings, 1983) for co-ordinating the actions of sub-units towards a common objective. Centralisationand standardisation have been seen as a defensive reaction by organisations when placed underincreasing external control (Mintzberg 1989), and also as a way to improve the efficiency ofinformation processing, at least for routine tasks (Galbraith 1974). An information system by definition

supports the centralisation of control (Markus & Robey 1988). Under conditions of uncertainty, there isa tendency to increase the amount of information sought (Thompson 1967). Earl & Hopwood (inLucas et al. 1980) refer to the tendency in the MIS area to perceive uncertainty as threatening ratherthan inevitable, and, rather than exploiting information for its educative (Gorry 1971) potential,information systems professionals tend to design models that mask reality with assumed certainties.

In parallel to the evolution of the capacity of systems to handle information at speed and in largevolumes, the determinist consequences of information systems for centralisation has been thesubject of much debate (Orlikowski in Knights, D. and Willmott, H. 1988). ERP systemssimultaneously centralise ownership of information resources and democratise access to thatinformation, mixing therefore, to an unprecedented extent, centralising and decentralising effects. Onthe one hand they promote local, cross-functional cooperation and control, breeding skills that areeminently transferable across sub-units. On the other, a best practice model of transaction processing

imposes a common standard across the organisation. As Davenport (1998) notes, the real challengeshinges on where to draw the boundary between centralization of control and autonomy of decisionmaking.


19/96

Fergal Carton and Frederic Adam


Researching the organisational impact of information systems is challenging because the interactionof people with technology in the execution of business processes is so intertwined that it becomesdifficult to differentiate organisational from technical factors (Markus et al. 2000). Researchers mustdifferentiate between the symptoms and the cure. Inefficiency and frustration may be the result ofpoor process design, or incomplete implementation, or a combination of both factors. When the

template falls short of actual physical procedures, data quality will inevitably be impacted.Researchers have noted that although ERP systems may be introduced, physical procedures are notalways changed, such that a mismatch develops between virtual processes and physical processes(Staehr, Shanks and Seddon in Adam and Sammon, 2004; Lee and Lee 2000).

The literature that provides the theoretical framework for this study of integration unites the themes oforganisations and control mechanisms. Organisations require co-ordination mechanisms to transformstrategy into operational targets. ERP systems are considered to be an advanced form of controlmechanism. Managers are the enactors of this co-ordination, and their decisions are informed by dataconcerning both performance targets and actual consumption of resources. ERP systemssuperimpose new patterns on this co-ordination, with an, as yet, poorly researched impact on thequality of decision making.

2. The research objectiveThe objective of this research is to propose a framework which would allow the question of integrationto be modelled alongside other related organisational attributes. It was anticipated that some of theseattributes, identified already in the literature, such as centralisation, standardisation andspecialisation, will figure in the framework. Equally, it was anticipated that other attributes will emergefrom the empirical work, which is based around the implementation of ERP in two multinationalmanufacturing organisations.

The research objective was operationalised into three separate research questions which togetheryield a picture of the scope of integration in the organisation, and a suggestion for theinterdependence between integration and other organisational attributes. Research Question 1 wasconcerned with discovering the key organisational attributes relating to the implementation of ERP.

Research Question 2 was concerned with developing a causality map between these attributes.Research Question 3 drew on the output from Question 1 and 2 to suggest a conceptual frameworkfor the consideration of integration as a set of interdependent organisational attributes. It is envisagedthat this framework could be used by managers to evaluate the potential impact of integration on theorganisation, as well as a providing a diagnostic tool for the isolation of dysfunctional behaviour andits potential causes.

2.1 Profile of the case studies

In this research, two case studies of successful multinational companies (KPC and SIT) are used toexplore the impact of integration on the organisation. Both cases studied are multi-nationalmanufacturing organisations with mature ERP implementations. Table 2 compares the profiles of thetwo cases studied.

