Evaluation and Program Planning 33 (2010) 487497

Contents lists available at ScienceDirect

Evaluation and Program Planning

journal homepage: www.elsevier.com/locate/evalprogplan

Evaluating a national science and technology program using the human capital
and relational asset perspectives
Chia-Liang Hung a,*, Jerome Chih-Lung Chou b, Hung-Wei Roan a
a
b

Department of Information Management, National Chi Nan University, 54561, Puli, Nantou county, Taiwan
Department of Information Management, Hwa-Hsia Institute of Technology, 23568, Taipei, Taiwan

A R T I C L E I N F O

A B S T R A C T

Article history:
Received 31 August 2009
Received in revised form 28 January 2010
Accepted 31 January 2010

The purpose of this research is to evaluate the performance of the National Science and Technology
Program (NSTP) by targeting the Taiwan National Telecommunication Program (NTP) initiated in 1998.
The Taiwan telecommunications industry has prospered, currently occupying key positions in global
markets even though NTP seldom contributes positively to patent citation performance. Hence, the
authors of this study investigate the qualitative perspective of intellectual capital rather than
quantitative technological indices. The current study focuses on both human capital and relational assets
through surveys of 53 principal investigators of NTP projects and 63 industrial R&D managers of
telecommunications corporations in the Taiwan market. Results show that NSTP member quality and the
ow of employment are good indicators of human capital and that both perform better than the middle
value in the case of Taiwan NTP. In addition, we nd that industrial participants are more likely to share
R&D resources than other academic researchers with higher intention of co-publishing, co-funding, and
sharing equipments and facilities. The industrial NTP participants also have higher expectations
regarding achieving advanced technology breakthroughs in contrast to non-NTP industrial interviewees.
Moreover, industrial participants with greater industryuniversity cooperation intensity indeed obtain a
particular advantage, that is, greater knowledge acquisition from other elds related to the effect of
knowledge spillovers through the particular NSTP linkage. Accordingly, from the perspectives of human
capital and relational assets, the authors conclude by articulating the importance of absorptive capacity
resulting from good human capital and knowledge spillover contributed by relational assets within
governmental technology policy and NSTP programming.
2010 Elsevier Ltd. All rights reserved.

1. Introduction
In the ever-changing 21st century, most industrialized countries allocate resources in pursuit of frontier technologies in order
to retain their economic leadership in the face of global
competition. Taiwan, the Silicon Island, has also invested large
sums in science and technology since 1998, through the initiation
of nine national science and technology programs (henceforth
abbreviated as NSTPs) in the diverse areas of telecommunications,
digital archives, digital learning, hazards mitigation, nanotechnology, biotechnology and pharmaceuticals, agricultural biotechnology, genomic medicine, and system on chip (NSC, 2007a).
Specically, the NT$24 billion (about US$0.73 billion) budget of
the Taiwan National Telecommunication Program (henceforth
abbreviated as NTP) accounts for 22.2% of the total NT$108 billion

* Corresponding author. Tel.: +886 49 2910960x4640; fax: +886 49 2915205.

E-mail address: clhung@ncnu.edu.tw (C.-L. Hung).
0149-7189/$ see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.evalprogplan.2010.01.003

(about US$3.3 billion) in NSTP investments. Fig. 1 shows the NTP

organizational chart, which represents the funding ows mainly
coming from four governmental ministries and one governmentbased telecom corporation; the supervision direction is coordinated by a special program committee. In reality, most research
projects are organized and conducted in the universities and
several national-level R&D institutes. Other NSTP programs were
also orchestrated by similar structures. However, in the case of NTP
outcome, few Taiwan telecom vendors have managed to dominate
the global telecommunications industry for an extended period
outside of Nokia (Finland), Ericsson (Sweden), Qualcomm and
Motorola (USA), DoCoMo (Japan), and Samsung (Korea). Although
the number of patents sponsored by NTP through 19982006
totalled to 925 (NSC, 2007b), only 225 of these patents were closely
related to telecommunications technology (i.e., those of the H04
group by the USPTO classication scheme), implying a mere 27.6%
of the total number of NTP patents. Even worse, the patent impact
of NTP has generally been low. According to the patent impact
identication method of Thomas and Breitzman (2006), hot

488

C.-L. Hung et al. / Evaluation and Program Planning 33 (2010) 487497

Fig. 1. Taiwan NTP organizational chart and fundingsupervision relationship.

patents with high impacts usually have a minimum of 10 recent

citations. There have only been 12 such hot NTP patents, mostly
produced by the Industrial Technology Research Institute (ITRI)
and the Institute for Information Industry (III) of Taiwan, both of
which are Taiwan government-sponsored research organizations.
In fact, over 73% of NTP participants had low-impact patents with
zero citations. However, the real performance of NTP should not be
solely evaluated based on the perspective of number of patents
issued or hot patents, because Taiwan ICT manufacturers continue
to occupy important positions in the global marketplace. For
example, Taiwan is the largest wafer foundry, CD, DVD, and Mask
ROM production center, and second largest in terms of IC design,
WLAN, xDSL CPE, Cable CPE, and routers (NSC, 2007a). Therefore, a
robust measurement of NTP performance should include other
dimensions such as the quality of human resources trained at NTP
and the innovation and entrepreneurship resulting from the
relational assets within the industry-university cooperation
network. The purpose of this paper is to discover the real value
of the NSTPs by constructing a non-patent evaluation framework
associated with the case of NTP.
2. Evaluation framework
Lepori (2006) points out that the national R&D budget for NSTPs
often plays the role of input indicator for future science and
technology performance. Hence, the Organization for Economic
Cooperation Development developed a Frascati Manual that
clearly identies several key R&D sources (OECD, 2002). However,
verifying the expenditure ow, especially the ows to research
personnel who actually embody the output knowledge, is even
more vital than focusing on R&D input (Stewart, 1997). If the
national R&D input ows mostly to education, training, instruction,
and mentoring of researchers, then growth in the capability for
technology innovation and exploitation will be expected. Stewart
(1997) argues that R&D input in research personnel as an
intellectual investment will create the following three types of
future competitive advantage: embodied human capital, internal
structural assets, and external customer assets. Bontis (1998)
further species that the third type of competitive advantage is
both the most sustainable and the most difcult to imitate,
followed by embodied human capital. Therefore, these two types of
competitive advantages will be the focus of our assessment of the
NSTP evaluation framework.

