See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/364315180

Assessing the effect of business intelligence quality on enterprise innovation

performance: A knowledge dynamic capability perspective

Article  in  International Journal of Technology Management · May 2023

DOI: 10.1504/IJTM.2022.10048678

CITATIONS READS

0 146

4 authors, including:

Kuan Yan Enjun Xia

Beijing Institute of Technology Beijing Institute of Technology
3 PUBLICATIONS   2 CITATIONS    14 PUBLICATIONS   609 CITATIONS   

SEE PROFILE SEE PROFILE

Jieping Huang
Beijing Institute of Technology
7 PUBLICATIONS   109 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Kuan Yan on 17 October 2022.

The user has requested enhancement of the downloaded file.

Assessing the Effect of Business Intelligence Quality on Enterprise

Innovation Performance: A Knowledge Dynamic Capability Perspective

Kuan Yan, Enjun Xia, Jun Li, Jieping Huang*

School of Management and Economics, Beijing Institute of Technology, Beijing, China

Citation

Yan, K., Xia, E., Li, J., & Huang, J. (2023) ‘Assessing the effect of business intelligence quality

on enterprise innovation performance: A knowledge dynamic capability perspective’,

International Journal of Technology Management, Vol. ahead-of-print No. ahead-of-print.

https://doi.org/10.1504/IJTM.2022.10048678

*
Corresponding author: [email protected]. Address: Zhongguancun South Street 5, Haidian District, Beijing 100081, China

1
Abstract: Enterprises leverage business intelligence (BI) to promote competitive advantages.

However, there is still a lack of understanding of BI quality and innovation performance. This

research constructed a model of the impact of BI’s system quality and information quality on

innovation performance through the IS success and dynamic capabilities framework,

investigated the mediation effect of knowledge sharing and absorptive capability and

examined heterogeneity with potential IT differences between industries. The results with

Chinese enterprise data show that both system quality and information quality are positively

related to knowledge sharing, and system quality is positively related to absorptive capability,

while information quality is not. Through the mediation effect of knowledge sharing and

absorptive capability, system quality partially facilitates innovation performance, while

information quality completely facilitates innovation performance, although indirectly. There

is no significant relationship between system quality and knowledge sharing among

traditional industries. These findings contribute new insights into BI-enabled innovation.

Keywords: business intelligence quality; system quality; information quality; innovation

performance; dynamic capabilities; knowledge sharing; absorptive capability

2
1. Introduction

In today's big data business world, the innovative activities of many firms are increasingly

dependent on the analysis and utilisation of data and the investment in data infrastructure

(Dewan and Min, 1997;Melville, Kraemer and Gurbaxani, 2004). Business intelligence (BI)

promotes innovation and gains a competitive advantage for enterprises through big data storage,

analysis and visualization (Chen, Chiang and Storey, 2012;Eidizadeh, Salehzadeh and Chitsaz

Esfahani, 2017). Although literature describes BI as a strategic initiative that leads to business

success, the successful implementation of BI innovations remains both a theoretical and a

business challenge (Ahmad, Ahmad and Hashim, 2016).

Despite the increasing use of BI in companies, research on the causal role between BI

quality and innovation performance is still lacking. BI is generally considered to be part of

innovation strategy since it helps analyse big data and decision-making, thereby increasing the

possibility of innovation success (Bernardino and Neves, 2016;Trieu, 2017). However, just

as more big data are not always better for innovation (Ghasemaghaei and Calic, 2020), BI does

not necessarily lead to greater innovation. The coupling between BI and organisational

capabilities is more critical (Park, El Sawy and Fiss, 2017). Compared to the use of BI, BI

quality focuses more on the transitions from technical to organisational value (Gorla, Somers

and Wong, 2010). BI quality is an important part of enterprises’ total quality management and,

thus, it makes sense to assess the success factors of BI (Ravichandran and Rai, 1999). Based

on the theory of IS critical success factors (CSFs) of DeLone and McLean (1992), many studies

argue business intelligence encompasses both system quality and information quality, which

reveals the BI value from two perspectives (Yeoh and Popovič, 2016;Kulkarni, Robles-Flores

3
and Popovič, 2017).

Therefore, it is essential to investigate whether and how BI quality contributes to

innovation performance. A significant focus when exploring the causality of BI quality on

innovation performance is the notion of knowledge (Ghasemaghaei, 2019), which especially

refers to BI-enabled knowledge (Cui et al., 2020). Knowledge is a dynamic competitive

resource for companies and is the basis of innovation (Roberts et al., 2012). It involves

knowledge sharing for interchange, as well as absorptive capability for absorption and

utilisation (Liao, Fei and Chen, 2007). However, there is still a lack of proof to show the

relationship of knowledge sharing and absorptive capability on BI quality and innovation

performance.

This research attempts to fill this research gap by proposing and validating a new model

to explain the relationship between BI quality and innovation performance from the perspective

of dynamic capabilities in knowledge, highlighting the significance of BI as a strategic

approach coupled with knowledge sharing and absorption. Specifically, BI drives external data

and information into the organisation and transforms them into innovation resources through

knowledge interchange, absorption, and utilisation (Shollo and Galliers, 2016;Steininger et al.,

2022). Similar to big data analytics, it will only benefit innovative performance if organisations

can fully synergise their capabilities with BI (Mikalef et al., 2020). Knowledge absorption is a

part of the dynamic capabilities and has a role in the processes, routines, structures, and

activities of organisations, constantly responding to internal and external environmental

changes (Lane, Koka and Pathak, 2006;Mikalef et al., 2019). This study seeks to investigate

the following two questions:

4
 (1) What are the individual and combined effects of BI’s system quality and information

quality on innovation performance?

(2) How do knowledge sharing and absorptive capability play a mediating mechanism

between BI quality and innovation performance?

The goals of this study are novel, as most previous studies have focused more on BI use

or big data analytics. This research demonstrated the impact of BI quality on innovation

performance under the framework of dynamic capabilities, which brought new insights into BI

success. Through the lens of knowledge absorption in dynamic capabilities, a mediating

mechanism for knowledge sharing and absorptive capability is hypothesised, expanding the

understanding of knowledge management in the IS domain. The theoretical model and research

hypotheses are validated based on data from Chinese enterprises. Finally, further heterogeneity

among industries with different potential IT capabilities is examined through a multigroup

analysis. It can be expected that this model will help managers assess the relationship between

BI quality and innovation performance and help enterprises to utilise BI more effectively to be

successful.

