MCM D 09 00747 Reviewed Copy

8/3/2019 MCM D 09 00747 Reviewed Copy

1/39

Mathematical and Computer Modelling

Manuscript Draft

Manuscript Number: MCM-D-09-00747

Title: Modeling the technology transfer process in petroleum industry: evidence from Libya

Article Type: Regular Article

Keywords: technology transfer, petroleum industry, Libya, structural equation model

Abstract: The purpose of this study was to propose a conceptual model for Technology Transfer (TT)

that houses the several factors. These factors believed to influence the processes effectiveness and

guided TT performance. In addition, to explore TT performance and the relationship between TT

government support, infrastructure, TT environment, and TT learning capability. The Oil production in

Libya is dependent on the foreign technology transferred into the country by foreign multinational

petroleum companies. During the eighties, the Libyan government launched a program of development

"Libyanazation" in the Libyan petroleum industry in effort to create an absorptive capacity to acquire

petroleum technology dominated by foreign companies. This study evaluates the level of technicalchange because of technology transfer programs and the impact on knowledge and competitiveness

performance of Libyan petroleum industry. A questionnaire survey was administered to companies in

the Libyan petroleum industry. There were 201 responses from industry professionals in the Libyan

petroleum industry that were analyzed using structural equation modeling (SEM) exploratory factor

analysis (EFA) and confirmatory factor analysis (CFA). In addition, the significant of direct and indirect

interrelationships between model factors was determined through (SEM). A path model was estimated

and specified to include three process enablers, namely government support, host characteristics, and

learning technology capability, and one outcome factor named TT performance. The results suggested

that government support factor (government support, laws and regulations, petroleum industry

strategy, international quality standards, and information technology) and technology learning

capability factor (i.e. supervision, adoption, teamwork, absorption, training, technology complexity,

and industry knowledge) were determined to be the key predictor of TT performance to the host

petroleum industry.


2/39

0

5

10

15

20

25

30

35

40

45

0-5 6-10 11-15 16-20 21+

Petroleum industry experience

Percentage

ofresponden

ts

Figure 2: Respondents petroleum experience

ure


3/39

0

10

20

30

40

50

60

70

1-2 3-4 5-6 7-8 9+

Past participation in TT processes

Percentage

ofrespondent

s

Figure 3: Respondents previous involvement in TT processes

ure


4/39

0

5

10

15

20

25

30

35

40

Manager

Project

manager

Superi

ntend

ent

Project

engin

eer

Supervi

sor

Engin

eer

Technic

al

Othe

rs

Respondents positions

Percentage

of

respondents

Figure 4: Positions of respondents.

ure


5/39

3.933.78

3.923.87 3.84

1

1.5

2

2.5

3

3.5

4

4.5

5

UK Germany Italy Austria Others

Nationality

TTs

uccessrate

Figure 5: Nationality success rate

ure


6/39

0

10

20

30

40

50

60

70

80

1990-1994 1995-1999 2000-2004 2005-2009

Technical process Training Advanced technology Petroleum project Study abroad

Figure 6: TT process description and year completed

ure


7/39

0

10

20

30

40

50

60

70

1990-1994 1995-1999 2000-2004 2005-2009

Year completeted

TTt

otalprocesses

UK Italy Germany Austria Others

Figure 7: TT processes count and nationality

ure


8/39

TLC

Training

1

Knowledge base

Technology complexity

Teamwork

Attitude

Capability

Supervision

GSLaws and regulations1

Government TT plan

NOC strategy

Standards and quality

Information technology

HCResearch and development

1

Sub-contractors

Communications

TTP

Competitiveness

1

Overall performance

Industry konwledge

Skill base

Financial performance

Schedule performance

e1

1

e2

1

e3

1

e4

1

e5

1

e6

1

e7

1

e8

1

e9

1

e10

1

e11

1

e12

1

e13

1

e14

1

e151

e16

1

e17

1

e18

1

e19 1

e20

1

e21

1

Figure 8: Confirmatory factor analysis (CFA) design

ure


9/39

Figure 9: Path model for TT in petroleum industry*Note: Value in parentheses are unstandardized path coefficients, values not in parentheses are standardized

path coefficients, ***p < 0.001.

Government

Support

TT

Performance

Host

Characteristics

Technology

Learning

Capability

0.42 (0.43) ***

0.45 (0.44) ***

0.36

(0.34)***

0.68

(0.72)***

ure


10/39

Figure 10: Respecified path model for TT in petroleum industry.

Note: Value in parentheses are unstandardized path coefficients, values not in parentheses are standardized

path coefficients, ***p < 0.001

Government

Support

TT

Performance

Host

Characteristics

Technology

Learning

Capability

0.46 (0.46) ***

0.45 (0.44) ***

0.40

(0.39)***

0.68

(0.72)***

ure


11/39

Figure 1: Conceptual model for Technology Transfer in Libyan petroleum industry

Technology

Transfer

Support

Technology

Transfer

Infrastructure

TechnologyTransfer

Performance

Petroleum

Industry

Learning

Capability

TechnologyTransfer

Environment

ure


12/39

Modeling the technology transfer process in petroleum industry: evidence fromLibya

1A.S. Mohamed, 2S.M. Sapuan, 3M.M.H. Megat Ahmad,4A.M.S. Hamouda and

2B.T. Hang Tuah Bin Baharudin

1Institute of Advanced Technology, University Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia2Faculty of Engineering, University Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia3Faculty of Engineering, University Pertahanan Nasional Malaysia, 57000 UPNM, Kem Sungai Besi,Kuala Lumpur, Malaysia4Mechanical and Industry Systems Engineering Qatar University, P.O.Box 2713, Doha,Qatar.

Abstract

The purpose of this study was to propose a conceptual model for Technology Transfer(TT) that houses the several factors. These factors believed to influence the processeseffectiveness and guided TT performance. In addition, to explore TT performance and therelationship between TT government support, infrastructure, TT environment, and TTlearning capability. The Oil production in Libya is dependent on the foreign technologytransferred into the country by foreign multinational petroleum companies. During theeighties, the Libyan government launched a program of development Libyanazation inthe Libyan petroleum industry in effort to create an absorptive capacity to acquirepetroleum technology dominated by foreign companies. This study evaluates the level oftechnical change because of technology transfer programs and the impact on knowledgeand competitiveness performance of Libyan petroleum industry. A questionnaire surveywas administered to companies in the Libyan petroleum industry. There were 201responses from industry professionals in the Libyan petroleum industry that wereanalyzed using structural equation modeling (SEM) exploratory factor analysis (EFA)and confirmatory factor analysis (CFA). In addition, the significant of direct and indirectinterrelationships between model factors was determined through (SEM). A path modelwas estimated and specified to include three process enablers, namely governmentsupport, host characteristics, and learning technology capability, and one outcome factornamed TT performance. The results suggested that government support factor(government support, laws and regulations, petroleum industry strategy, internationalquality standards, and information technology) and technology learning capability factor(i.e. supervision, adoption, teamwork, absorption, training, technology complexity, andindustry knowledge) were determined to be the key predictor of TT performance to thehost petroleum industry.

Keywords: technology transfer, petroleum industry, Libya, structural equation model

1. Introduction

In the developing countries, technology transfer is the solution to the improvement ofindustrial and economic sectors. However, the success of any transfer depends on theproper choice of the proper technology from the right provider as well as the absorptivecapacity of the technology. Evaluating technology transfer effect in the petroleumindustry is crucial to any petroleum company. More than ever as the price of petroleumproducts and the cost of oil production were increased dramatically in recent years.

nuscript

k here to view linked References
http://ees.elsevier.com/mcm/viewRCResults.aspx?pdf=1&docID=1171&rev=0&fileID=8349&msid={D185EE64-D00D-417D-BA94-3FCE5320B6F4}http://ees.elsevier.com/mcm/viewRCResults.aspx?pdf=1&docID=1171&rev=0&fileID=8349&msid={D185EE64-D00D-417D-BA94-3FCE5320B6F4}


13/39

Although, most petroleum-producing countries were committed to develop its petroleumindustries so become competitive, compatible, and reliable. This often involves setting uptechnology transfer infrastructure considered necessary for the petroleum industry.Nevertheless, most of these countries lack the managerial and technical expertise tomanage such large projects. On the other hand, the petroleum industry was slow to acceptand adopt new technologies. When oil was discovered in Libya 1959, and oil exportsbegan in 1961, the country then had very little human resources to manage and operate asizable modern petroleum industry. The foreign oil companies not to mention a veryweak government then played major role in establishing the conditions for theestablishment of linkage and minor change capabilities, especially in developing humanresources for the petroleum industry. Technology Transfer (TT) continues to be a keyenergizer to industrialization and economic expansion in developing countries, mostly inthe fast growing oil producing countries such as Libya, Algeria, Nigeria, etc.