2.1. Perspective of human capital

Lanzi (2007) measured the quality of research human capital
based on the three elements of basic skill, professional capability,
and continuous learning abilities. Wu (2006) also emphasizes that
the real quality of research personnel is often reected in job
salary, job status, and problem solving capability. It is important to
consider the absorptive capability of human resources in order to
be better equipped to acquire knowledge and raise the possibility
of invention in the future (Cohen & Levinthal, 1990). Kundu and
Kumar (2006) claim that special industrial training and development programs actually help employees achieve corporate goals
and personal career objectives as well as improve their productivity. Mann and Robertson (2006) further suggest that training
evaluation programs should not focus only on training satisfaction
and planned behavior but also on the enhancement of employees
learning capabilities and organizational contribution after training.
In addition, the notion of perceived high quality of human
capital also requires the appreciation of demanding employers
(Cozzens, 1997). A correct match that does not waste the talent
pool can only perform to the standards of the actual quality of
human capital. Therefore, Luwel (2005) nds that the ow of talent
is a good indication of knowledge diffusion and industrial
development resulting from national programs. It is important
for most newly industrialized or less-developed countries to focus
their limited national resources on diffusion-oriented technology
policy for expertise training and knowledge diffusion widely
(Chiang, 1990). According to Callon, Laredo, and Mustar (1997),
intermediaries who can transfer, translate, and disseminate
information to bridge the poles of science, technology, and market
are also required throughout in the value exploitation process of
innovation. Further, Wu (2006) emphasizes that the appropriate
ow of talent toward emerging industries can only contribute to
one of the national program objectivesindustrial enhancement
and upgrading.
2.2. Perspective of relational assets
Beyond measuring the quality and ow of human capital from
NSTP, the relational asset that Bontis (1998) species as the most
important advantage in terms of the external customer asset is also
a useful indicator of NSTP research personnel performance. Dalpe
(1993) and Georghiou (1998) both propose that the number of

C.-L. Hung et al. / Evaluation and Program Planning 33 (2010) 487497

Fig. 2. The NSTP evaluation framework.

publishing co-authors, the amount of mutual R&D funding, and the

intensity of sharing equipment and research data are possible
indicators for measuring the potential of relational assets.
Callon et al. (1997) argue that a technology-economic network
model is necessary to meet the emerging challenges when bringing
together scientists from different elds, engineers, marketing and
nancial agents, and end user representatives. According to
Bozeman, Dietz, and Gaughan (2001), scientic and technical
human capital includes both human capital endowments, such as
formal education and training, and social relations and network
ties that bind scientists and the users of science together. In fact,
the latter argument of Bozeman et al. (2001) articulated that the
potential assets value resulting from specic cooperative relationships range from science and technology research to market
research. Consequently, the concept of knowledge value alliance
presents a framework as an alternative focus for thoroughly
understanding and evaluating scientic and technical work
(Rogers & Bozeman, 2001). In fact, industrial participants usually
place greater emphasis on the development of relational assets
through connections to the knowledge alliance network of
government-sponsored technology programs. Sakakibara (1997)

489

argues that the ex ante perceived benets will determine the

degree of involvement within the NSTPs; in contrast, the ex post
participation gains will inuence the possibility of further
realization of relational assets woven among the industry
university cooperation networks.
As a result, the current authors elaborate the NSTP performance
evaluation framework from the perspective of intellectual assets,
namely, human capital and relational assets (Fig. 2). In this
framework, human capital is depicted through two groups of
subindices: one group measuring the quality of research personnel
as judged by industrial employers, and the other relating to the
tness of the talent ow to the optimum position as evaluated by
NSTP project investigators. In addition, the authors divide
relational assets into two groups, one pertaining to the degree
of R&D resource sharing within the NSTP, while the other to the
perceived participation benets associated with NSTP.
3. Research design
3.1. Development of measurement items
According to the proposed NSTP evaluation framework, the
authors developed four constructs of performance measurement
as shown in Table 1. The rst construct measures the quality of
R&D talent. Lanzi (2007) and Kundu and Kumar (2006) both
suggest that a higher level of willingness of the rm to expend
resources on recruitment, salary, and opportunities for promotion
implies a higher quality of employees. Cohen and Levinthal (1990)
and Mann and Robertson (2006) propose that highly talented
individuals reect higher performance and higher capability (or
absorptive capacity) to acquire knowledge and achieve organizational objectives. Wu (2006) argues that highly talented people are
usually assigned to R&D-oriented tasks, management of innovation, or organizational problem solving.

Table 1
Measurement items of the NSTP evaluation framework.
Measurement items
Human capital

Quality of R&D talent

Flow of R&D talents

Relational assets

R&D resources sharing

Perceived participation benets

Literature
Recruitment willingness
Salary level
Promotion opportunity
Performance
Knowledge acquisition capability
Achievement of organization objectives
R&D-oriented tasks
Innovation management
Organizational problem solving
Personnel spillover into domestic industries
Personnel spillover into NSTP-related technological elds
Expertise imported from NSTP
External brokers of expertise

Lanzi (2007)
Kundu and Kumar (2006)
Kundu and Kumar (2006)

Co-publishing the research results

Cooperation for commercialization
Licensing between participants
Technical meetings and conferences
Reorganization for new research agendas
Research co-funding
Experimental data sharing
Sharing equipment and facilities
Researcher training
Breakthrough in a critical technology
Increasing awareness on the importance of R&D
Accelerated development of the technology
Decrease in private R&D budget
Acquisition of knowledge from other participants
Establishment of an ongoing informational network
Commercialization of a product or process
Increase in the number of patent applications
A superior competitive position
Establishment of a standard in a target industry

Dalpe (1993), Georghiou (1998)

Cohen and Levinthal (1990),

Mann and Robertson (2006)
Wu (2006)
Luwel (2005), Chiang (1990)
Cozzens (1997)
Callon et al. (1997)

Callon et al. (1997)

Brown (1997)

Sakakibara (1997)

Rogers and Bozeman (2001),

Bozeman et al. (2001)