2. Theoretical background
2.1 System quality and information quality of BI

The underlying dimensions of BI quality are system quality and information quality (Yeoh

and Koronios, 2010), which are the key exogenous variables contributing to user satisfaction

and business value creation and are the main dimensions of total BI-based quality management

for enterprises (Popovič et al., 2012). It is grounded in DeLone and McLean’s (1992) critical

5
success factor model of information systems, which assumes that IS value is influenced by the

system dimensions of willingness to use and perceived value, which in turn create user value

(Nelson, Todd and Wixom, 2005). Previous studies adopting this framework have focused on

both the organisational value level and the level of individual use, yet most have used the

individual level, such as information use and personal satisfaction (Gorla, Somers and Wong,

2010). Studies targeting the individual level, overwhelmingly following the D&M framework,

confirm the positive impact of BI’s system quality and information quality on information use

(Yeoh and Koronios, 2010;Popovič et al., 2012;Popovič et al., 2014). At the organisational

level, however, the lack is still present in understanding the effect of BI quality on innovation

performance. Innovation is one of the most essential long-term IS strategies (Prahalad and

Krishnan, 2002). Thus, managers seek how to assess how BI quality contributes to innovative

performance (Rai and Tang, 2010;IşıK, Jones and Sidorova, 2013). In sum, whether as an

exogenous variable of BI success or as an important dimension of BI total quality management,

the system quality and information quality of BI are of great significance for BI value

assessment.

The system quality of BI is the assessment and perception of the system performance of

BI, which describes system reliability, response time, ease of use, flexibility, etc (DeLone and

McLean, 2003;Chan and Lau, 2018). Researchers argue that BI system quality can improve

performance and decision-making processes (Chan and Lau, 2018). A higher system quality of

BI provides more stable and reliable workflows and users’ satisfaction with systems (Au, Ngai

and Cheng, 2002). BI information quality is the quality of information output from BI, which

is represented by information characteristics such as accuracy, objectivity, and completeness,

6
etc (Nelson, Todd and Wixom, 2005;Kulkarni, Robles-Flores and Popovič, 2017). Higher BI

information quality provides managers with accurate data information and bases for more

timely decisions, enabling process optimisation and agility (Ahn and Sura, 2020;Popovič et al.,

2014).

However, the literature lacks an understanding of the role of BI quality in innovation

performance. Current studies focus on the relationship between the use of BI and innovation

(Ain et al., 2019;Božič and Dimovski, 2019b). Innovation is conducted when companies

respond to unconventional issues and implement improvements in the original resource

allocation that comes from organisational learning and knowledge (Ghasemaghaei and Calic,

2020;Kranz, 2021). Studies discuss the relationship between BI, knowledge management, and

firm innovation. Eidizadeh, Salehzadeh and Chitsaz Esfahani (2017) emphasise the positive

effect of BI on knowledge sharing and organisational innovation. Some literature argues that

knowledge sharing and organisational learning play a mediating role between BI and

innovation (Giménez-Figueroa, Martín-Rojas and García-Morales, 2018;Caseiro and Coelho,

2019). During the process of leveraging BI for innovation, organisations must first acquire

internal and external data and information and then utilise BI for data analysis and visualisation

(Ain et al., 2019). Therefore, organisational knowledge capability might play a critical

potential mediating role in the effect of BI quality on innovation performance.

In addition, it is interesting that the success of BI varies in different industries (Yeoh and

Koronios, 2010). This may be because the strategic positioning and uses of BI in different

industries are different, e.g., big-data-driven companies require real-time monitoring of data

from users by BI, manufacturing companies require fast feedback on data from the supply chain,

7
hospitals require objective and reliable analysis of patient data processing, etc (Hung et al.,

2010;Rikhardsson and Yigitbasioglu, 2018). The potential differences in IT infrastructure and

human resources across industries may result in heterogeneous understandings of BI and lead

to different outcomes (Melville, Kraemer and Gurbaxani, 2004).

2.2 Dynamic capabilities of knowledge absorption

Knowledge is a critical and competitive element of an organisation, influencing its

structure, processes, human resources, etc. (Mikalef et al., 2019). There are three different

perspectives on organisational knowledge management in the literature: knowledge assets,

knowledge capability, and knowledge absorption based on the dynamic capabilities view (Lane,

Koka and Pathak, 2006;Steininger et al., 2022). The first two perspectives represent the

resource and capacity perspective, but they focus on knowledge within the organisation and

ignore the dynamic nature of knowledge and the impact of environmental uncertainty (Božič

and Dimovski, 2019a;Conboy et al., 2020). Digital enterprises emphasise the fluidity of

knowledge, yet relying solely on knowledge within an organisation limits the ability to respond

to changes in the external environment. The concept of dynamic capabilities originated from

strategic management research and is now widely used in information systems management

(Winter, 2003;Warner and Wäger, 2019). Dynamic capabilities are the ability of organisations

to continuously respond to uncertain external opportunities and challenges and gain

competitive advantage by acquiring, integrating and leveraging resources (Teece, Pisano and

Shuen, 1997;Barreto, 2010;Karimi and Walter, 2015). Dynamic capability theory provides a

better understanding of how firms utilise IT to profit from the uncertainty of a dynamic

8
environment and is widely used in research on BI, BA, big data, etc. and organisational

performance (Conboy et al., 2020;Steininger et al., 2022).

Knowledge absorption is characterised by dynamic capabilities (Miroshnychenko et al.,

2020;Steininger et al., 2022), especially in BI-enabled digital organisations (Philip, 2018). It

refers to the ability of enterprises to identify and leverage internal knowledge and to absorb

and transform external knowledge. Knowledge sharing within organisations promotes the

interchange of knowledge and resource reorganisation (Ghasemaghaei, 2019;Maoret,

Tortoriello and Iubatti, 2020). It is possible for organisation members to create new knowledge

through the cooperation of knowledge sharing (Eidizadeh, Salehzadeh and Chitsaz Esfahani,

2017). For external knowledge, organisations scan and absorb external data, information, and

knowledge and convert them into internal knowledge resources, which are then utilised and

transformed into business value (Duan, Cao and Edwards, 2020;Tang, Fang and Qualls, 2020).

