A Structural Equation Modeling on Factors Affecting Lecturer

Knowledge Sharing in Islamic Universities for Strengthening Islamic

Economy

Wawan Djunaedi

, Yetti Supriyati

, Yuliatri Sastrawijaya

and Bahrul Hayat

Student of Postgraduate School, State University of Jakarta

Faculty of Mathematics and Natural Science, State University of Jakarta

Faculty of Engineering, State University of Jakarta

Faculty of Psychology, Syarif Hidayatullah State Islamic University Jakarta

Keywords: Knowledge Sharing, Research Skills, Islamic University Lecturers, Islamic Economy

Abstract: In the context of developing and strengthening Islamic economy, the practice of knowledge sharing among

Islamic university lecturers is very important. This is because lecturers are knowledge workers. Knowledge

workers are not just people who have basic jobs or routine activities related to intellectual studies, but must

also have innovative-creative power in utilizing, creating, and disseminating knowledge. Of course the

knowledge shared must be based on quality research results. Through the practice of effective knowledge

sharing based on quality research results, there will be a lot of motivation and innovation related to Islamic

economy products and services. Based on the reasons why this study was done, namely to understand what

factors that improve the practice of knowledge sharing among lecturers in Islamic universities. This study

involved 350 lecturers from a number of Islamic universities in Indonesia. Through analysis of structural

equation modelling (SEM), this study examines the proposed theoretical model of the 1st order

Confirmatory Factor Analysis (CFA) causal relationship between a number of variables predicted to

influence the practice of knowledge sharing. This study also examines a number of hypotheses arranged in a

hypothetical model through a series of empirical data analyzes. By identifying overall model fit for the

practice of knowledge sharing, the process of developing and strengthening research-based Islamic

economy is expected to be carried out better.

1 INTRODUCTION

Indonesia has great potential to become the center of

the global Islamic economy as a country with a

Muslim population reaching 85 percent of the total

population. Unfortunately, this potential still cannot

be maximized by Islamic economic actors. This can

be proven from the results of the 2016 National

Financial Literacy and Inclusion Survey conducted

by the Indonesian Financial Services Authority

(OJK), the Indonesian Islamic financial literacy

index at 8.1%. That means that out of every 100

Indonesians, only 8 persons know about the Islamic

financial services industry. This figure is far lower

than the conventional financial literacy index which

is at 29.5% (OJK, 2017).

A similar report also came from The National

Development Planning Agency (BAPPENAS) that

the overall impact of the Islamic economic industry

on the national economy remained small compared

to the conventional financial industry. Even though

the Islamic economic system in Indonesia has been

officially present more than two decades ago. Of

course this is a not-so-pleasant reality. The

landscape of Islamic economic industry in Indonesia

is very different compared to other countries such as

Malaysia which is far more progressive

(BAPPENAS, 2016).

Islamic economy is not only about religious

preferences. Through the aim of Sharia (Maqasid al

Shariah), Islamic economy has the latent power to

play an important role in empowering individuals

and communities. Islamic economy can promote an

entrepreneurial culture and influence people to

invest in a real and sustainable economy, thus

Djunaedi, W., Supriyati, Y., Sastrawijaya, Y. and Hayat, B.

A Structural Equation Modeling on Factors Affecting Lecturer Knowledge Sharing in Islamic Universities for Strengthening Islamic Economy.

DOI: 10.5220/0009507608250832

In Proceedings of the 1st Unimed International Conference on Economics Education and Social Science (UNICEES 2018), pages 825-832

ISBN: 978-989-758-432-9

825

bringing benefits to the wider community and the

Indonesian economy.

Reflection from the description above, it is

urgent to evaluate the Islamic economic

performance in order to be able to empower the

community. It can be ascertained that there is

something wrong regarding the education and

socialization of the Islamic economy, so that the

level of Islamic financial literacy in the Indonesian

community is at a low level. One of the factors

predicted as the cause of the low level of Islamic

economic literacy is the low participation of

universities in the socialization and education of

Islamic economics among the public.