490

C.-L. Hung et al. / Evaluation and Program Planning 33 (2010) 487497

The second construct concerns the measurement items of NSTP

R&D talent pertaining to industrial demanders. Luwel (2005)
identies indicators from the supply side of expertise and lays
emphasis on the R&D personnel owing into the domestic
industries and working in NSTP-related technological elds. The
trained expertise matching the demand of industrial development
is often the key purpose of diffusion-based technology policy in the
newly industrialized countries (Chiang, 1990). Contrastingly,
Cozzens (1997) argues that the R&D talent ows standing on
the viewpoint of the demand side, and observes the degree of
demand for external expertise and talent brokers/hunters, along
with the recruitment willingness of industrial demanders, which is
also an important mechanism to facilitate further knowledge
integration for technology commercialization (Callon et al., 1997).
The third construct relates to the measurement items regarding
R&D resource sharing. Georghiou (1998) identies as pertinent the
amount of licensing between NSTP participants, the number of
research meetings and conferences, an ongoing channel for sharing
research experience and innovative ideas, and the possibility of
reorganizing research agendas based on continuous cooperation.
Further, Dalpe (1993) emphasizes the intensity of co-funding, copublishing, and commercializing behaviors, while Brown (1997)
and Callon et al. (1997) focus on the establishment of sharing
mechanisms pertaining to the periodic communique, experimental data, work-in-process output, and equipment and facilities. The
fourth construct concerns the participation incentive in terms of
perceived participation benets, articulated predominantly by
Sakakibara (1997) and supported by Rogers and Bozeman (2001)
and Bozeman et al. (2001) in the knowledge acquisition and
information linkage.
Correct sampling of respondents is necessary for obtaining valid
data that can verify the measurement construct. Therefore, this
study focused on the following two sample sources of respondents:
NSTP project investigators with regard to the construct of R&D
resource sharing and the supply side of R&D talent ow, and
industrial participants involved in the NSTP in terms of measuring
all constructs other than the supply side of R&D talent ow. In this
study, the authors apply the case of Taiwan NTP for further
validation of the measurement constructs.
3.2. Research hypotheses
Apart from measuring the quantity of researchers, Lanzi (2007)
argues that researcher quality should also be considered as an
indicator of NSTP performance. A quality of NSTP manpower
higher than that of non-NSTP manpower may be attributable to
particular NSTP training and development programs. Further, Wu
(2006) elaborates on human quality within the socio-economic
dimension. Talent trained by NSTP will be often given more
challenging tasks. Kundu and Kumar (2006) also argue that
talented people trained by NSTP are usually expected by industrial
employers to have greater capabilities in terms of achieving
corporate objectives. As a result, high performance NSTP researchers may receive higher salaries and more opportunity for
promotion. Hence, the authors articulate the following hypothesis
to reect the human quality of NSTP performance:
H1. From the perspective of human capital, NSTP performance is
positively correlated with higher quality of R&D talent as evaluated
by industrial employers.
According to Cozzens (1997), the knowledge pool contributed
by government-sponsored NSTPs is an important supply channel
for aggregating talented people in technological elds where
demand for them exists. Wu (2006) emphasizes the allocation
efciency of human resources and claims that the t of education,

training, and development of researchers to industrial requirements is critical to NSTP achievements for speeding the creation of
the new industry. In addition, Luwel (2005) observes that
researcher ow into a specic industry implies the intended
diffusion of technological knowledge, which is usually the main
purpose for intervention by the government to drive industry
upgrades and enhancement as supported by governmental NSTPs.
Hence, the authors articulate the following hypothesis to evaluate
the t of the talent ow:
H2. From the perspective of human capital, NSTP performance is
positively correlated with the appropriate ow of talent to the
targeted technology.
Moreover, Dalpe (1993) observed obstacles that hinder universities and industrial corporations in terms of deterring information
exchanges, the former emphasizing academic publishing and the
latter, commercialization and market positioning. If NSTPs establish
cooperative mechanisms that invite greater research communication and resource sharing among participants, the cooperative
relationship will promote the effect of technology spillover and
thereby stimulate emerging businesses. Georghiou (1998) argues
that good cooperative experience between universities and industry
implies a good t of research topics, information, results, and
partnerships. Moreover, such special linkages are often a valuable
relational asset in commercializing frontier technologies or even
accelerating new business ventures through spinoffs, joint ventures,
or strategic alliances. Hence, the authors propose the following
hypothesis to reect the value of NSTP relational assets:
H3. From the perspective of relational assets, NSTP performance is
positively correlated with the higher incentives associated with
R&D resource sharing between NSTP participants.
Sakakibara (1997) argues that the high perceived participation
value of involvement with NSTPs induces industrial players to
accept invitations of NSTP and co-fund industryuniversity
cooperative projects. Moreover, increasingly small gaps between
the ex ante perceived values and ex post actual gains in other
words, higher satisfaction on the part of project participants
extend the relationship and increase the opportunity for future
cooperation. A comparison of the participation before-and-after
gap is a necessary aspect of the questionnaire design. Hence, the
authors articulate two hypotheses, H4 and H5, to reect the value
of relational assets:
H4. From the perspective of relational assets, NSTP performance is
positively correlated with higher perceived benet resulting from
participation in the NSTP.
H5. From the perspective of relational assets, NSTP performance is
positively correlated with higher satisfaction resulting from cooperation with the NSTP.
Hypotheses H1H4 will be veried using factor analysis in
order to ensure measurement validity for the two constructs,
human capital and relational assets. Then, the researchers will use
the t-test approach to compare the mean value of each
measurement item with the middle value of the measurement
scale in order to verify whether NSTPs do pursue a higher level of
performance. The middle value of this survey is the value of four for
the Likert-type measurement scale ranging from one to seven in
the research questionnaire. For hypothesis H5, the authors will use
a paired t-test comparison to verify whether industrial participants
have a before-and-after gap, so as to reect the satisfaction
associated with participating in NSTP and to predict the
industrialization or commercialization potential of future collaboration based on the particular relational linkage.

C.-L. Hung et al. / Evaluation and Program Planning 33 (2010) 487497

491

analysis by their representativeness even though the response rate

was unexpectedly low.