2. Literature Review

In the study of TT, numerous models had been developed to analyze or the TT process[1-4]. None of these studies concentrated on the petroleum industry. Nevertheless,

measuring the impact of transferred technology changes researchers and evaluators. Inaddition, not all TT studies models were backed by strong observed data analysis. Inmany models, finding out the meaning of TT effectiveness was intimidating. Theresearchers had tried to define the TT term in a different ways because of their individualareas of study. Chacko [5] came up with a definition of TT in a scientific manner asconverting physical or mental matter or energy into direct usable alternate form. Williamsand Gibson [6] defined TT as the shared responsibility between the source and thedestination by ensuring the technology was accepted, or at least understood, by a userwho owns the required knowledge and the resources to apply the technology. In theconstruction sector, Simkoko [7] tried to model on this definition by identifyingindividual construction resources, as either materials or permanent equipment (e.g. steel

beam, elevators, material) or construction-applied resources (e.g. information, skill).According to Waroonkun and Stewart [8] TT has been defined as when all types ofknowledge about the construction field (e.g. design, construction process, material use,equipment utilization, etc.) are transferred from a foreign party (transferor) to a host party(transferee) that arranges to receive it. For the purpose of this study, TT process inpetroleum industry has been defined as when some form of knowledge, material, orequipments are transferred from one foreign party such as person or organization toanother local party as a person or organization that arrange to receive it. Explicitly, thehost industry refers only to Libya petroleum companies fully owned by National oilassociation (NOC) or joint ventures and the foreign refers to the foreign company ororganization working with Libyan petroleum companies to acquire projects or perform

TT process. As expected, most foreign companies had origins in developed nations suchas the United States, United Kingdom, Italy, Germany, France, Australia, etc.In this study, we developed and empirically tested a model that related to severalantecedents factors to transfer of foreign developed technology by petroleum companiesin Libya. The model, shown in Fig.1, was based on extensive literature in TT processesand literature on technology transfer. The TT model was specifically designed to beapplied to the study of technology transfer from developed countries to the Libyanpetroleum industry. The TT process was not been tried before, and we seek evidence


14/39

supporting the relationship identified in our model via exploratory and confirmatoryfactor analysis. We believe in the importance and the uniqueness of the Libyan petroleumindustry TT processes. The exploratory test of our model was conducted considering thefactors adopted from previous studies. Libya is arguably one of the most prominentpetroleum producing countries in the world today. Libya's petroleum industry is currentlyundergoing rapid expansion, and technology transfer for petroleum industry developmentis likely to be an important engine for economic developments in Libya. We used aconvince sample of respondent representing over 30 petroleum companies across severalpetroleum industries in Libya: oil production, oil exploration, petroleum technology, oilrefining, and petroleum marketing. The specific behavior being modeled was the transferof foreign developed technology by Libyan petroleum employees in the industry.Calantone et al., [9] model consisted mainly of five components that capture TT process,which was constructed based on Boddewyns [10] study on comparative marketingresearch. Measuring TT process feedback was the main objective of the model. However,the model failed to include TT process performance indicators. Furthermore, the complexdesign model was not been empirically verified. Elements of the model would be suitableto be adapted to the petroleum industry TT model.Simkoko [7] focused on TT in the construction industry of developing countries. Based

on case studies of 12 international construction projects in developing countries ofAfrica, South America, and Asia in 1987 and 1988, the study was performed. Datacollection was conducted into two schedules. One involved examination of project filesand two involved site visits and further interviews project participants. The objective ofthis study was to examine the impact of TT programs and other internal and externalenvironment factors on construction project performance. This study is now oldconsidering the development of more advanced TT mechanisms. This study was onlyinvestigates the development of technological and management, rather than attempting tomodel the TT process.Kumar et al., [11] identified key elements that affect the ability of firms in developingcountries. The paper studied the Indonesian manufacturing sector and it developments in

recent decades, which has become the largest segment of the economy and is growing bynearly 10% annually. Kumar et al., [11] study was concentrating on small scalemanufacturing industry which considered a major hinder as petroleum industry in mostcases involves large scale technological changes. Nevertheless, some of the learningcapability model is suitable to be incorporated in the petroleum industry proposed TTmodel; these factors namely are government role and sub factors in the learningcapability that include training and R&D.Lin and Berg [2] carried out an exploratory study into the effects of cultural difference onTT projects. The aim of this study was to provide empirical evidence that confirms theconceptual models developed by other researchers in the field of TT. The Lin and Bergstudy focused on TT projects involving Taiwanese manufacturing companies. Three

groups of factors, previously examined in conceptual studies were investigated: nature oftechnology; previous international experience, and the cultural difference between thetechnology provider and receiver. An important conclusion made in their study was thatTT study investigations should not be limited to only examining the direct effects ofidentified factors and associated variables. It was also important to examine causalinteractions between factors to achieve an accurate representation of the TT process.Many of the factors and associated variables identified in this investigation were utilizedto develop the conceptual model for TT in petroleum industry described later. However,


15/39

important influences such as government policy, mode of transfer, to name a few, havebeen neglected. The study does not adequately address all aspects of cultural differenceleaving the framework somewhat incomplete.Another model was developed by Malik [3] concentrated on intra-firm TT. Because themodel was tested only on one manufacturing company, the supporting empirical evidencemight be having some biased testing. It should be noted that Maliks model was useful indeveloping the conceptual model for TT in petroleum industry projects in terms ofidentifying the interactive nature of the communication process.The study of Wang et al., [12] followed from extensive previous research in the field ofTT. His paper is primarily concerned with the transfer of knowledge from a multinationalcompany to a subsidiary. The Wang et al., [12] model was developed from semi-structured interviews with 62 multinational companies operating in China. However, themodel was also limited by the scope of the TT process that was examined. Scope wasconfined to the amount of knowledge that a subsidiary of a multinational companyacquires because of the transferor and transferee characteristics. The model was notsuccessful to examine other influential factors such as government influence; technologycharacteristics, etc. economic advancement, and competitiveness were factors to beadopted in the model for TT for the petroleum industry.

Waroonkun and Stewart [8] model attempted to estimate performance rates of TT indeveloping countries, the study proposed a conceptual model for TT that accommodatesthe several factors thought to impact on the processes effectiveness and derivedoutcomes. In their study, the transferee refers only to Thai Architectural, Engineering,and Construction (AEC) firms and the transferor refers to the foreign AEC firms workingwith Thai firms to secure projects. The model depended on the maturity level oftechnology of the host nation. Justifiably, the model was designed for constructionindustry and may not be accurate to other industries. However, some variables of theirmodel were adopted in this study.

3. Conceptual model of technology transfer in petroleum industry

The development of conceptual model for TT in the petroleum industry has been pointedtoward the capture all of the significant factors that influence the effectiveness of the TTprocess and the resulting performance. These relevant factors have been adapted from theexamined leading studies into the TT phenomenon with the objective to develop a model,which explains the TT process in petroleum industry. Through a process of categorizingvariables taken from the previous studies and conceptualizing their relationship with oneanother in the petroleum industry context a number of factors were identified. The factorsidentified were classified as enabling and TT outcome factors. The classification ofvariables into their relevant factors; namely, TT support, TT infrastructure, TTenvironment, TT learning capability, and TT performance has not been solely based on

other studies but rather is a conceptualization based on understanding of TT and thepetroleum industry. The structure and links between the model constructs have also beenconceptualized based on some empirical understanding and therefore require testing toconfirm their appropriateness and validity. Figure 1 illustrates the conceptual model onhow the developed enabling factors interact to create value for the host petroleumindustry. The four main TT enabling factors are shown at the left and middle of themodel as the constructs TT support, TT infrastructure, TT environment, and TT learningcapability. The outcome factor, TT performance, has been presented at the right of the


16/39

model. The arrows represent the hypothesized causal paths between each enabling andthe outcome factor. Each of these causal paths described in the following paragraphs.

INSERT > [Figure 1: Conceptual model for Technology Transfer in Libyan oil industry]

The variables contained in the TT support factor were found to have direct impact onthose variables contained within the TT Infrastructure factor. Therefore, the link from TTSupport to TT Infrastructure was constructed in Fig.1. Similarly, the literature providedsome evidence that the following causal relationships also existed: TT support TTenvironment; TT Support, petroleum industry learning capability; TT Infrastructure TTenvironment; TT infrastructure TT performance; TT infrastructure TT petroleum industrylearning capability; TT environment petroleum industry learning capability; TTEnvironment TT performance; and petroleum industry learning capability TTPerformance. Although these links were said to have been described in past literature,this does not mean that every variable contained in each factor impacts on every variablein another, rather that the factor when considered as a whole has impact on another. Thethree links from TT Support to TT infrastructure, TT environment, and petroleumindustry learning capability were insufficiently supported in the literature and therefore

their validity was closely examined. Many enablers have been identified as havingpotential to impact on the effectiveness of the TT process and have been divided into fourmain categories in this study: transfer support, transfer infrastructure, transferenvironment, and petroleum industry learning capability. The following sections describethe persuasion for including each variable in the aforementioned enabling and outcomefactors.