This framework involves two significant concepts, namely, knowledge sharing, and

absorptive capability. Knowledge sharing refers to sharing and exchanging effective

knowledge, information, experience, and insights among organisational members (Haas and

Hansen, 2007). Knowledge sharing may take place across organisational levels and among

different organisational members through two different types of knowledge sharing, formal and

informal, for either explicit or tacit knowledge (Herschel and Jones, 2005). It is the capability

of organisations to disseminate and diffuse knowledge (Roberts et al., 2012). It could

effectively enhance the knowledge assets of an organisation and thus have an impact on the

performance of the organisation (Liao, Fei and Chen, 2007). Knowledge sharing among

organisation members can reduce the cost of acquiring knowledge resources, which is a

9
necessary condition to facilitate sustainable competitive advantage such as innovation (Ritala

et al., 2015).

Absorptive capability is the ability to identify, absorb, transform and apply knowledge

and then utilise it for business purposes and convert it into business value (Roberts et al.,

2012;Božič and Dimovski, 2019a). Absorption depends on knowledge assets, but the mere

possession of knowledge assets does not ensure that an organisation will be able to fully utilise

the knowledge and transform it into innovation, and different capabilities will lead to different

innovation performances (Miroshnychenko et al., 2020). Furthermore, absorptive capability

and innovative capability are two very related notions. Absorptive capability refers to the

organisation's absorption and application of external knowledge to internal business activities,

while innovation capability refers to leveraging organisational knowledge to create new

knowledge (Lichtenthaler and Lichtenthaler, 2009). The potential positive effect of absorptive

capability on innovation performance stems from the transformation and application of internal

and external knowledge, forming a dynamic knowledge flow (Pai, Chang and City, 2013).

3. Research model and hypotheses

The research model is shown in Figure 1. It is proposed that knowledge sharing and

absorptive capability mediate the effect of BI quality on innovation performance. Meanwhile,

the path between knowledge sharing and innovation performance is fully mediated by

absorptive capability based on the literature (Liao, Fei and Chen, 2007;Kang and Lee, 2017).

The subsections below expound on the hypotheses.

Figure 1 Research model.

10
3.1 System quality and information quality of BI towards innovation performance

The system quality and information quality of BI are both increase innovation

performance through data use and data-enabled innovation (Ahmad, Ahmad and Hashim,

2016). Data are part of production and operation factors in firms, such as labour, capital, and

management. Thus, effectively and efficiently utilising data is a necessary for promoting

innovation (Božič and Dimovski, 2019b). In a complex, competitive market environment,

managers need the analysis ability and flexibility of information systems for agile decision-

making and innovation (Srinivasan and Swink, 2018). Business intelligence and big data

analytics help organisations scan for environmental changes, absorb knowledge resources and

implement governance measures to facilitate innovation (Duan, Cao and Edwards,

2020;Mikalef et al., 2020).

BI provides a flexible integrated system to help firms analyse data more quickly and more

flexibly (IşıK, Jones and Sidorova, 2013), which can facilitate innovation performance by

efficiently utilising data (Jaklič, Grublješič and Popovič, 2018). System quality optimises the

decision-making process, and organisational structure improves the efficiency of management

decisions and advances the possibility of integrating data elements into innovation decisions,

which in turn improves the potential innovation performance of enterprises (Cui et al.,

2015;Chan and Lau, 2018). Information quality can provide managers with accurate data

information and a decision foundation in a more timely way, which is of great value for quickly

integrating data information and responding to fluctuations from the environment, thus

improving the efficiency and agility of management decisions and playing a positive role in

the effective use of external data resources for innovative decision-making (Rai and Tang,

11
2010;Park, El Sawy and Fiss, 2017). Thus:

H1: BI system quality has a positive effect on innovation performance;

H2: BI information quality has a positive effect on innovation performance.

3.2 Knowledge sharing, absorptive capability and innovation performance

BI-enabled enterprise innovation is essentially the transformation of an enterprise's data

resources first into knowledge resources and then into innovation resources (Hung et al.,

2010;Ghasemaghaei, 2019). Some studies have considered that data are difficult to understand

and utilise directly. From data to innovation, performance involves a complex process of data

assimilation into knowledge (Gunasekaran et al., 2017). Thus, knowledge, as a "processed"

information product, shares certain characteristics with data, such as being disseminated and

processed, comprehensibility and independence (Shollo and Galliers, 2016).

Knowledge sharing promotes knowledge interchange and dynamic resource allocation

within the enterprise. External digital information enters the organisation through BI and

analytics, which then extends the coverage of knowledge within the organisation through

knowledge sharing to form a dynamic flow of knowledge (Giménez-Figueroa, Martín-Rojas

and García-Morales, 2018). BI system quality has a potential facilitating effect on knowledge

sharing because BI system quality characteristics determine the quality of the process of

knowledge acquisition. Individuals of the company must utilise the knowledge provided by the

BI system to ask questions, provide answers or assist with problem-solving (Ho and Kuo, 2013).

Based on cross-departmental cooperation and the multiterminal online properties of BI

analytics, managers across departments are able to share their knowledge obtained through

external data (Hidalgo and Albors, 2008;Kyriakou and Loukis, 2019). Therefore:

12
H3: Greater BI system quality leverages a more positive effect on knowledge sharing.

The significant interdependence of information and knowledge contributes to a potential

positive relationship between information quality and knowledge sharing. Information is not

only the necessary condition of knowledge creation and utilisation but also the medium of

knowledge transfer (Taylor and Wright, 2004). Data come from outside the organisation and

will not be processed again because of the organisational preferences, which means that data-

based knowledge is more objective when facing an uncertain environment (Mikalef, Pateli and

van de Wetering, 2016). Accordingly, information quality motivates organisational members to

share knowledge at a lower cost of interchange with a more convenient model (Taylor and

Wright, 2004). Through BI, an information system infrastructure, employees are empowered

to participate in management decisions, which in turn provides individuals with the motivation

to share knowledge. In turn, the enhanced availability of information, whether based on formal

or informal means, allows the knowledge interchange process to take place in a more direct

and portable way (Cassiman and Valentini, 2016). Thus:

H4: Greater BI information quality leverages a more positive effect on knowledge sharing..