Table 1: Comparison of case study profiles

Firm A Firm B

Industry Pharmaceutical Data management

Turnover 05 ($bn) 38.72 9.66

Employees 100,000 26,500

WW operations 119 52

Manufacturing sites 80 3

Head Office London, UK Boston, USA

ERP System SAP R/3 Oracle 11.03

Architecture Single instance Single instance

Server location Pennsylvania, USA Boston, USA

Go-live Phased 2004 Big-bang 2001


20/96



The Key Pharma Company, KPC (real name withheld to allow more detailed reporting), is a leadingmanufacturer of pharmaceutical products, with a highly successful product portfolio in consumerhealthcare, prescriptions drugs and vaccines. With annual sales of nearly 30 billion, and a R&Dbudget of 5 billion, KPC is in a dominant position in its marketplace. The manufacturing organisationinvolves more than 20 autonomous plants worldwide. Managing the supply chain to efficiently satisfydemand is extremely complex. W ith over 30,000 Stock Keeping Units (SKUs) or lines of product, any

one of the 600 sources of demand could be ordering 300-600 SKUs each. The case study focuses onthe KPC plant in Cork, Ireland, a bulk site which is part of the global Manufacturing and Supplyorganisation. KPC Cork ships 4,000 batches of goods per year and local managers are proud of theircustomer service record for deliveries, in the context of the complex and sometimes unpredictablescenario described above. KPC Cork was part of the roll-out programme for a new ERP system(based on SAP version 4.0), which involved all the sites in the Manufacturing and Supplyorganisation. The ERP project had the overall goal of implementing FDA compliant businessprocesses throughout KPC, using the best practice templates that had been designed by KPC aroundSAP standard functionality. The scope of the ERP project at KPC Cork was the integration ofprocesses in all the main business areas (Production, Finance, Sales, Quality), excluding processcontrol at the manufacturing execution level.

The second case is SIT, a market leader in data management solutions, which sees itself as

specialising in helping customers to derive more value from their corporate data. The company isfollowing an aggressive growth path, with 17% growth in consolidated revenue in 2005, which was theyear in which the case study was carried out. Revenues have since then continued their upwardgrowth trend, topping $11.2 billion in 2006. SIT Ltd employs over 26,500 people in 52 operationsworldwide. Manufacturing is concentrated in three sites, one of which is Cork, Ireland. SIT hasevolved into a solutions company, delivering not just hardware, but also information lifecycle toolsand consulting services. A key complexity of such full service offerings is the management of theinformation flows related to executing a single customer order, which increasingly is constituted ofhardware, software and services. Many of these revenue lines are executed by multiple locations,over different time horizons, yet the customer will require a single sales order, single invoice andsingle goods shipment. SIT implemented a single instance global ERP system in 2001. This big bangimplementation addressed user requirements for transaction processing in all back office activitiesrelating to sales order processing, manufacturing, materials planning, distribution and finance. The

Oracle based system supports 4,500 users in 52 countries worldwide, 3 of which involvemanufacturing operations.

3. The research methodology

The premise of this study is that data and process integration are inherent parts of an ERP package.Organisations make considerable assumptions about the level of integration between functions,plants, and headquarters when implementing an ERP system. Although differences exist between thedifferent ERP systems, and these differences will have an impact on the fit with individual businessprocess requirements, it is assumed, as with Gattiker & Goodhue (2005), that the differences are lessimportant than the similarities. The similarities are that all ERP systems are highly integrated at a datalevel, and all ERP systems use workflow logic to automate the flows of information through thedifferent stages of its transaction process. It is via this data and process integration that different

functions can access and transact information relating to the use of common enterprise resources.

Following the Barua et al. (1995) recommendations for evaluating IT investments, Gattiker & Goodhue(2005) selected a fine grained unit of analysis (the plant) and within that the functions examined areclose to the operating core of the business (manufacturing planning and control, as well as executionprocesses). The logic here is that longer lasting transformative benefits on the organisation are morelikely to be derived from core value-adding activities of the company, rather than from administrativefunctions Barua et al. (1995). This research study includes both core operational and supportfunctions, as ERP is inherently an administrative (support function) tool, yet its impact is most felt atthe transactional level (operations). The data from interviews regarding ERP impact was classifiedwith respect to organisational parameters, and these parameters had been identified as seedcategories from the literature, but other themes also emerged from the analysis of managerialperceptions. The seed categories included themes such as centralisation of responsibility,

standardisation of processes and gaps between template process and reality. The emergentattributes include granularity of information, process flexibility and manual manipulation of data.


21/96



Business processes were used to analyse findings as an embedded unit of analysis, in order to betteridentify areas of the organisation where the integration impact was most strongly felt. ERP systemssuperimpose a process based view of business activity, with integration at data and workflow level. Inmany cases, these new processes cross traditional functional boundaries, thus it was consideredimportant to be able to view the impact equally from the process level. The definition of processes is

closely allied the physical management of resources in the organisation, such as inventory, customerorders, cash, labour or plant capacity. Information flows in a logical and chronological order betweendifferent stages of the process as the physical resources are transferred, consumed and transformedthrough different stages of the supply and demand cycles.