3.1 TT Support

This factor was predominately concerned with the impact of government relatedinfluence on the TT process[13]. According to Kwon and Zmud [14] , The availability of

financial resources for the petroleum manufacturing industry technology must beconsidered during the TT process. The financial sub factor has been recognized in severalstudies as an influential factor that impacts on the effectiveness of TT process [15]. Theorganizations strategy towards the technology to be transferred affects the efficiency andinteraction pattern between the holding and its sub companies during TT process [2]. Thebusiness strategy is concerned with the overall purpose and long term direction of theparent organization and its financial viability [16]. A TT process supported bygovernment can decrease the technological gap between local and foreign companies byestablishing innovation system and policies that encourage technology research anddevelopments (R&D) [17]. There is also a direct link between the level of governmentsupport to the industry R&D and training to apply TT process. The government must plan

ahead of time in global petroleum industry to make the industry competitive and investorfriendly in the world [18]. Government support of petroleum industry technology hasbeen identified as an important consideration in the success of a TT process as it hasimpact on several other influential factors. The study adopted four variables for thisfactor (laws and regulations, government TT plan, NOC strategy, and NOC rewardsystem) from previous studies, which had been incorporated into the conceptual model.

3.2 TT Infrastructure


17/39

The information technology (IT) and its impact on the TT process is unquestioned amajor concern when managing TT process[19]. The innovative use of a variety of ITtools maybe provide benefits to facilitate TT process [20]. According to Nazmun et al.,[21] IT can increase capacity as well as decrease expenditure of information handlingwhich will in turn enhances the success of TT process. Local industry should have aninteraction with local R&D centers and universities[22, 23]. However, The TT may notbe materialized if technology gap between foreign company and local is too large, itsgenerally believed that local participation with foreign firms reveals the proprietaryknowledge in that way facilitate TT to domestic industry [24]. Notwithstanding, trainingis important component of TT process. TT through training could be in the form ofpractical training where local employees are exposed to working methods and required towork at highly developed industry environment to adopt new skills and techniques [25].Growing attention has been turned to the possible role of TT agreements as part of thearchitecture of TT process [26]. The foreign company shall take all viable steps topromote, facilitate, and finance as appropriate the transfer or access of soundtechnologies and know-how to local industry [27]. Management as an important subfactor can take actions to develop an infrastructure that is supportive of TT process.

Management approach would significantly contribute to low or high TT processperformance[28]. The six variables of this factor (IT, R&D, Sub-contractor, training,standards and quality, and management practice) had been incorporated in the conceptualmodel from the previous studies.

3.3 TT Environment

A major concerns of managing TT is the environment in which the interaction betweenthe foreign technology provider and the host industry and its effect on the success of TTprocess performance [2, 29]. Williams and Gibson [6] suggested that TT should beconceptualized as communication process where gaps between foreign and local

environment will affect efficiency of inter firm communication and the overalleffectiveness of TT process. Wei [30] reported that prior international technologyexperience on international operations is helpful for the host to gather relevantinformation during TT process . On the contrary, Lin [2] suggested that previous foreignexperience of the host can increase its capability to preserve core technology from theforeign and eventually resulting in the host becoming a serious competitor of the foreigncompany. Actually, among all the resources of a firm, knowledge is the most strategicallyimportant resource [31]. Knowledge provides the capacity for organizational action andnew knowledge provides the capacity for organizational renewal [32]. Nonaka [33] forinstance argues that tacit knowledge accounts for three quarters of all knowledge used byfirms. A complex system or technology may need a longer time, more technical people,

and higher capital investment to be transferred. Madeuf [34] therefore suggests thatnature of technology will affect TT effectiveness and need to be carefully investigated inmanaging a TT process. Most technologies are very difficult to transfer because theyinclude a large portion of tacit knowledge. According to Nanoka [16] tacit knowledge isnot easily visible; not easily expressible; highly personal; hard to formalize; and difficultto communicate. Several dimensions are proposed to characterize the nature of atechnology to describe its transferability [35]. Robinson [36] proposes that the skill andeducation level required to adopt a technology by the technology receiving team is an


18/39

indicator of the complexity of a technology. Therefore, a TT process to transfer acomplex technology is likely to have a lower success rate. Achieving success in TTprocess will require the information to be conveyed clearly and effectively in a total errorfree communication setting. Successful TT process requires many factors in particular ahigh level of commitment to shared goals. Saunders [37] provided a model that ischaracterized by frequent communication both formally and informally, open sharing ofinformation. Carolynn et al., [38] indicated that effective communications was given avery high rating by all categories and by both organizations involved in the process.Zeller [39] describes the introduction of cross functional teams as part of the reorganizingof R&D activities within pharmaceutical companies in response the increasingglobalization of R&D. More recently, Michie and Sheehan [40] considered firms withhigh levels of participation in teamwork as part of their examination of the impact ofalternative systems TT process. The five factor variables (experience, knowledge base,technology complexity, communications, and teamwork) had been adopted from severalstudies to be used for the conceptual model.

3.4 TT Learning Capability

The learning capability is concerned with the affects of the sub-factors that facilitate thetechnology that being transferred between host and local companies. The issue of culturein TT process has been studied by Kedia [41] and concluded that if foreign and hostcompanies did not emphasis the issue of culture, the result maybe unproductive TTprocess. The importance of recognizing the apparent and hidden components of the hostcountry culture involved in TT process depends on several factors such as attitudetowards foreigners and the company reputation [42]. The cultural traits of the two partiescan have a significant impact on the effectiveness and hence the success of TT process[43]. A TT process where cultural gap between host and foreign is high is expected toresult in unsuccessful TT process [2]. Adoption of new technology requires somemodification to fit with changes in the working environment by controlling the working

environment variables or making an adjustment to synchronize between host and foreigncompany policies [44]. The importance of adoption is crucial because maybe the inputsfrom host is not the same as for those the equipments was designed. The ability of anycompany to absorb the advanced technology depends on the organizational and technicalcapabilities of the company [45, 46]. Structural systems should adopt the measure ofquality, performance and promote learning within the company [47]. The company'scurrent absorption capacity will be determined to what extent their ability to participate inthe transfer of technology [30, 48, 49]. Recent research by Escribano et al., [50]suggested that the capacity for absorption is in fact a source of competitiveness. In TTprocess, exposure occurs when employees become informed and educated about thetechnical and manufacturing systems and their applications that was not diffused or

applied in their industry environment previously. In their research, Arbose [51] arguesthat many of the users who have an earlier exposure to technology have a greater capacityto accept TT process. Recently, Liu [52] suggested that employees of technologicalknowledge had not been exposed to external expertise will need to expose to foreignknowledge. Respectively, the company, which had been exposed to foreign expertise,will develop methods for local R&D, meanwhile the company, which has not beenexposed, has to depend and more foreign expertise. More recently, Feldman andBercovitz [53] in their study on in university technology transfer using data on individual


19/39

researchers from the medical schools of Duke University and Johns Hopkins University.They found that a high percentage of supporters of TT process and new organizationalstrategic initiatives were the more recently trained the staff members, the more likelyencountered an environment supportive of commercialization activity. There is littleinformation on the role and impact of the supervisory in the TT process. However, Miles[54] suggested that one of the reasons for the failure of TT process is incautious choice ofthe supervisors for the TT process, without considering the required knowledge for TTprocess. Susan E. Cromwell [55] suggested that supervisors who did not have the propersetup and their management support and has participated previously in TT processeswould excel in future TT process. In contradiction, supervisors who reported lessparticipation and did not have time and enough support in previous TT processes wouldbe considered a real barriers to TT process. Five variables of this construct (culture,attitude, capability, exposure, and supervision) were combined based on several studiesand applied in the conceptual model.

3.5 TT Performance

Performance of TT process could be examined from numerous attributes in the literature.