Among important agents who can educate and

socialize Islamic economic products and services are

Islamic university lecturers. Islamic economics

lecturers need to intensively share knowledge

regarding Islamic economic products and services to

the wider community, thereby encouraging the

development and strengthening of the Islamic

economic sector. Based on that problem, this study

was conducted. The aim is to test a proposed

theoretical model regarding factors that affect the

knowledge sharing of lecturers at Islamic

universities to strengthen the Islamic economy.

2 THEORICAL FRAMEWORK

Knowledge sharing is a key concept in knowledge

management (Hung and Chuang, 2009). At least

there are three basic components supporting the

process of forming knowledge management, namely

the knowledge acquisition, knowledge sharing, and

knowledge utilization. If knowledge acquisition is a

process in which knowledge is developed and

formed, knowledge sharing is a process of

disseminating knowledge and at the same time

making it ready for use. (Daud, Salina; Dali, Nuradli

Ridzwan Shah Mohd; Hamid, 2006). That is why the

concept of knowledge sharing is considered very

important in an institution, including Islamic

universities which are one of the institutions that are

very competent to develop and strengthen the

Islamic economy.

As an endogenous variable in this study,

knowledge sharing (KS) among lecturers should be

thought to be significantly influenced by three

exogenous variables, namely the lecturers' research

skills (RS), research self-efficacy (RSE), and

institutional support (IS). The following is briefly

reviewed the theoretical framework related to the

proposed theoretical model presented in this study.

The success of an institution in developing a

particular idea can be predicted from the application

of effective KS among its members. Experts

strongly believe that an institution —including a

university— can generate enormous profits if it

consistently applies KS (Cheng, Ho and Lau, 2009).

KS will only work effectively if lecturers in Islamic

universities have good RS. Lugkana Worasinchai

and Aurilla Aurelie Arntzen Bechina concluded that

strengthening RS is essential for the success of KS

(Worasinchai, Aurelie and Bechina, 2010).

Farkhondeh Hassandoust in his research also stated

that the practice of KS was influenced by RS

(Hassandoust; and Vimala Perumal, 2011).

Unfortunately, the development of the RS of

lecturers in the university itself has received

peripheral attention. According to Emmanuel

Chinamasa, negligence towards the development of

lecturers' RS often occurs due to the assumption that

lecturers have adequate RS (Emmanuel, 2014). Even

though the low of RS can have an impact on the low

level of KS.

As Islamic universities that involve many

knowledge workers, the practice of KS —especially

related to Islamic economics— among lecturers

should have a positive impact. Niels-Ingvar Boer

and K. Kumar have mentioned in his research results

that KS can produce very broad collective outcomes

(Boer and Kumar, 2005). KS is able to provide

benefits not only for lecturers, but also for the wider

community as a target of Islamic economic products

and services.

Chun-Lin and Mei-chi Chen's research stated

that one of the factors that drives someone to

practice KS is because the person has RSE (Lin and

Chen, 2009). Without high RSE, a person will not

have the courage to practice KS. The research of

Hsu et al. prove that one's RSE has a positive effect

on KS behavior. Lin and Hung's survey results also

stated that RSE is one of a number of variables that

positively influence KS behavior (Hsu et al., 2007).

While the results of the Elham Aliakbar survey also

stated that RSE is one of a number of variables that

positively effect KS behaviour (Aliakbar et al.,

2012).

The practice of KS is also predicted to be

influenced by IS. The practice of KS among

lecturers will easily occur when there is support

from the university. They will be stimulated to

transfer knowledge when the university provides

maximum support. This is confirmed by the results

of a study by Nicolette Bakhuisen who concluded

that the most important supporting factor for the

practice of KS was the presence of IS (Bakhuisen,

2012).

Connelly and Kelloway's research on employee

perceptions of KS culture also stated that IS is one

of the factors that can improve the culture of KS

(Connelly and Kevin Kelloway, 2003).