3.3. Data collection

First, the authors designed a questionnaire consisting of all
measurement items associated with the NSTP evaluation framework as Table 1. Then all the questions were pre-tested by the
professors involved in the NTP of Taiwan. The revised version is
shown as Table A-1 in the Appendix. An identical web-based
questionnaire was also developed using Microsoft Access, ASP, and
FrontPage software tools. Next, the authors sent the paper-based
questionnaire as well as the URL of the web-based questionnaire to
231 NTP project investigators and 250 key industrial telecommunications players, including telecom equipment and service
vendors in Taiwan. The respondents were free to complete either
version of the questionnaire; those who completed the paperbased version mailed their responses, while completed online
versions were directly e-mailed to the researchers. The collection
process lasted from May to June 2008. The total number of valid
responses included 63 industrial players and 53 NTP project
investigators from universities or research organizations, with
respective response rates of 26.8% and 23.4%. Most of the 63
industrial players were telecommunication manufacturers (53
respondents), while only 31 industrial respondents had participated in the NTP industryuniversity collaboration. A prole of the
respondents is given in Table 2.
When faced with a low response rate during data collection, the
researchers made several attempts of follow-up tracing and called
for responses of academic and industrial interviewees. However,
they were not obligated to answer the questionnaire. In particular,
some academic professors replied that they were only responsible
to the program funders, National Science Council of Taiwan, and
not to unrelated bystanders or other researchers. They were rather
reluctant to be measured. Similarly, several industrial interviewees
refused to answer for fear of leaking the rms recruitment policy
and the industrial cooperative relationship. However, the authors
further checked the respondent prole and found that most
academic responses were provided by main top-tier universities
such as National Taiwan University, National Chiao-Tung University, National Tsing-Hua University, National Cheng-Kung University, and National Central University. On the other hand, the
industrial respondents ranged from the most important nationallevel R&D organization in Taiwan, namely, Industrial Technology
Research Institute, to top equipment manufacturers of Internet and
Wi-Fi, key mobile phone OEM/ODM vendors, and ve major mobile
network operators, who had been enthusiastic supporters of the
M-Taiwan mobile application program, a sub-program of NTP. In
addition, the surveyed telecom service providers, that is, the
mobile operators and ISP providers, are usually bigger than the
telecommunication manufacturers. The latter have been horizontally disintegrated into multiple specialized vendors from chip
designing, fabrication and packaging to testing, assembling, and
rmware, since the global telecommunication liberalization
(Steinbock, 2002). Thus, the unbalanced size of respondents
between service and manufacturing sector is inevitable. As a
result, the sampling size and the number of service responses are
both comparatively smaller than those of the manufacturing
sector. Hence, the authors judged the collected data for further
Table 2
Prole of respondents.
Classication of respondents

Ratio (%)

Academic researchers vs. industrial players

Manufacturers vs. service providers
NTP vs. non-NTP participants
Awareness vs. unawareness of NTP
Recruiting vs. non-recruiting from NTP
In-sourcing vs. outsourcing expertise

46:54
84:16
49:51
83:17
49:51
48:52

4. Data analysis and discussion

4.1. Validity of measurement
The authors conducted a factor analysis using the principal
component method with Varimax rotation to verify the measurement validity (Sharma, 2005). The SPSS 13.0 statistical package
was used. The KMO (Kaiser-Meyer-Olkin) test demonstrates the
value of 0.525 and 0.752 for human capital and relational assets
respectively, which both show the acceptable level of sampling
data for factoring (Sharma, 2005). In addition, the Bartletts test
also reveals signicantly high correlations greater than zero
(p < .0000) between measurement items, implying the correlation
matrix of present data is appropriate for factoring (Sharma, 2005).
Table 3 shows the signicant classication of four factors. The two
factors in the upper half of Table 3 indicating the quality and ow
of R&D talent explain 52.69% of the measurement variance of
human capital, while the other two factors in the lower half of
Table 3 pertaining to R&D resource sharing and perceived
participation benets reect 61.42% of the variance associated
with measuring relational assets. Furthermore, the authors applied
the SAS Factor procedure for residual correlations analysis to
assess the estimated factor solutions. The derived overall RMSE
(Root-Mean-Square Errors) values are 0.098 and 0.084 for human
capital and relational assets respectively. Thus, the estimated
factor models can be considered to be appropriate due to small
RMSE values (Sharma, 2005). The results indeed t the four
measurement constructs of NSTP performance evaluation elaborated by the authors.
4.2. Evaluation of human capital
The authors then applied the human capital construct to
evaluate the performance of the NTP case. The t-test method tests
each measurement item to discover whether the mean exceeds the
median value. Table 4 shows the t-test results. All the measurements pertaining to human capital signicantly exceeded the
median value, except the item associated with external brokers of
expertise. This may indicate that the employment matching
process/talent brokerage instituted by the NTP must, from the
industrial viewpoint, be improved. Nevertheless, the quality of
R&D talent as perceived by industrial demanders is apparently
high enough to be a high priority. To induce NTP-trained talent to
ow into telecommunications jobs within the domestic industry,
higher salaries and greater promotion opportunities are offered,
and more challenging tasks assigned. Further, this higher quality of
employment reects the strategic purposes of governmental
support for national-level science and technology research. It also
reects the point of the absorptive capability of human resources
in order to be better equipped to scan the technology frontier,
facilitate international technology transfers, and raise the possibility of invention in the future (Cohen & Levinthal, 1990).
Therefore, the data analysis supports hypotheses H1 and H2
(mostly).
The authors further investigated whether the different types of
industrial rms differ in terms of their quality perceptions of
talented individuals. As shown in Table 5, NTP industrial
participants generally had higher demand for importing telecom-related expertise from NTP than from non-NTP participants.
Further, industrial telecommunication demanders were classied
into manufacturers, such as equipment vendors, and service
providers, such as telecom operators, after which the effects of
each industrial role were compared. The sole signicant difference

C.-L. Hung et al. / Evaluation and Program Planning 33 (2010) 487497

492
Table 3
Factor analysis of measurements.

Human capital

Quality of R&D talent

Flow of R&D talent

Relational assets

R&D resource sharing

Perceived participation benets

Measurement items

Factor 1

Factor 2

! Recruitment willingness
! Salary level
! Promotion opportunity
! Performance
! Knowledge acquisition capability
! Achievement of organization objectives
! R&D-oriented tasks
! Innovation management
! Organizational problem solving
! Personnel spillover into domestic industries
! Personnel spillover into NSTP-related technological elds
! Expertise imported from NSTP
! External brokers of expertise
l=
Cumulative=

0.21743
0.53171
0.94296
0.88577
0.79836
0.75323
0.66741
0.82513
0.73197
0.33276
0.14367
0.11469
0.23114
5.0698
0.5269

0.13278
0.01397
0.00516
0.08467
0.05305
0.00363
0.35654
0.04334
0.12184
0.52349
0.79101
0.65244
0.53257
1.7795