From the viewpoint of efficiency, Teece [56] tried to measure the effectiveness of TTprocess through the calculation of the cost of TT. Viewing From the perspective of TTprocess within the organization, Schwarz [57] defined by the effectiveness of TT as anadequate R&D in the local organization. Similar foundations evident in the work byAlam [58]. Zakaria [59] discussed the transfer of technology to the petroleum industry onthe bearing capacity of the country to purchase or lease of the best technologicalequipment. Manson [60] determined that the effective transfer of technology must bemeasured if it facilitates to develop the methods and the acquisition of new skills.Mytelka [49] suggested that the incorporation of technology is the best and only way topossibly modify, improve and extend it later. One of the leading motivations fordeveloping countries to adopt and apply TT programs is the expectation that TT would

enhance the standard of living [61, 62]. Economic development sub-factor is concernedwith the extent of competitiveness between the Libyan petroleum industry companies inlocal markets and global markets. In addition to the economic benefits expected to beobtained by the transfer of technology, the local petroleum industry could also benefitfrom cognitive development at the level of individual users, as well as at the enterpriselevel [63, 64].Preliminary results of TT process are the transfer to the individual employee the tacitknowledge to explicit knowledge [65]. Knowledge gain sub-factor is concerned with theimprovement cause by TT process in the knowledge of local industry employees,improvements caused by the TT process and its impact on the methods and technicalskills of the employees. The TT process has a number of results at each stage and

between stages and at the final stage, which sometimes delayed because of the length ofthe development process [66]. According to Devapriya and Ganesan [67] the keymotivations of the TT process in any industry are: highly effective financial performance,efficient schedule performance, and significant quality operating performance. Enhancedcommendable performance must be the key outcome of the TT process. The projectperformance factor main objective is to measure the impact of TT effectiveness on theindustry [68]. In particular, from an evaluation perspective of performance it would be interms of progress of financial, schedule, and quality assessment. The output of TT


20/39

process is measured in comparison with the objectives identified in advance with mostemphasis on time, cost and quality [7]. The variables (competitiveness, performance,working practice, skill base, financial performance, and schedule performance) of the TTperformance construct were adopted from previous studies.

4. Research methods

4.1 Data Collection

Data collection for this study was undertaken with Libyan petroleum professionals in thefourth quarter of 2008. The target group of respondents technicals, engineers,supervisors, managers of departments of the Libyan petroleum industry and haverelationship to TT process in their companies. This study only solicited the perceptions oflocal petroleum professionals (Libyans) since TT initiatives are ultimately undertakenfor the purpose of improving knowledge levels and enhancing the industry capacity oflocal participants. Accordingly, individuals from host nation were considered the bestrespondents to evaluate the importance and effectiveness of variables pertaining to the TTprocess and the outcomes it can potentially generate. As expected, it was difficult to

determine the adequate number of sample participants for this study. However, thisprocess made use of available statistics on the NOC and statistics of the Libyanpetroleum companies available [69]. Furthermore, counseling with professionalacademics in the relevant statistical research. The total petroleum employees in Libya,according to NOC statistics are 45000 employees. The approximate number of employeesin TT related past and present petroleum processes were about 5000 employees. In thetotal number of 300 were distributed a questionnaire were received 205 completed surveyquestionnaire responses, a rate of 68%. Statistical methods used to analyze the datashown below. The statistical package (SPSS ver. 11) was utilized to this research due toits accuracy and effectiveness, which is suitable for quantitative analysis. The surveyquestionnaire survey for the study compiled mainly of three parts and included 63

questions in total, 26 questions were used for analysis, 26 were utilized for descriptiveanalysis, and 11 questions for background of the respondents. To confirm that data wasobtained from reliable source; the background section contained questions about theparticipant years of work experience, position held, and education degree, and sex,number of TT projects involved with. The definition of each part and factor of thequestionnaire survey was offered. In addition, a translated copy of the questionnaire intoArabic language has been prepared in order for participants to understand thequestionnaire correctly. The questionnaire survey was prepared based on two keyindicators. First, by picking out the answer, the participant would provide his opinion onthe effectiveness and success of a factor in the TT process. Subsequently, the participantswould rate the impact of this factor on TT process. The choices of the questionnaire

answer was designed on a five-point Likert scale ranges from "strongly agree" to"strongly disagree" at the first part, while impact assessment of the answer rages fromstrongly positive" to strongly negative. Moreover, they enabled causal links betweenvariables to be established. In summary, the data set obtained was utilized to ensure thatvariables were perceived to be sufficiently important to be considered as essential TTenabler and outcome variables, examined the successfulness effectiveness of such TTvariables, and was utilized for evaluation purpose, formulating TT constructs anddetermining causal paths. Statistical techniques including exploratory factor analysis


21/39

(EFA), confirmatory factor analysis (CFA), and structural equation modeling (SEM)were implemented for this purpose. A complete description of each of these stagesprovided.

4.2 Data Screening

Data screening is a vital precaution before proceeding with data analysis to ensure thedata accurately reflect the responses made by participants of the study. To check if someof the data is missing and if there is a pattern to the missing data. In addition, to look forextreme responses present in data set that may distort the understanding under study.Moreover, to ensure that multivariate statistical assumptions are met, and what to do ifviolations are there. Data cleaning were performed used (SPSS ver. 11) based on visualinspection of the box plot, distribution diagnosis, frequency tables, histograms, bargraphs, scatter matrices, and outliers cases. Additionally, data screening applied to detectmultivariate outliers and validate multivariate assumptions (normality, linearity,homoscedasticity). Data screening of data set indicated four unusual cases attributes codeviolation that was eliminated after inspecting each case. On the other hand, two variablesviolated the multivariate assumptions (experience, culture) were removed from the data

set due to extreme scores that resulted in a measure of central tendency that does notreally represent the majority of the scores. Additionally, analysis of variance (ANOVA)was performed to ensure that respondents having different positions (e.g. supervisor,superintendent, engineer, etc.) and from different specializations of petroleum companies(production, exploration, etc.) could be considered as a single sample. ANOVAconfirmed that correspondent between position types at the 0.05 level of significance. Acareful examination of data, variance was not widespread and only resulted in twocombinations the data was treated as one useable sample. After critically examining thequality of data collected to prepare it to data analysis, resulted in retaining 201 cases afterdeleting 4 cases and 24 questions after the removal of two violating variables.

5. Descriptive Statistics

5.1 Respondent Profile

The acknowledged number of valid respondents involved in the questionnaire survey was201 respondents. Determining the experience of process participants was critical forensuring the validity of results. The greater the experience of the respondent in thepetroleum industry means greater understanding of process outcomes and influences. Thehighest frequency of respondents had 11-15 years experience as shown in (Fig. 2). Thisgroup accounted for almost 39.3% of the 201 respondents; the lowest group was the 0-5years of experience with 1.5%. However, there was a fairly well distributed frequency of

respondents in each category of experience. This spread of respondent years ofexperience should provide a balanced view on how the TT process was perceived by theactual Libyan petroleum industry.

INSERT > [Figure 2: Respondents petroleum experience]

Respondents were requested to detail the number of processes they have been involvedwith where TT was incorporated (Fig. 3). Almost 65% of the questionnaire participants


22/39

have been involved with at least three past processes involving TT. 1.5 % of the samplehas been involved in more than seven TT processes. These provide a good basis forevaluating the importance and success of individual TT processes and outcome variables.Advantageously, there was relatively low frequency of respondents that had beeninvolved with only two past TT processes (5.5%). As a result, this may not bound theirunderstanding of the TT process; they may still have a good understanding of processsuccess. As expected, a very few (about 0.5%) participated in more than nine TTprocesses.INSERT > [Figure 3: Respondents previous involvement in TT processes]

Table 1 details a break down of respondent experience verse the number of TT processesinvolvements. Obviously, the analysis determined that the experience of respondents andthe amount of past TT processes they were involved in were to some extent related.

Table 1: Experience and number of TT previous processes

Experience(Years)

No of TT processes

1-2 3-4 5-6 7-8 9 +

0-5 2 0 1 0 0

6-10 4 36 1 0 0

11-15 1 56 20 1 1

16-20 2 31 25 1 0

20 + 2 6 10 1 0

Most participants with experience on only (1, 2) TT processes have between 6-10 yearsexperience. Respondents with experience on (3, 4) TT processes generally had about (11,15) years of experience. Obviously, employees with more than 10 years of experiencewould involve in more than (3) TT processes. Very few respondents had more than (15)years experience in addition had been involved with more than (7) TT processes. Theserespondents are likely to have an excellent understanding of process outcomes andinfluences, consequently, their opinions are of great value to this study.The position held by respondents in their respective organization was detailed in Fig. 4. Itis essential to evaluate the position held by the respondents not only to give credibility tothe results but also to understand the perspective from which the survey questions havebeen interpreted. This will prove to be valuable when examining the factor analysisresults and should make grouping of factors and rating of factor importance easier. Therewere 33.7% of respondents for these studies were supervisors, followed by engineerswith 30.2%. These respondents will have an informed perspective of all daily operations.Thus, they would be able to critically evaluate all process issues especially thoseconcerning the enablers like teamwork, understanding, and communication. There were

also relatively moderate proportions of project engineers 21%. Other positions likeproject manager, superintendent, technicians, and others accounted for about 16% percent of the respondents each (approximately 31 employee).

INSERT > [Figure 4: Positions of respondents.]

Evaluating education levels was necessary to demonstrate that the respondents weresufficiently educated to develop a professional opinion about the petroleum industry.