The

UNICEES 2018 - Unimed International Conference on Economics Education and Social Science

826

conclusion is also confirmed by the research

findings of Afshin Mousavi Chalak et al. He

mentioned that IS was very much required so that

the practice of KS could work well (Chalak, Ziaei

and Nafei, 2014). Unfortunately, the results of

Castells' research show that there are phenomena of

bottlenecks on the issue of IS in the campuses

(Musiige, Gordon; Maassen, 2015). Therefore it can

be understood if the level of Indonesian Islamic

financial literacy is very low.

Based on the conceptual and theoretical

framework described above, at least three research

hypotheses can be formulated regarding any

variables that influence the practice of KS among

Islamic university lecturers. The following are three

hypotheses in this study:

H1: RS has a positive direct effect on KS.

H2: RSE has a positive direct effect on KS.

H3: IS has a direct positive effect on KS.

3 RESEARCH METHOD

This study uses inferential statistical analysis

techniques of structural equation modeling (SEM) of

order Confirmatory Factor Analysis (CFA). The

causal relationship between the exogenous and

endogenous variables described in the theoretical

model will be empirically tested. The aim is to prove

whether there is a fit of hypothetical models with

empirical data collected.

Respondents involved in this study were 350

lecturers from several Islamic universities in

Indonesia, consisting of 241 state Islamic university

lecturers and 109 private Islamic university

lecturers. In order to be able to represent lecturers

from state and private universities, the method of

sampling chosen is stratified random sampling.

The data collection technique used was the

Likert scale questionnaire, namely for KS, RSE, and

IS variables. The data related to RS is collected

through multiple choice research objective tests. All

instruments are sent to respondents via email.

SEM techniques of the 1

order CFA type were

chosen because according to Bagozzi and Fornel,

analysis using SEM techniques enabled researchers

to examine complex relationships between variables,

both recursive and non-recursive. Through SEM

analysis, researchers can also distinguish various

kinds of relationship effects, both those that are

direct effect, indirect effect, and total effect (Bagozzi

and Fornell, 1982).

The process of SEM analysis used in this study

follows the five stages of analysis submitted by

Bollen and Long, namely (a) model specification,

(b) model identification, (c) parameter estimation,

(d) model fit, and (e) respecification model (Bollen,

Kenneth A.; Long, 1993). The entire process of

analysis in this study, starting from the normality

test, multicollinearity test, analysis of the

measurement model and structural models are

calculated using the LISREL program.

4 ANALYSIS

As a multivariate analysis technique, SEM requires a

number of fundamental assumptions that must be

met. There are four data sets of research variables

that will be tested for multicollinearity and

normality, namely KS, RS, RSE, and IS variables.

The results of the analysis show that all data sets do

not contain multicollinearity and fulfill multivariate

normality assumptions with details of all p-

value

skewness-kurtosis

<0.05 as follows:

Table 1. Results of Variable Data Multivariate

Normality Test

Variables p-value

skewness-kurtosis

KS 0.061

RS 0.071

RSE 0.078

IS 0.067

The next step is to ensure the four measurement

models of each variable do not contain offending

estimates; do a validity test; check fifteen measures

of Goodness of Fit (GOF); and finally do a

reliability test. Here are the observed variables of

the four measurement models.

The KS measurement model consists of manifes

variables of Desire to Share Knowledge (DtSK),

Academic Community Interaction (ACT),

Information Technology Availability (ITA),

University Award (UA), and Academic Culture

(AC). The RS latent variable consists of observed

variables of Ability to Design Research (AtDR),

Ability to Choose a Method (AtCM), Data

Gathering Ability (DGA), Data Analysis Capability

(DAC), and Ability to Communicate Research

Results (AtCRR ). For RSE latent variables consists

of Intrinsic Motivation (IM), Research Preparation

Conceptualization (RPC), Data Sources Utilization

(DSU), Research Procedures Application (RPA),

Data analysis (DA), and Research Results

Communication (RRC). The IS measurement model

consists of Policy Support (PS), Financial support

(FS), Administrative Support (US), and

Infrastructure Support (InfS).