With the BI system, enterprises can effectively promote the potential of absorption

capability. Božič and Dimovski (2019a) argue that the external information identification and

data processing characteristics of BI significantly enhance the ability of an organisation to

absorb external information and external knowledge, which is one of the key paths of BI value

creation. Absorptive capability is a critical organisational capability for organisations to

continuously acquire external information as well as innovation resources and plays a role in

the dynamic capability to acquire external resources (Shollo and Galliers, 2016;Peters et al.,

13
2016). The resource acquisition aspect of absorptive capability overlaps greatly with the

application scenarios of BI and relies on the excellent performance of BI systems, which

enables organisations to improve their absorptive capability. This depends on the reliability of

business processes brought about by system quality improvement on the one hand and from

the high quality of external knowledge brought about by information quality improvement on

the other hand(Ali, Musawir and Ali, 2018). Therefore, BI application scenarios are highly

aligned with the enhancement of organisational absorptive capability; hence,

H5: Greater BI system quality leverages a more positive effect on absorptive capability;

H6: Greater BI information quality leverages a more positive effect on absorptive capability.

Knowledge sharing promotes absorptive capability (Haas and Hansen, 2007;Zhou and Li,

2012). Due to the knowledge sharing mechanism of the organisation, members can acquire

external knowledge through knowledge sharing in addition to acquiring data-based knowledge

directly through the BI system. Knowledge sharing can take place formally through regular

project meetings, work summaries, business processes, or informally. A portion of nonbusiness

data analysis employees and managers can easily acquire external knowledge and skills

through knowledge sharing, which, in turn, enhances the organisation's absorptive capability

(Giménez-Figueroa, Martín-Rojas and García-Morales, 2018). The knowledge sharing

mechanism enables the integration and reconfiguration of knowledge resources within the

organisation, creates an organisational learning environment (Tang, Fang and Qualls, 2020),

and promotes organisational knowledge absorptive capability, leading to a dynamic allocation

of resources to the organisation's knowledge capabilities (Wang and Wang, 2008). Therefore,

H7: Greater knowledge sharing leverages a more positive effect on absorptive capability .

14
 Absorptive capability is seen as a necessary condition for data-driven and knowledge-

driven innovation, particularly regarding the absorption of external knowledge (Chen, Lin and

Chang, 2009;Kang and Lee, 2017). Innovation is generally considered to be the offering of

new products and services to consumers or vendors, or a modification of production processes

or service approaches (Baregheh, Rowley and Sambrook, 2009;Francis and Bessant, 2005).

Sustained innovation is regarded as the core driver of a firm's competitive advantage in the

long run, and it is influenced by two factors: the flexibility of the organisation to adjust its

innovation strategy (Mikalef, Pateli and van de Wetering, 2016), and managers’ continuous

innovative decision-making (Eidizadeh, Salehzadeh and Chitsaz Esfahani, 2017). These two

factors require organisations to be able to keep abreast of changing trends in external markets,

i.e., to absorb external ideas for innovation (Duan, Cao and Edwards, 2020). Thus, a higher

absorptive capability promotes the absorption and utilisation of external knowledge, which is

one of the main sources of innovative ideas for both organisations and individuals

(Miroshnychenko et al., 2020). Thus, absorptive capability enables organisations to more

accurately analyse internal and external knowledge based on BI and respond to it in terms of

both innovation activities and strategies, thus contributing to the overall innovation

performance of the company. Therefore,

H8: Greater absorptive capability. leverages a more positive effect on innovation performance.

3.3 Mediation effect of knowledge sharing and absorptive capability

BI quality is closely related to the organisation's dynamic knowledge capabilities to share,

configure, absorb, and utilise external data resources through knowledge sharing and

absorption capability. This mechanism effectively expands the organisation's knowledge

15
resource base, allowing the internalisation of external data resources into the organisation's

internal knowledge and innovation resources, which in turn enhances the firm's innovation

performance (Cassiman and Valentini, 2016). Thus, there is a potential mediating path between

BI quality and firm innovation performance in knowledge sharing and assimilation capabilities.

Absorptive capability is necessary for innovation performance because absorptive

capability emphasises absorbing and applying knowledge resources to business value (Kang

and Lee, 2017). However, knowledge sharing is not necessary for innovation performance

because knowledge sharing focuses on integrating and allocating resources. Many empirical

studies show that absorptive capability fully mediate the positive association between

knowledge sharing and innovation performance (Liao, Fei and Chen, 2007;Ali, Musawir and

Ali, 2018). Therefore, based on the theoretical analysis and Hypotheses H5, H6, and H8, BI

quality enhances innovation performance by promoting organisational absorptive capability

and, in turn, innovation performance (Pai, Chang and City, 2013). System quality enhances the

organisation's ability to absorb and utilise external knowledge through digital business

processes and then transforms such resources into innovation performance. Information quality

enables organisations to efficiently absorb external knowledge through objective and effective

data and information, which reduces potential screening and decision-making costs and thus

enhances innovation performance. Therefore, it is assumed that absorptive capability plays a

mediating role in the path:

H9a: A higher level of organisational absorptive capability leads to a stronger relationship

between BI system quality and innovation performance;

H9b: A higher level of organisational absorptive capability leads to a stronger relationship

16
between BI information quality and innovation performance.

The literature has highlighted the full mediating role of absorptive capability in the path

of knowledge sharing on innovation performance. It is assumed that BI quality effectively

improves formal or informal knowledge sharing, thereby promoting absorptive capability.

Higher system quality brings a greater willingness to utilise BI for data analysis and then to

diffuse knowledge to a larger group through knowledge sharing (Ho and Kuo, 2013). Then,

with absorption, transformation, and application, firms are more likely to engage in innovation

activities efficiently, which ultimately improves innovation performance. Similarly, higher

information quality leads to a better quality of data output, brings a higher quality of knowledge

through BI output and knowledge sharing, and eventually forms better innovation performance

of the organisation (Taylor and Wright, 2004). Thus:

H10a: Knowledge sharing and absorptive capability mediate the path from BI system quality

to innovation performance;

H10b: Knowledge sharing and absorptive capability mediate the path from BI information

quality to innovation performance.