23/39

Obviously, the petroleum industry in Libya was dominated by male gender (100%). Asshown in Table 2, if for example, there were a high frequency of respondents with only ahigh school education, the results obtained may carry questionable authority and may notbe accommodating as an accurate and valid interpretation of the TT process.

Table 2: Respondents personal characteristics

Item Personal characteristics No. of respondents Percentage

GenderMale 201 100

Female 0 0

Age

>30 3 1.5

31-40 88 43.8

41-50 103 51.2

51 + 7 3.5

Education

Diploma 23 11.4

Bachelor Degree 148 73.6

Master Degree 27 13.4Doctorate Degree 2 1.0

Others 1 0.5

Fortunately, the highest frequency of respondents had a bachelor degree (73.6%). Mastersdegree qualified respondents were also quite common (13.4%), followed by diplomaqualifications (11.4%) and doctoral degree (1.0%). Unexpectedly, very few respondentshad less than high diploma in the petroleum field (0.5%).

4.2 TT project profile

Survey participants were requested to detail the number of processes they have beeninvolved with where TT was incorporated. More than 80% of the questionnaireparticipants have been involved with at least three past projects involving TT. Less than5% of the sample has been involved in just one TT project. This gave them a good basisfor evaluating the importance and success of individual TT process and outcomevariables. Understandably, very few respondents (about 2%) participated in more thaneight TT projects, since planned TT was a relatively new concept in the Libyanpetroleum industry. Respondents were requested to provide a range of information on thelast three petroleum processes they have been involved in where TT from a foreignpartner was integrated. In total, the respondents provided detailed information on 302processes performed. The information collected for the processes included year

completed, process description, foreign nationality, skills transferred, mode of transfer,and a general rating on the success of the process. A descriptive summary for each ofthese items was provided. The primary mode of transfer for processes in the petroleumindustry where TT was implemented was management contracting (38.7%), turnkey(26.8%); closely followed by joint venture (25.2%), followed by other modes of transfer(9.3%). This drift confirms the remarks made by Hill [70] that turn key projects arecommon in petrochemical plants and oil refineries. Also, this will reinforce other studiesreported in the literature [1].


24/39

Additionally, the nationality of the technology transferor (i.e. foreigner) for each of the302 listed processes was requested. United Kingdom (UK) were involved in the highestnumbers of processes (35.1%), followed by Germany (16.2%), Italy (15.6%), Austria(5%), and other countries (23.5%), Other transferor nationalities included Canadian,French, Spanish, Korean, and Irish, to name a few (Table 3).

Table 3: TT transfer mode and nationality

NationalityMode of transfer

CountJoint venture Turnkey Managementagreement

Othermodes

1. United Kingdom 25 25 46 10 1062. Germany 7 19 19 4 493. Italy 20 10 14 3 474. Austria 4 6 4 1 155. others 20 21 34 10 85

Total 76 81 117 28 302

This result support the information published recently by Otman and Karlberg [71] that

the United Kingdom is by far the largest source of foreign direct investments in Libya.Not surprisingly, Germany and Italy, two countries which have played a major role inLibyas petroleum industry during the sanction period [72].Table 4 shows that there were four groupings of skills transferred during the TT process.The skills included management, technical, new technology and other. Technical skillswere predominately transferred (44%), followed by technical and new technology skills,both being transferred on 27.8% of the processes.Skill types other than those previously mentioned (e.g. research and development) weretransferred on only 2% of the processes. Most processes examined transferred more thanthree types of skill.


Nationality

Skill Transferred

Count

1 2 3 Skills

Management TechnicalSkill

NewTechnology

1+2 2+3 1+2+3

UK 5 50 11 6 0 34 106Germany 1 21 12 2 3 10 49Italy 2 14 6 8 1 16 47Austria 0 7 0 2 0 6 15Others 10 31 9 5 2 24 85Total 18 133 38 23 6 84 302

To realize a general how successful the TT process on individual processes, therespondents were requested to rate each process on scale from very low to very high. Atotal of (270) process represents the majority of processes were rated as having eithermoderate or high success. However, only (2.65%) indicated a low or very low successrating combined (Table 5).


25/39


Rating

NationalityTotal Percentage

UK Germany Italy Austria Others

Very Low 0 0 1 0 2 3 1%

Low 1 4 0 0 1 5 1.65%

Moderate 15 6 6 2 17 46 15.23%

High 80 37 35 13 59 224 74.17%

Very High 10 2 5 0 7 24 7.95%

Excusably, it was difficult to indicate particular country transferred technology betterthan other country given that the sample size for each country is unequal.Notwithstanding, evaluating the mean of the rating provided insight to the TT process(Fig. 5), the United Kingdom received the highest score of (3.93) and Italy followed at(3.92) while others somewhat received lower rates.

INSERT > [Figure 5: Nationality success rate]

Respondents provided a list of process completion dates which grouped into 5-yearcategories from 1990 until 2009. Obviously, from Table 6, the years period (20002004), have a high number of technologies related processes were performed in Libyanpetroleum industry. This highlights that most petroleum technology in Libyan industrywas transferred on the large-scale recently due to the full lifting of sanctions, along withpossible changes to Libya's 1955 hydrocarbons legislation.

Table 6: TT process description and year completed

YearTechnicalprocess

TrainingAdvancedtechnology

Petroleumproject

Studyabroad

Total

1990-1994 2 4 0 7 0 131995-1999 21 11 7 14 5 58

2000-2004 69 39 28 52 2 190

2005-2009 12 10 5 11 3 41

Total 104 64 40 84 10 302

Sanctions had caused delays in a number of field development and Enhanced OilRecovery (EOR) projects, and had deterred foreign capital investment to a significantextent [73]. This study found that petroleum technology was rarely transferred during thesanction years (1990-2000) Fig. 6. The reason for this low performance was that manyinternational companies having the advanced petroleum technologies, many of which

they owned under patent were not allowed to participate in Libyan petroleum industry.

INSERT > [Figure6: TT process description and year completed]

However, Libya is seeking foreign company help to increase the country's oil productioncapacity from 1.60 million bbl/d at present to 2 million bbl/d by 2008-2010, and to 3million bbl/d by 2015[73]. In order to achieve this goal, and to upgrade its oilinfrastructure in general, Libya is seeking as much as $30 billion in foreign investment


26/39

over that period. The last period years (2005-2009) showed lower than expected becausedata for years 2008 and 2009 were not available.The main study provided some indications that United Kingdom has been the mostcommon transferor of technology in Libyan petroleum processes over the past 20 years(Fig. 7). Italy and the Germany also have a strong representation in the field of TT inLibyas petroleum industry. Austria has had a relatively low TT involvement over the

past 20 years. The Other nationality group included Canada, Korea, and Spain.

INSERT > [Figure 7: TT processes count and nationality]

Respondents were asked whether they knew if TT was planned prior to the beginning ofthe process (Table 7). This will help to develop a better understanding of the respondent sperception of the TT process and how completion planned could improve skills.

Table 7: TT process planning count and year completed

Year Yes No Dont know Total

1990-1994 13 0 0 13

1995-1999 54 1 3 582000-2004 180 5 5 190

2005-2009 37 3 1 41

Percentage 94.04% 2.98% 2.98% 302

For the 302 listed projects where TT was integrated almost 6% of the respondentsacknowledged that TT was either not planned or that do not know. Thus, the majority ofrespondents knew whether TT was proactively planned prior to the project executionphase over the past 20 years. During the 1990-1994 period, it had been found thattechnology had been slowly transferred into petroleum industry in Libya. In the 1995-1999 period technology had been transferred at almost 17.9%, meanwhile the percentage

in the years 2000-2004 period improved to 59.6% of the total processes surveyed. Thus,it appears that Libya is not only undertaking a massive upgrade its petroleuminfrastructure in general in recent years but also utilizing what was considered a highlyattractive petroleum state due to its low cost of oil recovery. Consequently, TT programswere utilized to improve the skillfulness of petroleum employees.

6. Data analysis and results

6.1 Rating TT variables

As previously described, the questionnaire respondent rated the impact for the retained 24

items on a 5-point Likert scale for the success or effectiveness of TT process. Table 8details the mean and standard deviation value for each variable in the conceptual model.The significant outcomes of this analysis are summarized below:

TT environment mean (4.05) was considered the most important TT processenabler followed by learning capability (4.04). However, TT infrastructure (3.92)and the TT support construct (3.90) were not considered as important as the otherswere. Possibly, due to respondents limited understanding on the impact of macro

factors on the TT process.


27/39

Industry knowledge (4.14) was considered the most important variable.Surprisingly, involving sub-contractors in the TT process (3.82) and theconstruction mode of transfer (3.86) were considered the least important enablingvariables.

Most of the TT outcome variables were deemed highly and equally important.Respondents perceived that the implementation of TT programs was essential forimproving schedule performance (4.11). Since schedule performance indeveloping countries is generally perceived to be quite low this result is notsurprising. However, the importance of TT for improving financial performance(3.93) and competitiveness (3.92) of host companies was notably lower than theother outcome variables.