A Structural Equation Modeling on Factors Affecting Lecturer Knowledge Sharing in Islamic Universities for Strengthening Islamic

Economy

827

The overall measurement models mentioned

above are confirmed its validity and realiability by

examining the t-values and standardized loading

factor (SLF); examining fifteen measures of

Goodness of Fit (GOF); and finally calculating its

construct reliability (CR) and variance extracted

(VE). The aim is to ensure that the estimated

standard loading factors of the measurement models

are good fit. The following is a summary of each

calculation result as referred to above:

Table 2. Validity and Reliability of Research

Variables

Variabl

Validitas Reliabilitas

value

error CR VE

KS 0.91 0.68

DtS

16.76 0.76 0.42

26.32 1.00 0.00

ITA 25.58 0.99 0.03

UA 14.72 0.69 0.53

AC 12.18 0.59 0.65

RS 0.92 0.70

AtD

24.62 0.97 0.06

AtC

21.26 0.89 0.21

14.09 0.67 0.55

13.33 0.64 0.59

AtC

24.27 0.96 0.08

RSE 0.93 0.67

IM 20.51 0.87 0.24

25.36 0.98 0.04

23.61 0.94 0.11

23.28 0.94 0.12

DA 16.33 0.75 0.44

7.45 0.39 0.85

IS 0.94 0.80

PS 23.52 0.95 0.10

FS 21.60 0.90 0.19

AS 17.54 0.79 0.38

InfS 22.41 0.92 0.15

From the summary table above, it can be seen

that each measurement model has a good level of

validity and reliability. It can be proven that every t-

value of all manifest variables has loading factors >

1.96 and all standardized loading factors > 0.70.

Likewise with all construct reliability of the latent

variables are > 0.70 and the variance extracted are >

0.50. Thus, it can be concluded that the four models

of measurement models match the good fit.

To continue on the analysis of the hybrid model

which is a combination of the four measurement

models, it is necessary to evaluate the overall model

fit or GOF statistics between the data and the

hypothetical model. Following are the results of the

tabulation of the overall model fit test of the hybrid

model:

Table 3. The Overall Model Fit Test Result of

Hybrid Model

Fit Index

Acceptable

Threshold

Level

Estimation

Result

Decision

Absolute Fit Measures

Chi-

Square

Low Chi-

Square

value

1388.74

Poor Fit

p p > 0.05 (P = 0.0)

NCP Low NCP

value

1224.74

Poor Fit

Interval Narrow

interval

value

(1109.50 ;

1347.42)

RMSEA RMSEA <

0.08

0.15

Poor Fit

p (close

fit)

p > 0.05 P = 0.00

ECVI

Small value

and close to

saturated

ECVI

M* = 4.24

Good Fit

S* = 1.20

I* = 83.87

RMR

Standardize

d RMR <

0.05

0.044 Good Fit

GFI GFI > 0.90 0.72 Poor Fit

Incremental Fit Measures

NFI NFI > 0.90 0.95 Good Fit

NNFI NNFI > 0.90 0.95 Good Fit

AGFI AGFI > 0.90 0.64 Poor Fit

RFI RFI > 0.90 0.94 Good Fit

IFI IFI > 0.90 0.96 Good Fit

CFI CFI > 0.90 0.96 Good Fit

Parsimonious Fit Measures

AIC

Small value

and close to

saturated

AIC

M* =

1480.74

Good Fit S* = 420.00

I* =

29272.12

UNICEES 2018 - Unimed International Conference on Economics Education and Social Science

828

Fit Index

Acceptable

Threshold

Level

Estimation

Result

Decision

Absolute Fit Measures

CAIC

Small value

and close to

saturated

CAIC

M* =

1704.21

Good Fit

S* =

1440.17

I* =

29369.28

Other GOFI

CN CN > 200 51.28 Poor Fit

Information:

M* = Model S* = Saturated I* =

Independence

From the results of the GOF Statistics test, it is

known that there are 9 of GOF statistics that have a

good fit, those are Expected Cross-Validation Index

(ECVI), Root Mean Square Residuan (RMR),

Normed Fit Index (NFI), Non-Normed Fit Index

(NNFI), Relative Fit Index (RFI), Incremental Fit

Index (IFI), Comparative Fit Index (CFI), Akaike

Information Criterion (AIC), and Consistent Akaike

Information Criterion (CAIC). The rest is 6 of GOF

that indicate poor fit, namely Chi-Square, Non-

Centraly Parameter (NCP), Root Mean Sequare

Error of Approximatipn (RMSEA), Goodness of Fit

Index (GFI), Adjusted Goodness of Fit Index

(AGFI), and Critical N (CN). Thus, it can be

concluded that the level of model fit of the structural

model is good fit. Through the Structural Fit Model

test, the path diagram of t-values is also obtained.

The path diagram shows the t-value of each

correlation between variables as follows:

Figure 1. Path Diagram of T-values of Hybrid Model

The structural model regression coefficients can be

seen from the path diagram of standardized solution

below:

Figure 2. Path Diagram of Standardized Solution of

Hybrid Model

Based on the results of the Structural Fit Model

test above, a number of parameters relating to the

exogenous and endogenous latent variables which

are also the basis for testing a number of hypotheses

can be summarized in the table below:

Table 4. Relationship Parameters between

Exogenous and Endogenous Latent Variables based

on the Research Hypothesis

Hypothesis T-value

Path

Coefficients

H1: RS  KS 8.09 0.52

H2: RSE  KS 8.38 0.73

H3: IS  KS 4.71 0.34

Analysis of the structural model fit also produced

a coefficient of determination (R2) for the KS latent

variables as below:

RS RSE IS  KS = 0.82

Based on the analysis of the level of the

structural model fit above, a number of hypotheses

proposed can be tested as follows:

Hypothesis 1:

It is known that the t-value for the RS  KS

parameter is 8.09. That means that the t-value is

significant between the acceptance region, namely

8.09 > 1.96. In addition, the estimated regression

coefficient between the two variables shows a

positive relationship of 0.52. Thus it can be

concluded that the alternative hypothesis (Ha) is

accepted and the null hypothesis (Ho) is rejected, so

that RS has a positive direct effect on KS.

A Structural Equation Modeling on Factors Affecting Lecturer Knowledge Sharing in Islamic Universities for Strengthening Islamic

Economy

829

Hypothesis 2:

It is known that the t-value for the RSE  KS

parameter is 8.38. With a t-value of 8.38 > 1.96, the

parameter is significant in the acceptance region. In

addition, the estimation results of the regression

coefficient between the two variables show a

positive relationship of 0.273. This means that it can

be concluded that Ha is accepted and Ho is rejected,

so that RSE has a direct positive effect on KS.

Hypothesis 3:

It is known that the t-value for the IS  KS

parameter is 4.71. With a t-value of 4.71 > 1.96, the

parameter is significant in the acceptance area. In

addition, the estimation results of the regression

coefficient between the two variables show a

positive relationship of 0.34. This means that it can

be concluded that Ha is accepted and Ho is rejected,

so IS has a positive direct effect on KS.

5 RESULTS

Analysis of the structural model fit shows that RS

has a positive direct effect on the practice of KS.

With the estimated value of the regression

coefficient parameter of 0.52, it can be interpreted

that the effect of RS on KS is 52%. In addition, the

standard error value of the SIMPLIS program

calculation output is 0.086 or 9%. With a standard

error value that is not too large, it can be interpreted

that the estimated regression coefficient of 52% is an

accurate parameter estimation value. This means that

RS factor can be used to accurately predict the

increase of KS.

With the acceptance of Hypothesis 1,

conclusions delivered by Lugkana Worasinchai and

Aurilla Aurelie Arntzen Bechina above is getting

stronger. It is true that strengthening RS is essential

for the success of KS practices. The results of this

study also confirm the conclusions of Farkhondeh

Hassandoust who stated that the practice of KS is

influenced by RS. The more a person has the high

level of RS, the higher the tendency to practice KS.