4. Methodology

This paper adopted a questionnaire survey for data collection and conducted an empirical

study to test the research hypotheses and model. Structural equation modelling (SEM) was

used to examine the path coefficients, and the robust method with bootstrapping was applied

to estimate the indirect coefficients, standard errors, and mediation effects. The heterogeneity

among industries with different potential IT foundations was estimated by the multigroup SEM

method. The lavaan package (Rosseel, 2012) and semTools package (Jorgensen et al., 2020)

17
were adopted for data analysis within the R Statistical Environment (R Core Team, 2020, v.

4.0.2). Many studies have shown that SEM and multigroup SEM methods using the lavaan and

semTools packages are robust and reliable compared to software such as AMOS, smartPLS,

and Mplus (Narayanan, 2012;Svetina, Rutkowski and Rutkowski, 2020).

4.1 Sample design

The measurement items were adapted from previously validated scales in the literature

with a 7-point Likert scale for Chinese firms. This research was conducted by business analysts

and managers in Chinese companies since they use BI systems directly for data analysis,

visualisation, and decision-making. There are three preliminary questions in the questionnaire,

and participants were required to meet at least one criteria: (a) Now working in the BI analysis

department; (b) Have previous work experience in BI analysis; or (c) Now are managers in the

companies. The survey was targeted at larger, innovative companies. Larger companies are

better able to deploy sophisticated BI tools, and innovative companies make it easier for us to

collect data on innovation performance.

Since the participants were all from Chinese companies, to ensure the accuracy and

comprehensibility of the scale, the items were translated and localised based on the English

scale, and the experts with BI/IS experience were invited to participate in the translation and

modification. The questionnaire was finally developed and distributed in Chinese to avoid the

interference of language factors for the participants. To ensure the validity and reliability of the

questionnaire, a prestudy was conducted, and some questions were adjusted and revised

according to the analysis results of the prestudy.

Trap questions were set in the questionnaire to prevent participants from answering

18
randomly or indiscriminately. Finally, 621 participants took part in the survey. After screening

out the questionnaires with inadequate responses and those with poor response quality, 352

questionnaires were suitable for further analysis, with a response rate of 56.7%. The

demographic characteristics of the participants is shown in Table 1.

During the survey, participants were not informed of the specific research and survey

content to prevent potential common method bias. This paper applied Harman's single-factor

test to estimate common method bias in the phase of data analysis. It showed that the percentage

of unrotated variance was 34.99%, which was lower than 40%. Nonresponse bias was also

tested in different periods, and the results showed that there was no significant difference

between the early and late participants; thus, it could be considered that there was no non-

response bias (Armstrong and Overton, 1977).

Table 1 Sample demographics

4.2 Measurement

This article used a questionnaire with validated scales in the literature (see Appendix A).

System quality was measured through four items and information quality was measured

through five items, which were adapted from research among BI/IS success literature (Nelson,

Todd and Wixom, 2005;Yeoh and Popovič, 2016;Kulkarni, Robles-Flores and Popovič,

2017;Ahn and Sura, 2020). Knowledge sharing was measured through four items (Kang and

Lee, 2017;Mola et al., 2020), and absorptive capability was measured through four items

(Rouhani et al., 2018;Božič and Dimovski, 2019b), adapted from the knowledge management

literature in IS. Innovation performance was measured through four items for the performance

19
measurement of new products, new services, new markets, etc. (Cui et al.,

2015;Miroshnychenko et al., 2020).

There were also four control variables considered in this study: firm size, sales revenue,

industry, and ownership (see Appendix A). Large and high-revenue companies, which usually

have a greater demand for mature BI systems, may have an impact on the performance of BI-

enabled innovation; therefore, this paper considered these two control variables. These two

variables are measured by the number of employees and sales revenue. There is evidence that

Chinese firms differ in their potential willingness to innovate due to different types of

ownership, so this paper controls for the ownership variable. Finally, industry control for the

effect of different industries on the model and in the multigroup analysis according to industries

was removed to avoid unrecognisable multicollinearity.

5. Results
5.1 Measurement model

First, the goodness-of-fit of the five-dimensional model was tested through confirmatory

factor analysis (CFA), and the result, which also contains the comparison of the 1-to 4-

dimensional model, is shown in Table 2. The fit of the five-dimensional model was significantly

better than that of the other models (Δχ2 significant). Both the RMSEA and CFI of the five-

dimensional model were significantly better than those of the other models (RMSEA < 0.05,

CFI > 0.9).

Table 2 Model competition results of CFA

Then, the reliability and validity of the scale were examined through indicators of α, CR,

20
and AVE (Ringle, Sarstedt and Straub, 2012). Table 3 shows the factor loading, Cronbach'α,

and CR. Most items had loadings above 0.7, with only one item having a slightly lower value

of 0.665. Additionally, the Cronbach's α and CR for all five constructions were above 0.7,

indicating that the internal consistency of the measurement constructs is reliable (Bagozzi and

Yi, 1988).

Table 3 Factor Loading, Cronbach'α, and CR

Then, Table 4 shows the correlation coefficient matrix, average variance extracted (AVE),

and the square root of AVE (shown in bold on the diagonal). Convergent validity measures the

aggregation of different indicators within the same factor. It is generally accepted to have good

convergent validity when factor loadings and AVEs are greater than 0.5 (Fornell and Larcker,

1981). All five constructions’ AVE exceeded the threshold of 0.5, indicating effective

convergence. Discriminant validity is the degree to which a given construct distinguishes itself

from all other constructions and is usually measured by the square root of AVE with the

magnitude of the correlation coefficient (Fornell and Larcker, 1981). The diagonal line of Table

4 shows the square root of AVE for each construction, the correlation among the latent variables

is lower than the square root of AVE, and the coefficients are all at low levels, which can be

considered valid for discrimination.