Table 8: TT process planning count and year completed

Code Description Mean Std. Deviation

Enablers A TT support 3.90 0.622

A2.1 Laws and regulations 3.87 0.688

A2.2 Government TT plan 3.99 0.845A2.3 NOC strategy 3.85 0.807A2.4 NOC reward system 3.92 0.865 B TT infrastructure 3.92 0.600

B2.1 Information technology 3.91 0.729B2.2 Research and development 4.03 0.854B2.3 Sub-contractors 3.82 0.799B2.4 Training 3.93 0.761B2.5 Standards and quality 3.86 0.913B2.6 Management practice 3.98 0.869C TT environment 4.05 0.645

C2.2 Knowledge base 4.14 0.796C2.3 Technology complexity 3.96 0.896

C2.4 Communications 4.05 0.756C2.5 Teamwork 4.05 0.870 D TT learning capability 4.04 0.669

D2.2 Attitude 4.05 0.882D2.3 Capability 4.00 0.815D2.4 Exposure 4.09 0.798D2.5 Supervision 4.02 0.827TT outcome

Economic performance 4.01 0.693

A4.1 Competitiveness 3.92 0.760A4.2 Performance 4.09 0.834

Knowledge performance 4.01 0.721

B4.2 Working practice 4.00 0.809B4.3 Skill base 4.01 0.803

Project performance 4.02 0.657C4.1 Financial performance 3.93 0.803C4.2 Schedule performance 4.11 0.747

The variables within all constructs were considered important (i.e. >3), therefore, 24variables of impact perspective was used for the initial factor analysis computation.


28/39

6.2 Exploratory Factor Analysis

EFA using principal component analysis (PCA), with varimax rotation, was conducted tocondense the information contained in the original 24 variables into a smaller set offactors with a minimum loss of information [74, 75]. Specifically, the aim was to searchfor and define the fundamental constructs assumed to underlie the original variables. Theratio of data sample observation to variable (8.3:1) was deemed adequate for factoranalysis, exceeding recommended by Hair [74]. Moreover, the value for the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy was 0.95, exceeding therecommended threshold level of 0.5 [76]. EFA retained a 21 variable solution, removingthree variables A2.4 (reward system), B2.6 (management), and D2.4 (exposure). Two ofthe removed variables (i.e. management and reward system) had very high loadingswithin their own individual constructs. These factors could be argued as being essentialenablers in the TT process; however, they were removed because there were factors,which consisted of only one generic variable. Five factors best represented the data interms of variance explained (64%) and grouping of variables. These factors includedgovernment support; host infrastructure; technology learning capability; hostcharacteristics; and TT performance. Table 9 details the factor loading, explained

variance, Eigen values, and Cronbachs alpha for the five-factor solution. All factorloadings exceeded the 0.5 threshold level with loadings ranging from 0.733 to 0.859.Additionally, Cronbachs alpha results ranged from 0.625 to 0.883 indicating that the

scale used was reliable [77, 78]. The results underline that the technology learningcapability factor is the key enabler of the TT process explaining almost half (46.65%) ofthe total variance in the data set (64%). The combined explained variance for the TTprocess enablers (i.e. government support, host characteristics, technology learningcapability and local characteristics) equates to more than two-thirds (58.7%) of the totalvariance (64%). unquestionably, these factors need to be carefully managed to ensure thatthe TT process derives the most value for the host country.

Table 9: Varimax rotated factor loading for the four-factor solution

Factor Items (identifying Questions) v (63.937%) explained Loading

1. Government supportVariance=4.675%Eigen Value= 4.675Cronbachs Alpha=0.720

Government TT plan 0.814NOC strategy 0.807Standards and quality 0.780

2. Host InfrastructureVariance=3.850%Eigen Value= 0.924Cronbachs Alpha=0.742

Communications 0.843Research and development 0.814Sub-contractors 0.779

3. Technology LearningCapabilityVariance=46.65%Eigen Value= 11.2Cronbachs Alpha=0.883

Supervision 0.806Attitude 0.797Teamwork 0.768Capability 0.759Training 0.755Technology complexity 0.741Knowledge base 0.738

4.Local Characteristics


29/39

Variance=3.476%Eigen Value= 0.834Cronbachs Alpha=0.625

Information technology 0.859Laws and regulations

0.859

5. TT PerformanceVariance=5.291%Eigen Value= 0.1.270Cronbachs Alpha=0.870

Overall performance 0.826Industry knowledge 0.818Skill base 0.799Schedule performance 0.760

Competitiveness 0.736Financial performance 0.733

Factors 1 and 4 are related to the TT support required for TT, and factor 4 contained onlytwo variables. These two factors would be better represented as one broader enablingfactor, as originally perceived in the conceptual model (Fig. 1). To substantiate thismodel, factor analysis was performed on these combined two factors (5 variables) asshown in Table 10.

Table 10: Varimax rotated factor loading for the single outcome factor solution

Factor Items (identifying Questions) Loading1. Government support

Variance=53.341%Eigen Value= 2.67Cronbachs Alpha=0.778

NOC strategy 0.758Standards and quality 0.739Laws and regulations 0.731Government TT plan 0.721Information technology 0.701

As suspected, only one factor resulted, explaining (55.064%) of the variance and allloadings exceeded 0.7 as shown in Table 11. Thus, factor analysis produced a TT modelconsisting of four factors, including, three enablers: Government Support (GS), HostCharacteristics (HC), Technology Learning Capability (TLC), and one outcome factor:

TT Performance (TTP). The following stage of analysis utilizes SEM to confirm themodel structure and causal paths between factors.

6.3 Structural Equation Modeling (SEM)

SEM using AMOS 4 software was performed to test the study model and inter-relationships between factors. SEM is an effective technique for conceptualizing atheoretical model, confirming relationships between variables and gaining insight into thecausal nature and strength of identified relationships [79]. However, SEM is a structuraltechnique requiring large samples with a minimum sample size of 15 cases per measuredvariable. Since factor analysis reduced the number of variables to four factors, a

satisfactory ratio of 50:1 cases per measured variable was achieved [80]. Moreover, thedeveloped model needs to satisfy conditions for a number offit indices. A rule-of-the-thumb is that Comparative Fit Index (CFI) and other incremental indexes with valuesgreater than 0.90 may indicate reasonably good fit of the studys model [81]; Goodnessof Fit Index (GFI) is an absolute index and requires values to be above 0.90 as well; TLIis the Tucker-Lewis coefficient, also called the Bentler-Bonett non-normed fit index(NNFI). TLI is not guaranteed to vary from 0 to 1, however, TLI close to 1 indicates agood fit; and the Root Mean Square of Approximation (RMSEA) is a lack of fit index


30/39

where a value of zero indicates the best fit and higher values indicate a worse fit. The Arule of thumb is that values less than 0.05 indicate a close approximate fit, valuesbetween 0.05 and 0.08 indicate reasonable error of approximation, and values greaterthan 0.10 indicate a poor fit [82]. For the purpose of this study, SEM was employed forthe following two main tasks: confirmatory factor analysis (CFA) was undertaken tocorroborate the four constructs established through exploratory factor analysis (i.e. testingthe measurement model); and determining significant causal paths between factors.

6.4 Measurement Model

Confirmatory Factor Analysis (CFA) was undertaken to substantiate results determinedthrough EFA (Fig. 8).The measurement model for these four constructs (latent variables)had acceptable goodness-of-fit indices: RMSEA= 0.042, GFI = 0.90 ,CFI = 0.97, TLI =0.96, [79, 83]. Moreover, the results shown in Table 11 indicated that the items used foreach factor were representative of that factor (regression weights >0.6, significant t-valueat the 0.05 level). In summary, CFA confirmed the EFA analysis and did not suggest theremoval of any variable, as a result the same constructs were utilized for path analysis inthe following section.