! Co-publishing the research results

! Cooperation for commercialization
! Licensing between participants
! Technical meetings and conferences
! Reorganization for new research agendas
! Research co-funding
! Experimental data sharing
! Sharing equipment and facilities
! Researcher training
! Breakthrough in a critical technology
! Increasing awareness on the importance of R&D
! Accelerated development of the technology
! Decrease in private R&D budget
! Acquisition of knowledge from other participants
! Establishment of an ongoing informational network
! Commercialization of a product or process
! Increase in the number of patent applications
! A superior competitive position
! Establishment of a standard in a target industry
l=
Cumulative=

0.01380
0.06053
0.09704
0.01283
0.08755
0.12238
0.09648
0.07005
0.75682
0.80906
0.74163
0.63615
0.72444
0.55589
0.79688
0.83355
0.83901
0.88560
0.86924
6.6521
0.6142

0.84081
0.79822
0.84883
0.91801
0.51310
0.70920
0.86604
0.73502
0.12574
0.01510
0.01649
0.03476
0.16188
0.05792
0.10719
0.00612
0.06382
0.04610
0.02878
5.0173

Table 4
Signicance of human capital performance.

Quality of R&D talents

Flow of R&D talents

*
**
+

Measurement items

Mean

t-value+

! Recruitment willingness
! Salary level
! Promotion opportunity
! Performance
! Knowledge acquisition capability
! Achievement of organization objectives
! R&D-oriented tasks
! Innovation management
! Organizational problem solving

6.0
5.1
5.4
5.1
5.6
5.2
4.9
5.3
5.1

16.70**
4.89**
6.78**
5.02**
8.32**
5.33**
3.59**
6.07**
4.25**

H1 (supported)

! Personnel spillover into domestic industries

! Personnel spillover into NSTP-related technological elds
! Expertise imported from NSTP
! External brokers of expertise

6.0
4.7
4.3
3.0

17.02**
2.23*
2.04*
4.81**

H2 (mostly supported)

Hypothesis

p < 0.05.
p < 0.01.
One-tailed t-tests are comparisons of mean scores with the middle (indifferent) value, 4.

Table 5
Signicant performance differences on human capital across recruiters.
Recruiters

Human Capital

Mean

t-value+

NTP participants vs. Non-NTP participants

Manufacturers vs. Service providers

! Demand Flow: Expertise imported from NSTP

! Salary level
! Promotion opportunity

4.65 vs. 3.91

5.34 vs. 4.00
5.46 vs. 4.80

2.961**
5.30**
2.169*

*
**
+

p < 0.05.
p < 0.01.
t-tests are comparisons of the mean scores of two samples.

C.-L. Hung et al. / Evaluation and Program Planning 33 (2010) 487497

493

Table 6
Signicance of relational assets performance.
Measurement items

Mean

t-value+

R&D resource sharing

! Co-publishing the research results

! Cooperation for commercialization
! Licensing between participants
! Technical meetings and conferences
! Reorganization for new research agendas
! Research co-funding
! Experimental data sharing
! Sharing equipment and facilities

3.8
3.4
3.5
4.2
4.4
3.1
4.2
3.2

0.89
1.99
1.73
0.71
1.52
3.16
1.00
3.24

Perceived participation benets

! Researcher training
! Breakthrough in a critical technology
! Increasing awareness on the importance of R&D
! Accelerated development of the technology
! Decrease in private R&D budget
! Acquisition of knowledge from other participants
! Establishment of an ongoing informational network
! Commercialization of a product or process
! Increase in the number of patent applications
! A superior competitive position
! Establishment of a standard in a target industry

5.1
5.2
5.8
5.0
5.5
5.2
5.5
5.3
5.5
5.8
5.5

7.15**
8.75**
13.00**
5.08**
9.64**
7.10**
8.43**
7.58**
8.43**
12.34**
10.4**

**
+

Hypothesis

H4 (supported)

p < 0.01.
One-tailed t-tests are comparisons of measurement mean with the middle (indifferent) value, 4.

Table 7
Signicant relational assets performance by NTP industrial participants.

R&D resource sharing

**
+

Measurement items

Mean

t-value+

Hypothesis

! Co-publishing the research results

! Research co-funding
! Sharing equipment and facilities

5.65
5.52
5.65

5.32**
4.50**
5.09**

H3
(Partially supported)

p < 0.01.
t-tests are comparisons of mean score with the middle (indifferent) value, 4.

is shown in Table 5, where both industrial participants, whether or

not they had ever worked with the NTP, evaluated the quality of
NTP talents equally, exclusive of salary and promotion opportunities. In reality, according to the information from NTP project
investigators, most of the trained talent ows into the manufacturing sector (manufacturing vs. service: 61.3% vs. 38.7%), even
though both telecom service providers and manufacturers demand
expertise from NTP equally, and both stated their intention to place
NTP-trained talent in challenging positions such as R&D, innovation, and organizational problem solving. Manufacturing rms
generally provide higher salaries and more promotion opportunities. This stronger incentive appears to explain why the majority
of trained talent ows into the manufacturing sector and
contributes to Taiwans ourishing global share of broadband
device manufacturing.
4.3. Evaluation of relational assets
In the same way, the authors adopted the veried measurement
construct of relational assets to evaluate the performance of the

NTP case and also employed t-tests to determine whether the score
for each measurement item exceeded the middle value. Table 6
shows the statistical results. Only the perceived participation
benet measurements signicantly exceeded the median value.
Therefore, hypothesis H4, which argues that greater perceived
participation benets enhance the value of relational assets within
the NSTP, is supported. Each of the measurement items regarding
participation benets was signicant. Further, these results
indicate that the spillover effect remains a top priority for
government-sponsored research programs aimed at technology
diffusion and industry position upgrading, as found in Georghiou
(1998). Contrastingly, the sharing of R&D resources between NTP
partners was non-signicant. This unusual nding may be due to
the fact that nearly half of the data was collected from university
professors, who primarily receive government-sponsored research
grants and often concentrate on academic publishing rather than
diffusion of technology to industry. For this reason, the authors
further scrutinized the questionnaire data and only veried the
signicance of 31 industrial responses from those who had
previously participated in the NTP; these results are shown in

Table 8
Signicant differences of relational asset performance across different industrial participants.
Industrial participants
! NTP participants vs. Non-NTP participants
! Ex ante vs. Ex post benets of NTP participants
! High- vs. low-degree of resource sharing by
NTP participants
! Manufacturers vs. service providers
*
**
y
+

Relational asset performance

Breakthrough in a critical technology
Accelerated development of the technologyy
Acquisition of knowledge from other participants

Mean

t-value+

Hypothesis

5.48 vs. 4.93

6.05 vs. 5.35
3.62 vs. 2.86

2.209
3.034**
2.718*

5.34 vs. 4.40

2.105*

H5 (Mostly Supported)

p < 0.05.
p < 0.01.
Only the signicantly unsatisfactory benet by the before-and-after comparison on eleven NTP participation benet measurements is presented here.
t-tests are comparisons of the mean scores of the two samples.