Interviews with managers from both cases were carried out in the period from April 2005 to August2005 and involved meetings with 76 managers from different functions affected by the implementationof the ERP system. SIT had gone live on their ERP system in October 2001, so these interviewsreflect the views of managers using a relatively mature system. Table 1 shows a summary of thenumbers of managers interviewed, broken down by case and by function.

Table 2: Breakdown of interviews by case and by functional area

SIT KPC

Function Cork US Total Cork UK TotalFinance 5 9 14 1 1 2

Manufacturing / Distribution 13 6 19 9 7 16

Sales 4 3 7 1 1

IS 4 2 6 3 3 6

Engineering 2 2 1 1

HR 2 2

Total 30 20 50 15 11 26

Interviews were carried out using a semi-structured format, and each interview lasted one hour. Theinterviews were recorded and transcribed, yielding over 400,000 words of raw research material. Arobust coding methodology was applied to reduce the data and avoid paralysis by data analysis (Yin

2003). Observations from the transcripts were extracted to a matrix structured by research question,yielding a total of 3,202 observations. Cell entries were either abridged versions of the original quote,summarised to capture the issue raised. Using hyperlink functionality between Microsoft Excel andWord, each extracted observation was linked back to the original transcript, thereby retaining richnessand avoiding too thin cell entries (Miles & Huberman 1994). The observations were coded at threelevels, identifying the interviewee, the business process commented, and the theme (organisationalattribute). The business process was recorded independently of the functional affiliation of theinterviewee. For example, observations regarding shipment decisions were classified as pertaining tothe Deliver process, and were commented upon by managers from Finance, Materials,Manufacturing, Distribution, and Sales. The themes began as a set of 13 seed categories, developedfrom the literature review, and complemented by themes which emerged from the data.

4. Findings of the case study

The empirical data was analysed to answer the three research questions mentioned in the researchobjective section of this paper. These questions form a vehicle to investigate the effect of ERP drivenintegration in todays multinational company.

Research question 1: Key organisational attributes

Research question 1 identifies the organisational attributes associated with ERP driven integration.These observations were coded based on whether the impact of the attribute was perceived to berelated to organisational, decisional or integration themes, and categorised by process area. The fielddata is aggregated across both cases (n=2,818), and plotted according to three dimensions. Figure 1summarises the results of this analysis of organisational attributes. Organisational themes (y-axisvalues) included Centralisation, Goals, Compliance and Skills. The 695 observations classified under

the organisational themes related to the positive aspects of centralisation, goal clarity, complianceand the development of new skills. Integration themes (x-axis) included Correlation, Granularity,


22/96



Accuracy / Consistency, Automation and Aggregation. Again, the 883 observations in this categoryemphasised the positive aspects of integration, particularly the analytical potential. The Decisionalthemes (denoted by size of bubble) were Manual, Flexibility, Gap Virtual vs. Physical, Uncertainty andLatency. The 1,060 observations under the decisional theme were largely negative feedbackregarding the difficulties of exploiting ERP data.

Plan

Buy

Make

Deliver Quote

to Cash

Report

Approve

& Pay

Forecast

-50

0

50

100

150

200

250

300

350

400

-50 0 50 100 150 200 250 300 350

Integration attributes

Organisatio

nalattributes

Figure 1: The organisational impact of ERP integration by business process areaThese results reveal a strong linear relationship between centralisation and integration. That is, thestronger the consensus around the need for a centralising force in the organisation, the stronger theperceived need (among managers) for integration. This relationship holds across the different processareas. It could be deduced from this data that organisational goals are well understood andassimilated in the different functions, and that centralisation of the organisation is wholly accepted.There is a good understanding among managers of the impact of centralisation for data integrity, anda good understanding of the potential to use the associated wealth of highly granular information fordifferent decision scenarios. This linear relationship would seem to suggest a broad acceptanceamong managers that centralisation is necessary from the point of view of goal focus, and that themore centralised the organisation, the more integration is required and accepted. The corollary of thisfinding, not tested in this sample, would be that more decentralised organisations would have difficultyaccepting integration. On the other hand, what is equally striking from these results is the level ofunanimity in the frustration felt throughout the business in exploiting corporate information for decisionsupport. Because of performance and security issues with the technology, and its fit with actualbusiness activities, managers are obliged to resort to much manual manipulation of the informationderived from the ERP system in order to get to the meaning they require.