Table 11: Measurement model results

Factor variable description Standardized regression weights t - value R2

Technology learning capability

Training 0.71 * 0.50

Knowledge base 0.70 9.45 0.49

Technology complexity 0.68 9.19 0.47

Teamwork 0.73 9.84 0.54

Attitude 0.75 10.13 0.57

Capability 0.70 9.44 0.49

Supervision 0.76 10.23 0.58Government support

Laws and regulations 0.61 * 0.37

Government TT plan 0.68 7.41 0.47

NOC strategy 0.68 7.75 0.47

Standards and quality 0.65 7.45 0.42

Information technology 0.60 7.00 0.35

Host characteristics

Sub contractor 0.69 8.72 0.47

Research and developments 0.71 * 0.50

Communications 0.71 8.99 0.50

TT performance

Competitiveness 0.64 * 0.41Overall performance 0.78 9.09 0.61

Industry knowledge 0.78 9.07 0.61

Skill base 0.77 9.02 0.59

Financial performance 0.66 9.25 0.44

Schedule performance 0.69 8.26 0.48

*, fixed for estimation


31/39

INSERT > [Figure 8: Confirmatory factor analysis (CFA) design]

6.5 The Path Analysis of the Model

Paths analysis was undertaken using the SEM technique to uncover the significantinterrelationships between the factors retained from EFA and CFA. From the analysis itwas determined that government support was the only exogenous factor in the model.The remaining enablers were considered endogenous factors (Fig. 9). In order to access,what is called a saturated model in the sense of every variable was hypothesized to berelated to every other variable. The model allows examining the direct effect ofgovernment support and hosting characteristics on TT performance. Nevertheless, it alsoallows examine some of the indirect effects as well. Not only are government support andhost characteristics said to directly affect TT performance, they are also hypnotized toexert an effect through the technology learning capability variable. Technology learningcapability takes a mediator role in this model, and it could be said that some of the causalinfluence of government support and host characteristics is mediated through technologylearning capability. Thus, government support and host characteristics are said to

influence TT performance in two ways: first by exerting a direct effect on TTperformance and secondly by exerting an indirect effect on TT performance by affectingtechnology-learning capability.

INSERT > [Figure 9: Path model for TT in petroleum industry]

The model considered recursive. Using the model-fitting method with AMOS, this allowsto quickly reviewing the results of the initial model. The standardized regression weightswere assigned to the appropriate paths from each analysis. The path coefficients showedthat government support and host characteristics are both significant predictors oftechnology learning capability. Because they exceed the (0.3) criterion, both would be

treated as having achieved practical significance as well. However, the path coefficientsleading from host characteristics to TT performance (0.09) is insignificant indicating thatthe host characteristics variable has indirect effect accomplished through the mediatorvariable of technology learning capability. Only two of the three predictors of TTperformance yielded significant coefficients. The average of the residuals between theobserved correlation/covariance (RMSEA) from the sample and the expected modelestimated from the population. Loehlin [84] proposes the following classifications: (1)less than 0.08 indicate good fit, (2) 0.08 to 0.1 indicates a moderate fit, (3) greater than0.1 indicates poor fit. The RMSEA in this model is 0.65 indicating a poor fit.

6.6 Respecifying the path model

In the model, one of the direct effects in the original did not statistically materialize. Pathcoefficients from host characteristics failed to achieve a statistical significance there forthe path were dropped from the re specified model shown in Fig.10. Moreover,discriminate validity analysis did not uncover any correlated endogenous perspective [85,86]. Additionally, scatter plots between the four factors were conducted to ensure that alinear trend best represented (i.e. highest R

2fit) their relationship. Fig. 11 presents the

formulated path model for international TT in petroleum industry.


32/39

[Figure 10: Respecified path model for TT in petroleum industry]

This model has the following fit coefficients: CMIN/DF = 1.96 where, the CMIN valueshown in AMOS is equal to the chi-square value divided by the degrees of freedom (DF),RMSEA = 0.069, GFI = 0.99, AGFI = 0.95, NFI = 0.99, CFI = 0.99, TLI = 0.98. In total,four structural equations explained the five causal relationships (paths) which existbetween the four retained TT enabling and outcome factors. A summary of the developedstructural equations, path coefficients, and significance levels was provided in Table 12.The following section discusses the practical implications of each structural equation andits associated predictor variables.

Table 12: Standardized path coefficients and structural equations

Paths Standardized equation Coefficients t R2

GS HC ZHC=0.86*(ZGS) = 0.68 6.75*** 0.46

GS TLC ZTLC=0.40*(ZGS)+0.45*(ZHC) = 0.40 6.75*** 0.61

GS TTP ZTTP=0.46*(ZGS) = 0.46 7.43***

HC TLC ZTLC=0.45*(ZHC) = 0.45 7.49***TLC TTP ZTTP=0.40*(ZTLC)+ 0.46*(ZGS) = 0.40 6.55*** 0.63

*P


33/39

promote solid relationships by encouraging only those foreign companies thatown satisfying TT reputation.

Technology transfer performance (ZTTP): The results confirm that governmentsupport is essential for achieving outcomes from the TT process. Governments inseveral developing countries are currently encouraging TT initiatives in anattempt to improve their industries, living standards, and economic prospects.This objective could only be achieved if host employees and professionalsperform at a higher level and become more competitive locally within petroleumproducing countries and international petroleum market, eventually becoming acompetitor of the foreign companies. Host and foreign companies with idealisticcharacteristics for TT need to be carefully selected to ensure that the host nationhas the best chance for receiving the most tacit and implicit knowledge from theprocess. Moreover, mutual trust developed through effective communication andunderstanding between the transferee and transferor will greatly enhance the hostfirms knowledge advancement, working practices and overall performance over

the long term. It should be noted that the developed structural equations might notbe as accurate when applied in another context (i.e. path coefficients may changewhen modeling TT initiatives in another countries petroleum industry). Whilst the

causal relationships should hold true in these different settings, the strength of therelationship will depend on the maturity of the host nation and its petroleumindustry, in relation to the development scale. Nonetheless, the identified pathequations can be used as a tool by governments and petroleum companies fromdeveloping and newly oil producing countries to monitor the TT process and itsgenerated outcomes for the host sector. Additionally, the core constructs orfactors of the model were generically named for easy adaptation and utilization inother industry sectors. With some minor modifications to the derived TT enablingand outcome sub factors presented herein, the developed TT model could beapplied to a wide range of industrial settings. The results confirm that appropriatehost characteristics are essential for technology learning capability. Foreign

companies that have experience working with foreigners, a strong knowledgebase and are willing to transfer their knowledge will create robust bonds withlocal workers, which are based on mutual trust, communication andunderstanding.

7. Discussion and Conclusions

As indicated before, two variables with high factor loadings (i.e. management and rewardsystem) were removed through the EFA process along with two variables (culture andexperience) that were removed by data screening. Moreover, CFA did not corroborate theinclusion of variables representing the technology infrastructure and the degree of

exposure to technology itself. Future research should address the limitations of thisresearch and include factors, with a number of specific variables, which focus on theimpact of technology characteristics, technology infrastructure, degree of exposure,management, and reward system on the TT process and its outcomes. First, it could beargued that if the technology being transferred is significantly more advanced than thecurrent working practices of the host employees they may not properly understand howand why it was implemented and thus will be unlikely to embrace it on future technologyacquisitions. Second, the technology infrastructure available for the TT process can


34/39

influence the degree to which TT performs. It could be argued that joint ventures are oneof the better vehicles for achieving higher rates of technology diffusion to the host sectorbecause they typically imply a shared management approach. Third, having compatibleculture may play a role in achieving effective outcomes from the TT process.Considerable cultural differences could potentially have an impact on a transferorswillingness to implement TT initiatives, which will in turn, can create barriers toachieving harmonious relationships. Moreover, culturally blind leadership, where noattention was paid to cultural difference and indigenous approaches may cause conflictsresulting in the disintegration of teamwork. Lastly, imbedding training into the projectschedule could also be considered as a key enabler in the TT process. Implementingtraining sessions into TT agreements should not only foster mutual trust, communicationand information sharing between the transferor and transferee, but will more rapidlyadvance local employees knowledge at the operational, functional and managementlevels. Certainly, TT programs, which were formally planned and managed (i.e. trainingtimes allocated, supervision specified, etc.), are more likely to transfer a greater degree ofknowledge to the host workers.

The implications for petroleum industry in developing countries, government

departments and industries, such as petroleum, manufacturing, and construction aregaining benefits and competitive advantages from the successful implementation of TTinitiatives. Persuading such TT initiatives is the first step to efficiently and effectivelytransforming or re-engineering traditional petroleum business processes, and ultimatelyimproving the productivity of the domestic petroleum industry. However, it is not enoughto only expect that TT will naturally occur. The processes, which underpin TT, should becontinuously evaluated to ensure that knowledge and indigenous workers are seamlesslyabsorbing skills. This research study has implications for the petroleum industry ofdeveloping and newly oil producing countries attempting to develop and promote aneffective TT process in petroleum industry. The derived international TT model could beutilized to assist government officers in developing countries to enhance the evaluation of

TT performance. Specifically, NOC owned petroleum companies managers would beinterested in the significant pathways to achieving value from the TT process.Understanding the dynamics of such pathways will assist them to better structure TTarrangements and concentrate on the most empowering enablers. This study providesevidence that when petroleum industry incorporating TT are established there must becareful selection of both transferee and transferor companies. Companies withappropriate characteristics for TT will form solid bonds that are based on trust,understanding, and communication. Thus, it is essential that a substantial investment isprovided for workshops and other technology learning capability activities to create thesebonds as early as possible in the petroleum industry. In essence, speeding up the TTprocess is the key to rapidly enhancing industry capacity and competitiveness.