C.-L. Hung et al. / Evaluation and Program Planning 33 (2010) 487497

494

Table 9
Signicance of relation between resource sharing and ow of R&D talent.
Resource sharing

Flow of R&D talents

Mean

t-value+

! High- vs. low-degree of resource sharing by industrial NTP participants

! High- vs. low-degree of Experimental data sharing by NTP project investigators
! High- vs. low-degree of cross-licensing between industryuniversity cooperation

Demand ow: expertise imported from NSTP

Supply ow: personnel spillover into
NSTP-related technological elds

4.52 vs. 3.96

5.08 vs. 4.28
5.03 vs. 4.17

2.187*
1.845*
4.557**

*
**
+

p < 0.05.
p < 0.01.
t-tests are comparisons of measurement means of two samples.

Table 7. The sharing of R&D resources between NTP partners from

the perspective of NTP industrial participants revealed signicance
only pertaining to the measurement items of co-funding, copublishing, and sharing of equipment and facilities. As a result, the
ndings partially support hypothesis H3.
Additionally, the authors investigated whether the different
types of industrial participants differed in terms of their perceived
NTP participation benets. The results are shown in Table 8. The ttest verication shows that the perceived possibility of a
breakthrough in critical technology is one signicant inducer for
NTP participants because of higher expected benets. If we focus
only on NTP participants, the before-and-after paired comparisons
of gaps between the expected benets and actual gains were nonsignicant except for the inadequacy of accelerated technology
development. This implies that NTP participants generally
experience a satisfactory level of cooperation with partners from
university professors or researchers of national R&D organizations
as they strive to achieve advanced telecommunications technologies. Hence, hypothesis H5 is also supported. However, achievement in the accelerated development of technology was lower
than we expected. This situation may have been caused by a lack of
common focus in the universityindustry cooperation; professors
in Taiwan universities usually concentrate on the time to publish,
while industrial participants emphasize the speed to commercialization for global competition. Further, the research found that
those industrial participants who engaged in a higher degree of
R&D resource sharing actually experience greater acquisition of
knowledge from other participants. In addition, another t-test
comparison also revealed that industrial manufacturing roles are
associated with higher benets from knowledge acquisition than
telecom service providers. The capability of knowledge acquisition
indeed enhances the new product development in integrating
complements and skills (Mohr, Sengupta, & Slater, 2005). Hence,
this may indicate that NTP performance facilitates Taiwan telecom
manufacturers to quickly congure all telecom systems and thus
rapidly supply mobile devices and broadband equipments to
markets, which in turn improves their global standing.

4.4. Further data exploration among measurement constructs

Finally, the authors explored the relationship between the NSTP
performance evaluations. Signicant and meaningful ndings are
shown in Tables 9 and 10. When focusing on the demand ow of
industrial participants, those who expressed greater willingness to
share R&D resources also indicated a greater demand for expertise
from NTP. By the same token, the t-test comparison for the supply
ow of R&D talent showed that academic researchers who share
more experimental data or licenses with NTP partners also offer
project-trained members more opportunities for employment
within telecom-related technological industries. Moreover, the
regression analysis in Table 10 revealed some meaningful ndings
between the perceived participation benets and the quality of the
NTP-trained R&D talent. Greater satisfaction in terms of increased
awareness of R&D importance was associated with greater
willingness to recruit NTP-trained talent. At the same time, greater
satisfaction in terms of the acquisition of knowledge from other
participants was associated with greater willingness to recruit
NTP-trained talent as well as a higher valuation for performance,
and greater likelihood of assigning challenging innovation
management tasks to new recruits. In addition, satisfaction with
the increase in the number of patent applications leads to
employees being evaluated higher in terms of achieving organizational objectives. Lastly, greater satisfaction with the commercialization of a product or process leads to a greater possibility of
raising employee salaries due to high performance. Thus, the
higher salaries attract more R&D talent to the industry, especially
to industrial employers who participate in the NSTP and
experience satisfaction through industry-university cooperation.
Accordingly, Table 11 summarizes the ndings from verications of and comparisons among the measurement constructs of
the NSTP evaluation framework. In Table 11, the arrows between
the measurement constructs indicate the potential inuences.
Consequently, if the NSTPs embed more incentives for interaction
and sharing behaviors between project partners, especially among
industrial participants and university professors, the resulting

Table 10
Signicance of relation between participation benets and quality of R&D talent.
Participation benets

Quality of R&D talent

t-value+

Satisfaction with increasing awareness of the importance of R&D

! Recruitment willingness

3.395**

Satisfaction with acquisition of knowledge from other participants

! Recruitment willingness
! Performance
! Innovation management

4.314**
2.562*
2.444*

Satisfaction with increase in the number of patent applications

! Recruitment willingness
! Performance
! Achievement of organization objectives
! Innovation management

3.37**
2.70*
4.36*
2.924*

Satisfaction with commercialization of a product or process

! Salary level
! Performance

3.18**
2.96*

p < 0.05.
p < 0.01.
t-tests verify the signicance of regression coefcients that are used to predict the relationship between participation benets (independent variables) and quality of R&D
talent (dependent variables).
**
+

C.-L. Hung et al. / Evaluation and Program Planning 33 (2010) 487497

495

Table 11
Summary of ndings.