It was of interest to note that the acceptance of centralisation varied between the two cases observed,with SIT showing more consensus on centralisation for Buying and Shipping, than KPC who weremore forceful in their support for centralisation for Planning and Making. It is thought that theregulatory environment explains the KPC willingness to accept compliant manufacturing processes,whereas SIT had moved Purchasing to a global process, and were debating how to change theDeliver process from a site specific activity to a global basis. These nuances demonstrate that theacceptance of integration should be considered from a site and functional level.

Research question 2: Causality map between attributes


23/96



The analysis of Research question 1 served to establish the existence of a strong relationshipbetween the centralisation attributes and the integration attributes. Integration is accepted bymanagers as it is understood that it is a pre-requisite for better control and visibility of organisationalactivity. The granularity of information available possibly encourages requirements for managers tocorrelate data in ways that were not envisaged by the integrated applications used, and withinformation from other sources. The increased accuracy of information benefits Manufacturing and

Finance particularly. The downside of this integration is that ERP systems are inflexible andunforthcoming with meaningful information. Organisations seek to exercise control and reduceuncertainty by multiplying the number of data points, but the implementation of tools and proceduresto collect the data in a changing reality becomes a cause of bureaucratic inflexibility. This inflexibility isa cause for users to resort to workarounds, which engender gaps between the physical reality and thevirtual picture used to monitor its progress. These gaps or virtual blind-spots mean managers return tomanual methods in manipulating information for decision making purposes. This additionalmanipulation introduces latency in reporting, which is compounded by response time issues derivingfrom the technical architecture. The net result is a requirement for skilled resources in themanipulation of corporate data to derive meaning. Thus, as integration increases with centralisation,there is a corresponding increase in the necessity for skills to manipulate and digest the ensuing glutof data. Finally, and perhaps most worryingly, users in the most sensitive execution areas(Manufacturing and Deliver processes) are handicapped by inflexibility in the transaction processing

system, whereas their focus should be on material availability and customer satisfaction respectively.

These elements are depicted graphically in Figure 2.

Figure 2: Understanding the causality between control, integration and its impact

Research question 3: Conceptual framework for ERP integration and functionality gaps

The first two research questions highlighted the themes related to the impact of ERP drivenintegration on the organisation, and the causality associated with these themes. The final researchquestion draws on this output to suggest a framework for the conceptualisation of integration in theorganisation, named here as the Zipper effect and shown in Figure 3. At go -live (t0), a certainnumber of gaps will exist between the configured ERP processes and the way the company actuallyworks. These will typically be dealt with by workarounds, but managers do not have visibility of theimpact for data integrity of these workarounds, or their cost. The suggestion behind the Zipper effect

is that these gaps between the physical and the virtual will inevitably emerge over time, but thatacknowledging that they exist is the first step in addressing them. The second step is to establish

Control

CentralisationCo-ordination

Integration

GranularityAccuracy / consistency

Inflexibility

LatencyManual

More analysis potential

Deteriorating data integrity

Technical skills required for reporting


24/96



processes for monitoring these gaps, analysing and explaining them. Once recognised, the third stepfor organisations is to have the resources and procedures in place to be able to resolve them.

Data integrity gaps evolve over time

t1

t2

tn

t0

Reality

(physical consumptionof resources)

ERP

(virtual picture ofuse of resources)

Workarounds found or

changes are made to

ways of working to

maintain ERP data integrity

Business model changes,

new channel partners,

outlsourcing supply chain,

volumes increase,

margins shrink,

Go-live

Go-live

+ n years

Gaps in data integrity

ERP templates customised

to allow for company

specific requirements

Gaps filled at transactional

level by more workarounds

or system customisation

or at reporting level with BI

Increasing use of tools to manipulate

data in order to derive meaning from data

(corporate DW, BI tools and spreadsheets)

Figure 3: The Zipper effect: modelling the divergence of ERP from reality

5. Conclusions

This research highlights a fundamental difficulty for the use of ERP systems in large corporations.Integration involves virtualising the communication process through the use of information systemssuch as ERP. The accuracy of the information is a function of the closeness of the templateparameters to the actual way of doing business. The two cases studied demonstrate that thevirtualisation of business processes engenders constraints that negate the original purpose ofimproved control. These constraints arise from either the virtual or the physical context. The

combination of a centralised technical architecture and large volumes of highly granular informationcreates latency issues (virtual context) which impacts at both an execution and a reporting level. Theinflexibility of the standardised ways of working to the evolving business imperatives (physical context)obliges managers to resort to manual methods when analysing performance variances, defeating theoriginal aim of an integrated approach to performance control, and incurring an additional cost.