Table 4 Correlations matrix, AVE, square root of AVE

5.2 Structural model

There are multiple possible solutions for SEM; hence, two competitive models were

21
constructed as contrasting models to the standard model. Model A is the partial mediation

model, competitive Model B is the completely mediated model, and competitive Model C is

the direct effect model. The results are shown in Table 5. Both Model A and Model B fit

significantly better than Model C (Δχ2 is significant). For the standard model (Model A) and

completely mediated model (Model B), all the fitted indices of RMSEA, DFI, GFI, NFI were

consistent, and the Δχ2 was insignificant. Therefore, Model B was better according to the

parsimonious principle. However, to examine H1 and H2, both Model A and Model B are

estimated in this paper.

Table 5 Model competition results

The structural model is summarised in Figure 2 and Table 6, which shows the standardised

results of path coefficients obtained by performing a bootstrap analysis with 5,000 resamples

and hypothesis tests. The direct effects of both the system (β=0.038, p>0.05) and information

quality (β=0.045, p>0.05) on innovation performance are not significant, thus rejecting H1 and

H2. BI quality has a positive effect on knowledge sharing, but the effect of system quality

(β=0.218, p<0.001) is slightly less than that of information quality (β=0.332, p<0.001). BI

system quality was found to have an impact on absorptive capability (β=0.266, p<0.001). In

contrast, no direct significant effect was found between information quality and absorptive

capability (β=0.034, p>0.05). Additionally, knowledge sharing significantly improves

absorptive capability (β=0.540, p<0.001), while there is a significant promotion effect of

absorptive capability on innovation performance (β=0.439, p<0.001), so H7 and H8 are

supported. None of the control variables had a significant direct effect on innovation

22
performance, with positive but nonsignificant coefficients for firm size and sales revenue, and

a negative nonsignificant effect for ownership.

Figure 2 Estimated relationships of the structural model

Table 6 Summary of the SEM path coefficients

5.3 Test for mediation

To examine the mediating effect of knowledge sharing and absorptive capability between

BI quality and innovation performance, a bootstrapping approach with resampling of 5000

subsamples was employed to calculate the standard error and the mediation effect. The results

summary is shown in Table 7. First, the path BISQ→AC→IP is confirmed to be significant,

but the direct effect of system quality on innovation performance is insignificant, which

indicates a fully mediated effect. The direct and indirect effects of BIIQ→AC→IP are both

insignificant, demonstrating no mediating effect and H9b is not supported. Furthermore, the

mediated path BISQ→KS→AC→IP is significant (β=0.052, p<0.01), and the direct effect from

system quality to absorptive capability is also significant (β=0.266, p<0.001), which indicates

partial mediation. Finally, the path BIIQ→KS→AC→IP is significant (β=0.079, p<0.001), and

both the direct effect from information quality to absorptive capability and innovation

performance are insignificant, meaning full mediation.

Table 7 Summary of mediation results and hypotheses

5.4 Post hoc multigroup analysis

As discussed, the results of the structural model show the different mediated paths of

23
BISQ and BIIQ on innovation performance. To further understand the heterogeneity from

potential differences in IT capability due to differences in IT infrastructure and staff in different

industries, a further multigroup analysis was examined to enhance the understanding of the

model. With industries as group variables, traditional manufacturing, service, sales, and

transportation industries were categorised as industries with low potential IT capability, and

internet companies, information technology services, and financial services were categorised

as industries with high potential IT capability. Table 8 shows the standardised results of

multigroup SEM coefficients with the 5,000 resamples bootstrap approach.

Table 8 Summary of the multigroup SEM path coefficients

The result of the multigroup analysis shows that for industries with high potential IT

capability, the path coefficients and mediation effects are essentially the same as those of

baseline model. For industries with low potential IT capability, unlike the baseline model, there

is no significant facilitative effect of system quality on knowledge sharing (β=0.173, p>0.05),

the partial mediation effect of knowledge sharing is also insignificant (β=0.044, p>0.05), and

absorptive capability completely mediates the effect of system quality on innovation

performance. For the control variables, there is a significant positive effect of firm size on

innovation performance (β=0.209, p<0.05).

6. Discussion

In the era of big data, companies use BI systems as a solution for decision-making and

firm innovation. However, some companies do not improve innovation performance through

BI (Puklavec, Oliveira and Popovič, 2018). The literature focuses more on BI use rather than

24
BI quality, which ignores the significant role of the coupling of technology and organisational

capabilities. This study addresses the above gap by utilising the BI success and the knowledge

absorption of dynamic capabilities theory as a theoretical framework to examine the role of BI

quality on innovation performance (Conboy et al., 2020;Steininger et al., 2022). Furthermore,

heterogeneity among the industries with different potential IT capabilities was explored.

The findings confirmed the enhancement of the system and information quality of BI on

firms' innovation performance, but the effects of these two variables differed significantly. The

literature generally supports the argument that the use of BI or big data analytics is beneficial

for firm innovation (Duan, Cao and Edwards, 2020;Božič and Dimovski, 2019b;Mikalef et al.,

2020), and this idea is further reinforced in this paper, based on the perspective of BI quality

and dynamic capabilities. The difference in motivating firms to innovate here is further found

from the perspective of system quality and information quality. Interestingly, the effect of

information quality on knowledge sharing is significantly higher than the effect of system

quality, which means that the relationship between information quality and innovation

performance is more sensitive to knowledge sharing (Ahn and Sura, 2020). Correspondingly,

the relationship between system quality and innovation performance is more sensitive to

absorptive capability (Chan and Lau, 2018). Moreover, the contribution of system quality to

knowledge sharing is not significant in industries with lower potential IT capability, which

notes that absorptive capability is more sensitive.

To summarise, this is the first study that explored the paths of BI quality on innovation

performance based on the perspective of dynamic capabilities. Based on the integration of

business data analysis with knowledge capabilities, the knowledge absorption of dynamic

25
capabilities is applied here to frame and explain why BI quality is beneficial to innovation

performance. The findings indicate complete mediation of knowledge sharing and absorptive

capability. Absorptive capability completely mediates the effect of system quality on

innovation performance, while knowledge sharing is partially mediated. Moreover, the result

also shows that the mediation of knowledge sharing on the relationship between system quality

and innovation performance is insignificant for industries with lower potential IT capability

(Prahalad and Krishnan, 2002;Park, El Sawy and Fiss, 2017). The framework and findings

provide novel and valuable insights for related domains and further research.