Note: The arrows imply possible relationships between measurement constructs.

trained human capital, enriched with sharing and knowledge

acquisition abilities, will be greatly demanded in the industry. It
appears that the NSTPs should promote circulation of knowledge
even though outside employment brokerages remain inadequate
when faced with highly immature and uncertain technology.
Moreover, if the NSTP ofce initiates coordinating mechanisms to
facilitate industry-university cooperation, participant satisfaction
may increase, and quality evaluations of NSTP-trained talent will
certainly increase. Consequently, the higher quality of human
capital will enhance the possibility of higher salaries during
recruitment and thereby contribute toward promoting NSTP
expertise across the industry. Hence, the important diffusion
purpose of most NSTP policies can be achieved successfully.
5. Conclusions
The current study adopts a perspective different from the
conventional quantitative approach of directly counting output in
terms of patent applications and citations to evaluate a research and
development program. Using the perspectives of human capital and
relational assets, it explains how the Taiwan NTP is successful
despite its lack of outstanding technological patents. In this study,
the authors emphasize the intellectual capital on and between
the trained talents themselves. The quality and employment ow of
human capital trained by the NSTP is the rst measurement of

concern pertaining to the on-talent dimension; the other is the

between-talent measurement dimension focusing on the relational
assets that can be attributed to sharing R&D resources and the
experience of participation benets. The research in the NTP case
generally supports the on-talent aspect; the quality of the NTPtrained human capital is perceived better than the middle value (the
indifferent point), which results in the manufacturing industry
offering higher salaries to NTP-trained individuals. NTP members
also effectively ow into the Taiwan telecommunications industry,
especially to the manufacturing sector. Yet, the authors show that
the construct of ex ante perceived participation benets is more
effective than the behavior of sharing R&D resources for evaluation
of relational assets. The measurement results of perceived
participation benets indicate that NTP industrial participants are
more eager to achieve breakthroughs in critical technologies by
cooperating with academic researchers than participants not
involved in NTP. Even though most industrial NTP participants
eventually become disappointed with the accelerating development
of technology associated with commercial applications, they are
generally satised with the collaboration experience because of the
very small expectation gap pre- and post-participation. In addition,
industrial participants with higher industry-university cooperation
intensity gain a particular advantage, namely, greater knowledge
acquisition from various other elds due to knowledge spillover.
Taiwan telecommunications manufacturers stress the acquisition of

496

C.-L. Hung et al. / Evaluation and Program Planning 33 (2010) 487497

knowledge from collaborative partners to improve and accelerate

the integration of telecommunications system products (e.g.,
wireless and broadband devices), and capture global markets. It is
thus reasonable that Taiwan telecommunications manufacturers
have a greater willingness to offer higher salaries and more
promotion opportunities to NTP-trained talents in contrast to
telecom service providers.
Accordingly, in terms of the national technology policy,
measuring performance according to the perspectives of human
capital and relational assets highlights the importance of the
increase in absorptive capacity and knowledge spillover contributed by NSTP, as opposed to the conventional quantitative method
that considers patent analysis and scientic impact. Thus, targeting
the performance criteria associated with human capital and
relational assets engendered by the government-sponsored NSTP
is more appropriate than investigating the technological output,
which only pertains to a few highly developed countries.
Furthermore, based on the disappointment with accelerating
application development of targeted technology among the
industrial participants, which corresponds with the results of
Sakakibara (1997) for the Japanese government-sponsored NSTP,
Taiwans NSTP program should not restrict itself to technological
breakthroughs, but should also consider the elds of marketing
research, regulation reform, infrastructure restructuring, and
education as well as other elds that complement and nourish
application innovation and knowledge diffusion associated with
new technology. In fact, Georghiou (1998) suggests that the
coordination function is the most important criterion when
evaluating the performance of any scientic program. Hence, the
NSTP technology policy design should emphasize liaison mechanisms to harmonize the objectives of the participants, thereby
creating a space where heterogeneous relational alliances can be
formed between public and industrial laboratories, between
industrial competitors, or between rms with complementary
resources. Further, while the academic sector usually pursues
personal scientic achievements, the industrial sector focuses on the
potential for commercialization according to a strict cost-benet
analysis. Hence, policies that can help to bridge this gap are
imperative. For example, university researchers could receive career
compensation pertaining to their role in carrying out applied
research, or a tax reduction could be offered to industrial NSTP cofunders for their support in pioneering research. In addition,
intermediary roles in marketing, public relations, and nance
should also be initiated based on market demand or new ventures
that attract the interests of industrial practitioners even when the
dominant purpose of NSTP focuses mainly on the technology
advancement. If carried out, NSTP participant satisfaction will
improve as recruitment of NSTP-trained members increases due to
their higher quality evaluations. Consequently, a coordination or
liaison mechanism installed in the NSTP that supports program
interaction will then support increasing returns in terms of
absorptive capability and technology spillover, thus efciently
achieving the goal of a diffusion-oriented technology policy.
The limits of the current study may be found in the problem of
generalization. The targeting case, telecommunications, is an
applied technology area that lacks a strong component of basic
research, unlike nanotechnology or biotechnology. For future
research, broadening the sample space encompassing different
orientations of NSTP is necessary to justify and adjust the
measurement constructs articulated in this study.
Acknowledgements
The authors wish to acknowledge the nancial support from the
BRIDGE project of National Science Council of Taiwan under grant
no. NSC-96-3114-P-260-001-Y.

Appendix A
Table A-1 Questionnaire of intellectual-based NSTP evaluation
framework.
Questions on quality of R&D talent (1 = not at all, 7 = very much)
1. Does (Did) your company (ever) like to recruit the member ever
trained by National Telecommunication Program (NTP)?
2. Does (Did) your company (ever) give a higher level of salary to
the employee trained by NTP?
3. Does (Did) your company (ever) give a higher promotion
opportunity to the employee trained by NTP?
4. Does (Did) your company (ever) get a higher level of
performance from the employee trained by NTP?
5. Do (Did) you (ever) think that the employee trained by NTP
possesses a higher level of knowledge acquisition capability?
6. Do (Did) you (ever) think that the employee trained by NTP has a
higher possibility of achieving the organization objectives?
7. Does (Did) your company (ever) assign a higher proportion of
R&D-oriented tasks to the employee trained by NTP?
8. Do (Did) you (ever) think that the employee trained by
NTP possesses a higher capability of managing innovation
process?
9. Do (Did) you (ever) think that the employee trained by NTP
possesses a higher capability to solve organizational problems?
Questions on ow of R&D talent (1 = not at all, 7 = very much)
10. Do you think that most of your NTP project members owed
into the domestic industry?
11. Do you think that most of your NTP project members owing
into the domestic industry targeted the telecom-related segments?
12. Does (Did) your company still (ever) has the demand to acquire
expertise from the NTP?
13. Does (Did) your company (ever) acquire telecommunication
expertise by the help of external agents?
Questions on R&D resource sharing (1 = not at all, 7 = very much)
14. Does (Did) your NTP project (ever) co-publish the research
results with external research centers, university laboratories, or
companies?
15. Does (Did) your NTP project (ever) cooperate with external
research centers, university laboratories, or companies for
commercializing the new technology?
16. Does (Did) your NTP project (ever) license technological
patents mutually between the NTP participants?
17. Does (Did) your NTP project (ever) meet and discuss with
external research centers, university laboratories, or companies in
the telecom-related conference?
18. Does (Did) your NTP project (ever) reorganize a new research
agenda with external research centers, university laboratories, or
companies after the NTP cooperation?
19. Does (Did) your NTP project (ever) receive research funds
from external research centers, university laboratories, or
companies?
20. Does (Did) your NTP project (ever) share experimental data
with external research centers, university laboratories or companies?
21. Does (Did) your NTP project (ever) share research equipments
and facilities with external research centers, university laboratories or companies?
Questions on perceived participation benets (1 = not at all,
7 = very much)