Likewise with RSE factor, it have a positive

direct effect on the practice of KS. The estimated

value of the regression coefficient between them is

known to be positive at 0.73. This means that the

effect of RSE on KS practices is 73%. Moreover,

the standard error value from the analysis results is

only 0.069 or 7%. This shows that the results of the

regression parameter estimate of 73% are accurate

values. Thus it can be concluded that the factor of

RSE is a very strong predictor for the improvement

of KS practices.

The acceptance of Hypothesis 2 confirms the

conclusions of Chun-Lin and Mei-chi Chen in the

above theoretical study that RSE is a very strong

factor to encourage someone to practice KS.

Similarly, the results prove the conclusions of Hsu et

al. that one's RSE does have a positive effect on KS

behavior.

Similar to the latent variables of IS, it was

proven to have a positive direct effect on the latent

variables of KS. The estimated value of the

regression coefficient between the two shows a

positive value of 0.34. This means that IS can affect

the practice of KS by 34%. Moreover, the standard

error value of the analysis results is only 0.067 or

7%. This shows that the results of the regression

parameter estimation of 34% are accurate values.

Therefore, it can be concluded that the factor of IS

can be used as a very strong predictor for the

improvement of KS practices.

The acceptance of Hypothesis 3 confirms the

conclusions of Nicolette Bakhuisen who stated that

the most important supporting factor for the practice

of KS is IS. Likewise with the conclusions of

Connelly and Kelloway who stated that IS is a very

significant supporting factor for the culture of KS.

Another conclusion confirmed by Hypothesis 3

reception is the study of Afshin Mousavi Chalak et

al. They stated that IS is highly required so that the

practice of KS can run well in an institution.

The results of other analyzes also show a

coefficient of determination (R2) of 0.82. That

means the three exogenous variables (RS, RSE, and

IS) have a total influence of 82% on endogenous

variables (KS). This means that the practice of KS

among Islamic university lecturers —particularly

related to Islamic economic products and services—

is largely determined by the factors of RS, RSE, and

IS in 82% and other factors only have an influence

of 18%. In other words, when the university focuses

on these three exogenous factors (RS, RSE, and IS),

the predictable success rate for improving the

practice of KS has reached 82%.

6 CONCLUSIONS

The results of testing a number of research

hypotheses on the proposed theoretical model in this

study proved empirically. The hypothetical model

that was previously designed proved to be supported

by empirical data with a number of significant

values. Based on the analysis above, it can be

UNICEES 2018 - Unimed International Conference on Economics Education and Social Science

830

concluded that one of the most effective efforts to

increase the level of literacy of Islamic finance in

Indonesia is to increase the willingness of Islamic

university lecturers to share their knowledge

regarding Islamic financial products and services.

This is because Islamic university lecturers are

people who have the competence to provide

information and education related to Islamic

economic materials. With a high level of KS among

Islamic university lecturers, the Islamic financial

literacy index in the future is not expected to be in a

very low position as the survey was reported by the

Indonesian Financial Services Authority (OJK) in

2016.

Therefore, it is important for Islamic universities

to provide support and commitment to increase the

KS of lecturers. The method is by strengthening the

research skills of the lecturers; improving their

research self-efficacy; and facilitating maximum

institutional support, both in terms of policy,

finance, administration, and infrastructure supports.

By boosting these three factors (RS, RSE, and IS), it

can be predicted that the level of KS of lecturers will

also increase. With the increasing practice of KS

among Islamic university lecturers, a positive impact

on developing and strengthening Islamic economy

will be felt.

Increasing KS will also produce a multiplayer

effect. In addition to information and education

related to the Islamic economy that can run well, KS

among lecturers will indirectly have a positive

impact on the overall Islamic economic industry

towards the national economy. The lecturers in

Islamic universities will also indirectly empower

individuals and communities. Thus, the

entrepreneurial culture and investment in the real

and sustainable sharia economic sector will continue

to grow, so bringing benefits to the wider

community and the Indonesian economy. Thus, the

opportunity for Indonesia to become a strong Islamic

economic player on a global scale can be realized.

REFERENCES

Aliakbar, E. et al. (2012) ‘Determinants of

Knowledge Sharing Behavior’, 2012 International

Conference on Economics, Business and Marketing

Management IPEDR.