6.1 Theoretical contributions

First, the findings of this paper extend the view of dynamic capabilities in BI. Dynamic

capability theory is an effective theoretical tool for linking BI, organisational capabilities, and

innovation (Conboy et al., 2020;Steininger et al., 2022), and this paper reinforces this view

particularly from a knowledge-based perspective on resource sensing (Mikalef, Pateli and van

de Wetering, 2016;Duan, Cao and Edwards, 2020), seizing (Tang, Fang and Qualls, 2020;Chen

and Siau, 2020), and transforming (Karimi and Walter, 2015;Miroshnychenko et al., 2020).

The framework is constructed through knowledge sharing and absorptive capability, and the

effects on innovation are further explored (Eidizadeh, Salehzadeh and Chitsaz Esfahani,

2017;Giménez-Figueroa, Martín-Rojas and García-Morales, 2018). Knowledge assets

facilitate the role of BI in innovation (Karimi and Walter, 2015;Philip, 2018), but system and

information quality have different effects on knowledge sharing and absorptive capability

(Eidizadeh, Salehzadeh and Chitsaz Esfahani, 2017;Giménez-Figueroa, Martín-Rojas and

García-Morales, 2018). In the relationship between system quality and innovation performance,

26
absorptive capability is more sensitive than knowledge sharing, and the contribution of system

quality to knowledge sharing is not significant in industries with low potential IT capability. In

contrast, the sensitivity of knowledge sharing was higher than that of absorptive capability in

the relationship between information quality and innovation performance (Lichtenthaler and

Lichtenthaler, 2009;Zhou and Li, 2012;Kang and Lee, 2017). The results of this study are an

important step in understanding the effect of absorptive capability of knowledge in enhancing

BI quality in innovation performance.

Second, building on the work of Božič and Dimovski (2019b);Mikalef et al. (2019);Duan,

Cao and Edwards (2020);and Mikalef et al. (2020) on big data analytics in business to advance

innovation, this paper further strengthens the theory of BI quality and innovation performance

and provides a theoretical contribution to total quality management (Ravichandran and Rai,

1999;Gorla, Somers and Wong, 2010). System quality and information quality, the key

dimensions of BI total quality management, have proven to be effective in combining dynamic

capabilities to promote innovation performance (Ravichandran and Rai, 1999). At the same

time, BI quality should be synergistic with dynamic capabilities to promote innovation

performance. If the organisation lacks knowledge sharing, then it is difficult to promote

innovation performance even if the information quality is improved. Similarly, if an

organisation lacks absorptive capability, then even higher system quality will not enhance

innovation performance.

Third, this study provided a new theoretical perspective on the BI success model of BI.

This research focused on the concepts of system quality and information quality in the CSFs

and applied these two exogenous variables to firms' innovation performance (DeLone and

27
McLean, 1992;Yeoh and Koronios, 2010;Popovič et al., 2012). Moreover, this paper treats BI

quality as organisational innovation and an exogenous variable at the organisational level rather

than at the individual level. Hence the impact of BI quality on information use is not considered,

which has already been validated in many studies (Olszak, 2016;Yeoh and Popovič, 2016). The

results of the study confirmed that there are differences in the paths of system quality and

information quality on innovation performance and that such differences are more significant

across industries. Future research can extend the findings of this study based on the success

model by discussing the differential effects of system quality and information quality of BI on

other business values (e.g., financial performance, strategic agility).

6.2 Practical contribution

Assessing the path differentiation of system quality and information quality in BI on

innovation performance gives companies a new perspective on the use of BI. From a practical

perspective, this may be due to the difference in the use of BI between data analysis

departments and decision-making departments, with data analysis departments preferring first-

hand information and decision-making departments preferring second-hand information, thus

leading to high sensitivity of information quality to knowledge sharing. However, at the same

time, this paper finds that a higher quality of information output does not directly contribute to

absorptive capability. This paper suggests that this could be because management has barriers

to the absorption of the primary information output of BI, preferring to obtain secondary

information for their knowledge source through knowledge sharing, and they are not sensitive

to the information quality of BI. However, higher system quality can make all employees and

managers in the organisation more inclined to acquire, share, absorb and utilise knowledge

28
through BI to improve innovation performance. Therefore, if knowledge is not shared

sufficiently, the utility of information quality for innovation performance will be reduced, and

this reduction will be higher than the utility of system quality.

The results also indicated that the system quality of BI in labour-intensive manufacturing

and service industries with a low level of information technology did not have a significant

impact on knowledge sharing. The low level of information technology in these industries

implies that the employees and managers of companies have poorer knowledge of information

systems and information workflows. In business practice, the data analysis departments and

management departments in these industries are relatively inefficient in using BI for work and

decision-making, and employees do not fully utilise BI for visualisation and analysis. Therefore,

the impact of system quality on knowledge sharing is not significant. In contrast, system quality

has a significant enhancement effect on absorptive capability. In terms of the dynamic

capability of knowledge, knowledge sharing can promote absorptive capability and thus firm

innovation. Therefore, the research infers that the main limitation of firms with low information

technology levels is the use of information rather than the use of IS. They have difficulty

effectively translating the information output of BI into intraorganisational knowledge to be

utilised. In other words, for these industries, an overall improvement in the information

utilisation and knowledge-sharing capabilities of employees and managers can effectively

promote the use of BI and data analysis by employees and managers, which in turn can

transform knowledge into innovation resources through absorptive capability and promote the

innovation performance of the firms (Melville, Kraemer and Gurbaxani, 2004).

To summarise, the findings of the study emphasise two points. The first is that firms

29
should focus on the importance of knowledge sharing and absorptive capability, which are

important for firms to leverage the value of business intelligence and facilitate innovative

performance. Additionally, when the resources of firms are limited, they should combine their

organisational and technological characteristics and invest their resources in departments that

are more conducive to enhancing their innovation performance based on the high sensitivity of

knowledge sharing to information quality and the high sensitivity of absorptive capability to

system quality. These findings contribute to a stronger understanding of what is necessary for

enterprises to adopt BI for innovation.