C.-L. Hung et al. / Evaluation and Program Planning 33 (2010) 487497

22. Is it an important purpose of participating into the NTP for

beneting from better researcher training?
23. Is it an important purpose of participating into the NTP for
beneting from earlier breakthrough in the critical technology?
24. Is it an important purpose of participating into the NTP for
beneting from increasing awareness on the importance of
telecommunication R&D?
25. Is it an important purpose of participating into the NTP for
beneting from accelerating application development of telecommunication technology?
26. Is it an important purpose of participating into the NTP for
beneting from subsiding research funds from public sector?
27. Is it an important purpose of participating into the NTP for
beneting from acquiring/sharing knowledge from other participants?
28. Is it an important purpose of participating into the NTP for
beneting from establishment of an ongoing informational
network with other participants including private and public
sectors?
29. Is it an important purpose of participating into the NTP for
beneting from speeding the new product commercialization
process?
30. Is it an important purpose of participating into the NTP for
beneting from increasing the number of patent applications?
31. Is it an important purpose of participating into the NTP for
beneting from obtaining a superior position in facing the
telecommunication competition?
32. Is it an important purpose of participating into the NTP for
beneting from quicker establishing an industrial standard of new
telecommunication technology?
References
Bontis, N. (1998). Intellectual capital: an exploratory study that develops measures and
models. Management Decision, 36(2), 6376.
Bozeman, B., Dietz, J., & Gaughan, M. (2001). Scientic and technical human capital: an
alternative model for research evaluation. International Journal of Technology
Management, 22(7/8), 636655.
Brown, E. A. (1997). Measuring performance at the army research laboratory: the
performance evaluation construct. Journal of Technology Transfer, 22(2), 2126.
Callon, M., Laredo, P., & Mustar, P. (1997). Strategic management of research and
technology: evaluation of programmes. Paris: Economica International.
Chiang, J. T. (1990). Management of national technology programs in a newly industrialized countryTaiwan. Technovation, 10(8), 531554.

497

Cohen, W. M., & Levinthal, D. A. (1990). Absorptive capacity: a new perspective on

learning and innovation. Administrative Science Quarterly, 35(1), 128152.
Cozzens, S. E. (1997). The knowledge pool: measurement challenges in evaluating
fundamental research programs. Evaluation and Program Planning, 20(1), 7789.
Dalpe, R. (1993). Evaluating the industrial relevance of public R&D laboratories. In B.
Bozeman & J. Melkers (Eds.), Evaluating R&D impacts: methods and practice. Boston:
Kluwer Academic Publishers.
Georghiou, L. (1998). Issues in the evaluation of innovation and technology policy.
Evaluation, 4(1), 3751.
Kundu, S. C., & Kumar, R. (2006). Evaluating benets of training and development: a
study of Indian and multinational companies. In Proceedings of 12th Asia Pacic
Management Conference (pp. 760768) [ISSN: 974-8257-30-4].
Lanzi, D. (2007). Capabilities, human capital and education. Journal of Socio-Economics,
36(3), 424435.
Lepori, B. (2006). Methodologies for the analysis of research funding and expenditure:
from input to positioning indicators. Research Evaluation, 15(2), 133143.
Luwel, M. (2005). Job advertisements as an indicator for mobility of researchers:
naturejobs as a case study. Research Evaluation, 14(1), 8092.
Mann, S., & Robertson, I. T. (2006). What should training evaluations evaluate? Journal
of European Industrial Training, 20(9), 1420.
Mohr, J., Sengupta, S., & Slater, S. (2005). Marketing of high-technology products and
innovations (2nd ed.). Prentice-Hall.
NSC. (2007a). White paper on science and technology (20072010). Taipei: National
Science Council.
NSC. (2007b). 2007 Yearbook of science and technology of Taiwan. Taipei: National
Science Council.
OECD (Organization for Economic Cooperation Development). (2002). Frascati manual:
proposed standard practice for surveys on research and experimental development.
Paris: OECD Publishing.
Rogers, J. D., & Bozeman, B. (2001). Knowledge value alliances: an alternative to the
R&D project focus in evaluation. Science, Technology & Human Values, 26(1), 2355.
Sakakibara, M. (1997). Evaluating government-sponsored R&D consortia in Japan: who
benets and how? Research Policy, 26(45), 447473.
Sharma, S. (2005). Applied multivariate techniques. John Wiley & Sons Inc.
Steinbock, D. (2002). Wireless horizon: strategy and competition in the worldwide mobile
marketplace. New York: AMA.
Stewart, T. (1997). Intellectual capital: the new wealth of organizations. Doubleday
Publishing.
Thomas, P., & Breitzman, A. (2006). A method for identifying hot patents and linking
them to government-funded scientic research. Research Evaluation, 15(2), 145
152.
Wu, H. L. (2006). Economic analysis of developing technological human capitals by
technology budget. Taipei: CIER.
Chia-Liang Hung is an assistant professor of department of information management
at National Chi Nan University in Taiwan. He holds a PhD from National Taiwan
University and now specializes in management of technology and strategy of electronic
commerce.
Jerome Chih-Lung Chou is an assistant professor of Department of Information
Management at Hwa-Hsia Institute of Technology in Taiwan. He holds a PhD from
National Taiwan University and now specializes in technology strategy and entrepreneurship of small business.
Hung-Wei Roan holds an MBA degree of National Chi Nan University in Taiwan. He
specializes in designing and implementing web-based software applications.