The model is especially important for publicly funded petroleum infrastructure where thegovernment is concerned that advanced technologies are being willingly and effectivelytransferred to local petroleum employees and professionals. Moreover, the model couldassist national economic council for planning, which would want to have tools to bettermonitor the performance of the TT process when they set up road maps for the country todevelop the necessary infrastructure for petroleum industry. One of the primaryobjectives of these planning councils is to actively encourage domestic industry indeveloping countries to improve the knowledge levels of their employees as well as


35/39

industry capacity; ultimately leading to improved standards of living for all indigenouspeople. Finally, the authors suggest that government and the petroleum industrycompanies in developing countries should seriously start to invest funds into furtherdeveloping acquired petroleum technology and management knowledge. Such fundingwill ensure that knowledge is perpetually building in domestic companies diminishing thedegree of reliance on foreign firms.

In conclusion, a path model was created to help both researchers and practitioners tounderstand the TT process in petroleum industry. The main emphasis of the developedmodel was to assess TT performance in the petroleum industry. The model provided anevident design on main variables influenced TT issues. The structural model consisted offour factors and five paths, representing the interrelationships between the four enablingand one outcome factor. Positively, EFA and CFA empirically validated that factorsreferring to technology learning capability, technology characteristics, and technologysupport could be incorporated to evaluate the transfer performance. However, variablessuch as degree of exposure, management, and culture were dropped by factor analysisfrom the model and should be investigated in future research. With suitable assumptions,the findings of this study and the developed TT path model have essential implications

for governments, policy makers, and Libyan petroleum companies seeking to enhancerates of transfer. The application of the developed path model to TT in other industries isrecommended.

Acknowledgments

The researchers would like to thank all the respondents who participated in the questionersurvey. The authors also thank the Institute of Advanced Technology (ITMA), UniversityPutra Malaysia for their cooperation and the facilities provided. This support is gratefullyacknowledged.

References

1. Calantone, R., M.T. Lee, and A.C. Gross,Evaluating international technology transfer in acomparative marketing framework. Journal of Global Marketing, 1990. 3(3): p. 23-46.

2. Lin, B.W. and D. Berg,Effects of cultural difference on technology transfer projects: an empiricalstudy of Taiwanese manufacturing companies. International Journal of Project Management, 2001.19(5): p. 287-293.

3. Malik, K.,Aiding the technology manager: a conceptual model for intra-firm technology transfer.Technovation, 2002. 22(7): p. 427-436.

4. Waroonkun, Stewart, and Mohamed,International technology transfer in construction: processperformance evaluation, in CIB W102 Conference on Information and Knowledge Management ina Global Economy: Challenges and Opportunities for Construction Organizations. 2005: Lisbon,Portugal. p. 119-127.

5. Chacko, G.K.,International technology transfer for improved production functions.EngineeringCosts and Production Economics, 1986. 10(1): p. 245-252.

6. Williams and Gibson, Technology Transfer: A communications prospective. 1990, Newbury Park:Sage Publications.

7. Simkoko, E.E.,Managing international construction projects for competence development withinlocal firms. International Journal of Project Management, 1992. 10(1): p. 12-22.

8. Waroonkun, T. and R.A. Stewart,Modeling the international technology transfer process inconstruction projects: evidence from Thailand. The Journal of Technology Transfer, 2008. 33(6):p. 667-687.


36/39

9. Calantone, M.T. Lee, and A.C. Gross.A comparative model of systematic forces on internationaltechnology transfer. in Proceedings of the International Conference on ComparativeManagement. 1988. Taipei.

10. Boddewyn, J.J., Comparative Marketing: The First Twenty-Five Year. Journal of InternationalBusiness Studies, 1981. Spring/Summer(61-74).

11. Kumar, V., U. Kumar, and A. Persaud,Building Technological Capability Through ImportingTechnology: The Case of Indonesian Manufacturing Industry. The Journal of TechnologyTransfer, 1999. 24(1): p. 81-96.

12. Wang, P., T.W. Tong, and C.P. Koh,An integrated model of knowledge transfer from MNC parentto China subsidiary. Journal of World Business, 2004. 39(2): p. 168-182.

13. Al-Mabrouk, K. and J. Soar,Identification of key issues for successful technology transfer in theArab countries: a Delphi study. International Journal of Technology Transfer andCommercialisation, 2009. 8(1): p. 22-50.

14. Kwon, T.H. and R.W. Zmud, Unifying the fragmented models of information systemsimplementation, in Critical issues in information systems research. 1987, John Wiley \&Sons, Inc. p. 227-251.

15. Robinson, R.D., The international transfer of technology : theory, issues, and practiceBallingerseries in business in a global environment 1988, Cambridge, Mass: Ballinger.

16. Nanoka, I. and H. Takeuchi, The Knowledge Creating Company: How Japanese CompaniesCreate the Dynamics of Innovation. 1995: Oxford University Press, USA

17. Hoekman, B., K.E. Maskus, and K. Saggi, Transfer of Technology to Developing Countries:

Unilateral and Multilateral Policy Options. 2004: SSRN.18. Ozighbo, N. Technological Capacity Building in the Nigeria's Oil and Gas Industry. inProceeding of the 19th annual International information Management Association. 2008. SanDiego, CA.

19. Al-Mabrouk, K. and J. Soar,A delphi examination of emerging issues for successful informationtechnology transfer in North Africa a case of Libya. African Journal of Business Management,2009. 3(3): p. 107-114.

20. Nazmun, N. and S. Vesa,IT-enabled International Promotion of Technology Transfer in theEnterprise Resource Planning Space. Informatics and Control, 2000. 9(3): p. 233-251.

21. Nazmun, N., et al., Success factors for information technology supported international technologytransfer: finding expert consensus. Inf. Manage., 2006. 43(5): p. 663-677.

22. UN, U.N.,Industryuniversity linkage with special reference to management, inReport of anExpert Group Meeting. 1974: New York.

23. Sanchez and Tejedor, Universityindustry relationships in peripheral regions: the case of Aragonin Spain. Technovation 1995. 15(10): p. 613625.

24. Blomstrom, M. and F. Sjoholm, Technology Transfer and spillovers: Does local participationwith multinationals matter. European Economic Review, 1999. 43: p. 915-923.

25. Gander, J., University/Industry research linkages and knowledge transfers: a general equilibriumappraoch. Technological Forcasting and social change, 1987. 31: p. 117-130.

26. Coninck, H.d., et al.,International Technology-Oriented Agreements to Address Climate Change,inResources for the future. 2008.

27. Barton, J.H.,Intellectual property and access to clean energy technologies in developingcountries, inInternational Environment House 2. 2007: Geneva, Switzerland.

28. Holsapple, C.W. and K.D. Joshi,An investigation of factors that influence the management ofknowledge in organizations. The Journal of Strategic Information Systems, 2000. 9(2-3): p. 235-261.

29. Kumaraswamy, M.M. and G.B. Shrestha, Targeting technology exchange for faster

organizational and industry development. Building Research and Information, 2002. 30(19): p.183-195.

30. Wei, L.,International technology transfer and development of technological capabilities: Atheoretical framework. Technology in Society, 1995. 17(1): p. 103-120.

31. Grant, R., Toward a Knowledge-Based Theory of the Firm. Strategic Management Journal, 1996.17(Special Issue: Knowledge and the Firm): p. 109-122.

32. Inkpen, A.C.,Learning through joint ventures: a framework of knowledge acquisition.Journal ofManagement Studies, 2000. 37: p. 1019-1043.


37/39

33. Nonaka, I., Knowledge management based on information technology is a mistake, in (in Korean),Maeil Kyungjae Shinmoon (an economic daily published in Seoul). 2001: Seoul.

34. Madeuf, B.,International technology transfers and international technology payments:Definitions, measurement and firms' behaviour. Research Policy, 1984. 13(3): p. 125-140.

35. Wu, F.S., University- industry technology transfer: an empirical study of the industrial firms'organizational practices. 1993, Rensselaer Polytechnic Institute: Troy, New York.

36. Robinson, R.D., The International Communication of Technology: ABook of Readings. 1991, New York: Taylor and Francis.

37. Saunders, M., Strategic Purchasing and Supply Chain Management2nd ed. 1977, London:Financial Times/ Prentice Hall.

38. Carolynn, Akintola, and Eamon,An analysis of success factors and benefits of partnering inconstruction. International Journal of Project Management, 2000. 18(6): p. 423-434.

39. Zeller, C., Project Teams as Means of Restructuring Research and Development in thePharmaceutical Industry. Regional Studies, 2002. 36(3): p. 275-289.

40. Michie, J. and M. Sheehan,Business strategy, human resources, labor market flexibility andcompetitive advantage. International Journal of Human Resource M

Documents

MCM D 09 00747 Reviewed Copy