Bagozzi, R. P. and Fornell, C. (1982)

‘Theoretical Concepts, Measurement and Mening’,

in A second generation of multivariate analysis.

Measurement and evaluation. doi:

10.1159/000358090.

Bakhuisen, N. (2012) Knowledge Sharing Using

Social Media in the Workplace. Amsterdam

University.

BAPPENAS (2016) ‘Masterplan Arsitektur

Keuangan Syariah Indonesia’. Badan Perencanaan

Pembangunan Nasional.

Boer, N.-I. and Kumar, K. (2005) Knowledge

sharing within organizations: a situated and

relational perspective, Erasmus Research Institute

of Management (ERIM).

Bollen, Kenneth A.; Long, J. S. (ed.) (1993)

Testing Structural Equation Models. Newbury Park:

Sage Publications.

Chalak, M. A., Ziaei, S. and Nafei, R. (2014) ‘A

Survey of Knowledge Sharing Among the Faculty

Members of Iranian Library and Information

Science ( LIS ) Departments’, Library Philosophy

and Practice (e-journal).

Cheng, M., Ho, J. and Lau, P. (2009)

‘Knowledge sharing in academic institutions: A

study of Multimedia University Malaysia’, Journal

of Knowledg Management. doi: ISSN 1479-4411.

Connelly, C. E. and Kevin Kelloway, E. (2003)

‘Predictors of employees’ perceptions of knowledge

sharing cultures’, Leadership & Organization

Development Journal. doi:

10.1108/01437730310485815.

Daud, Salina; Dali, Nuradli Ridzwan Shah

Mohd; Hamid, H. A. (2006) ‘Successful Knowledge

Sharing Among the Staffs of Economics and

Business Management Faculties in Higher

Institutions Education’, in. Proceeding of 12th

National Symposium of Business Management

organized by Universiti Tenaga Nasional 2006, pp.

1–8.

Emmanuel, C. (2014) ‘Evaluation of the

University Research Policy: Implications For

Lecturer Research Output’, International Journal of

Advanced Research in Management and Social

Sciences, 3(6), pp. 90–105.

Hassandoust;, F. and Vimala Perumal (2011)

‘Online Knowledge Sharing in Institutes of Higher

Learning : A Malaysian Perspective’, Journal of

Knowledge Management Practice.

Hsu, M. H. et al. (2007) ‘Knowledge sharing

behavior in virtual communities: The relationship

between trust, self-efficacy, and outcome

expectations’, International Journal of Human

Computer Studies. doi: 10.1016/j.ijhcs.2006.09.003.

Hung, Y. C. and Chuang, Y. H. (2009) ‘Factors

Affecting Knowledge Sharinf Behavior: a Content

Analysis of Empirical Findings’, Myacmeorg.

A Structural Equation Modeling on Factors Affecting Lecturer Knowledge Sharing in Islamic Universities for Strengthening Islamic

Economy

831

Lin, C. and Chen, M. (2009) ‘Factors Affecting

Teachers ’ Knowledge Sharing Behaviors and

Motivation : System Functions that Work’, eLAC.

Musiige, Gordon; Maassen, P. (2015) ‘Faculty

Perceptions of The Factors That Influence Research

Productivity at Makerere University’, in Cloete,

Nico ; Maassen Peter ; Baile, T. (ed.) Knowledge

Production and Contradictory Functions in African

Higher Education. Cape Town: African Minds, pp.

109–127.

OJK (2017) ‘Strategi Nasional Literasi

Keuangan Indonesia (Revisit 2017)’, Otoritas Jasa

Keuangan.

Worasinchai, L., Aurelie, A. and Bechina, A.

(2010) ‘The Role of Multinational Corporations (

MNC ’ s ) in Developing R & D in Thailand : the

Knowledge Flow Between MNC ’ s and University’,

Journal of Knowledge Management. doi:

10.1017/S0305000909990444.

UNICEES 2018 - Unimed International Conference on Economics Education and Social Science

832