Another key lesson from this study has been that an integrated model of how businesses function istheoretical and ill adapted to the reality of how businesses operate. Managers in both studiesconfessed to their inability to build conceptual models of the decision processes required in thedemand and supply processes of the firm because of their sheer complexity. In such complexscenarios, the imposition of standardised procedures appears to inhibit rather than encourage anunderstanding of the key variables. Accordingly, the notion of integration should be treated as anormative force in organisational administration, along with other organisational attributes, but not as

a once-off solution embodied in the ERP software. Raising the awareness of integration as a


25/96



parameter of organisational design should be an objective for researchers and practitioners alike,rather than being left in the hands of vendors of solutions.

References

Adam F. and Sammon D. (2004) The Enterprise Resource Planning Decade: Lessons Learned And Issues ForThe Future, IPG, Hershey, PA

Barua, A., Kriebel, C. H., and Mukhopadhyay, T. (1995) Information Technologies and Business Value: AnAnalytic and Empirical Investigation, Information Systems Research, Vol. 6, No.1, pp 3-23

Child, J. (1973) Strategies of Control and Organisational Behaviour, Administrative Science QuarterlyDavenport , T.H. (1998) Putting the Enterprise into the Enterprise System, Harvard Business Review, Jul-AugDearden, J. (1972) MIS is a mirage, Harvard Business Review, Jan/Feb, Vol. 50, No. 1, pp 90-99Gattiker, T.F. and Goodhue. D.L. (2005) What Happens After ERP Implementation: Understanding the Impact of

Interdependence and Differentiation on Plant-Level Outcomes, MIS Quarterly, Vol. 29, No. 3,September, pp 559-585,

Galbraith J.R. (1974) Organisation design: an information processing view, Interfaces, Vol. 4, No. 3, pp 8-37.Gorry, G. (1971) The Development of Managerial Models, Sloan Management Review, Winter, pp 1-16Gorry G. & Scott Morton, M. (1971) A Framework for Management Information Systems, Sloan Management

Review, Fall, Vol. 13 No. 1, p 49-61Inkson, J., Hickson, D. & Pugh, D. (1968), Administrative reduction of variance in organization and behaviour,

unpublished paper given to the British Psychological Society, Annual Conference, April

Knights, D. and Willmott, H. (1988), New technology and the labour process, Macmillian, London, pp 20-49Leavitt, H. and Whisler, T. (1958) Management in the 1980s : New information flows cut new organisation

channels, Harvard Business Review, Nov-Dec, pp 41-48Lee, Z. and Lee J. (2000), An ERP implementation case study from a knowledge transfer perspective, Journal of

Information Technology, Vol. 15, pp 281288Lucas et al. (1980) The Information Systems Environment, North-Holland Publishing Company, IFIPMarkus, M.L. and Robey, D. (1988) Information technology and organisational change causal structure in theory

and research. Management Science, Vol. 34, No. 5, May, pp 583-598Markus et al. (2000), Learning from adopters experiences with ERP: Problems encountered and success

achieved, Journal of Information Technology, Vol.15, 245265Miles, M., and Huberman, A. (1994) Qualitative Data Analysis: An Expanded Sourcebook, 2

ndedition, Sage

Publications, CAMintzberg, H. (1989) Mintzberg on Management, The Free Press, New YorkPennings (1983) Decision Making : an Organisational Behaviour Approach, Markus Wiener Publishing Inc., New

York, pp 131-139Pugh, D.S. (1997), Organization Theory: Selected Readings, 4th

edition, Penguin Books, EnglandSimon, H. (1977) The new Science of Management Decision, 3

rdedition, Prentice Hall, Englewood Cliffs, NJ

Thompson J. (1967) Organisations in Action, McGraw-Hill, New YorkYin, R.K. (2003) Case Study Research, Design and Methods, 3

rdedition, Sage Publications, CA


26/96




27/96

ISSN 1566-6379 27 Academic Conferences LtdReference this paper as:Cragg, P, Mills, A and Suraweera, T. (2010) Understanding IT Management in SMEs The Electronic JournalInformation Systems Evaluation Volume 13 Issue 1 2010, (pp27 - 34), available online at www.ejise.com

Understanding IT Management in SMEs

Paul Cragg, Annette Mills and Theek SuraweeraUniversity of Canterbury, Christchurch, New [email protected]

Abstract: There is evidence in the IT literature indicating that IT management is one factor that influences ITsuccess. In addition, there is much literature indicating that IT management is important in the SME context.However, much of this literature has focused on the important role of the

ejise volume 13 issue 1

Documents