6.3 Limitations and future research

There are several limitations of the research. First, this paper has used various methods to

minimise the potential common method bias as much as possible during the questionnaire

design and distribution process, and the results of the statistical tests provide evidence that

common method bias is within an acceptable range. However, future research should try to

avoid potential common method bias through some exante methods at the research design stage,

such as collecting questionnaire data through different information sources and collecting data

at different time points, and other methods. Second, BI, as a strategic tool for business

management, is a long-term strategy for corporate information systems. That is, the impact of

BI on innovation performance may require longer-term tracking and observation. The research

design and research data in this paper are derived from short-term data, further research could

extend the extra findings through a longitudinal perspective. Finally, this study may have

omitted variables. Future research can focus on the possible omitted variables in this study to

construct a new research framework as a way to provide new insights into the impact of BI and

30
corporate innovation.

7. Conclusion

This study provides a new understanding of BI quality, BI success, and innovation

performance. The study explores the differential impact of BI system quality and information

quality on firms' innovation performance through a structural model mediated by knowledge

sharing and absorptive capability and highlights heterogeneous industries with different

potential IT capabilities. Overall, the results provide insights for managers on how BI quality

can promote innovation performance. The findings show: (a) there are two facilitating paths of

system quality on innovation performance, i.e., fully mediated by absorptive capability and

partially mediated by knowledge sharing; (b) there is only one positive path for information

quality on innovation performance which is completely mediated by knowledge sharing and

absorptive capability; and (c) for traditional manufacturing and service industries with low

potential IT capability, system quality cannot improve knowledge sharing significantly, and a

partial mediation effect does not exist. In general, there is a positive correlation between BI

quality and innovation performance, but managers need to target organisational knowledge

sharing and absorptive capability as an essential approach to improve innovation performance.

References

Ahmad, A., Ahmad, R. and Hashim, K.F. (2016) ‘Innovation Traits for Business Intelligence Successful

Deployment’. Journal of Theoretical & Applied Information Technology, 89 (1), pp. 96-107.

Ahn, J. and Sura, S. (2020) ‘The effect of information quality on Social Networking Site (SNS)-Based

commerce: From the perspective of Malaysian SNS users’. Journal of Organizational and End User

Computing, 32 (1), pp. 1-18.

31
Ain, N. et al. (2019) ‘Two decades of research on business intelligence system adoption, utilization and

success——A systematic literature review’. Decision Support Systems, 125 1-13.

Ali, I., Musawir, A.U. and Ali, M. (2018) ‘Impact of knowledge sharing and absorptive capacity on

project performance: The moderating role of social processes’. Journal of Knowledge Management,

22 (2), pp. 453-477.

Armstrong, J.S. and Overton, T.S. (1977) ‘Estimating nonresponse bias in mail surveys’. Journal of

Marketing Research, 14 (3), pp. 396-402.

Au, N., Ngai, E.W. and Cheng, T.E. (2002) ‘A critical review of end-user information system

satisfaction research and a new research framework’. Omega, 30 (6), pp. 451-478.

Bagozzi, R.P. and Yi, Y. (1988) ‘On the evaluation of structural equation models’. Journal of the

Academy of Marketing Science, 16 (1), pp. 74-94.

Baregheh, A., Rowley, J. and Sambrook, S. (2009) ‘Towards a multidisciplinary definition of

innovation’. Management Decision, 47 (8), pp. 1323-1339.

Barreto, I. (2010) ‘Dynamic capabilities: A review of past research and an agenda for the future’.

Journal of Management, 36 (1), pp. 256-280.

Bernardino, J. and Neves, P.C. (2016) 'Decision-making with big data using open source business

intelligence systems'. Human Development and Interaction in the Age of Ubiquitous Technology.

Hershey: IGI Global, pp. 120-147.

Božič, K. and Dimovski, V. (2019a) ‘Business intelligence and analytics for value creation: The role of

absorptive capacity’. International Journal of Information Management, 46 93-103.

Božič, K. and Dimovski, V. (2019b) ‘Business intelligence and analytics use, innovation ambidexterity,

and firm performance: A dynamic capabilities perspective’. The Journal of Strategic Information

Systems, 28 (4), pp. 1-20.

Caseiro, N. and Coelho, A. (2019) ‘The influence of Business Intelligence capacity, network learning

and innovativeness on startups performance’. Journal of Innovation & Knowledge, 4 (3), pp. 139-

145.

Cassiman, B. and Valentini, G. (2016) ‘Open innovation: Are inbound and outbound knowledge flows

really complementary?’. Strategic Management Journal, 37 (6), pp. 1034-1046.

Chan, L.-K. and Lau, P.-Y. (2018) ‘Investigating the Impact of System Quality on Service-Oriented

Business Intelligence Architecture’. SAGE Open, 8 (4), pp. 1-14.

32
Chen, H., Chiang, R.H. and Storey, V.C. (2012) ‘Business intelligence and analytics: From big data to

big impact’. MIS Quarterly, 36 (4), pp. 1165-1188.

Chen, X. and Siau, K. (2020) ‘Business analytics/business intelligence and IT infrastructure: impact on

organizational agility’. Journal of Organizational and End User Computing (JOEUC), 32 (4), pp.

138-161.

Chen, Y.-S., Lin, M.-J.J. and Chang, C.-H. (2009) ‘The positive effects of relationship learning and

absorptive capacity on innovation performance and competitive advantage in industrial markets’.

Industrial Marketing Management, 38 (2), pp. 152-158.

Conboy, K. et al. (2020) ‘Using business analytics to enhance dynamic capabilities in operations

research: A case analysis and research agenda’. European Journal of Operational Research, 281

(3), pp. 656-672.

Cui, T. et al. (2020) ‘Managing Knowledge Distance: IT-Enabled Inter-Firm Knowledge Capabilities

in Collaborative Innovation’. Journal of Management Information Systems, 37 (1), pp. 217-250.

Cui, T. et al. (2015) ‘Information technology and open innovation: A strategic alignment perspective’.

Information & Management, 52 (3), pp. 348-358.

DeLone, W.H. and McLean, E.R. (1992) ‘Information systems success: The quest for the dependent

variable’. Information Systems Research, 3 (1), pp. 60-95.

DeLone, W.H. and McLean, E.R. (2003) ‘The DeLone and McLean model of information systems

success: a ten-year update’. Journal of Management Information Systems, 19 (4), pp. 9-30.

Dewan, S. and Min, C.-k. (1997) ‘The substitution of information technology for other factors of

production: A firm level analysis’. Management Science, 43 (12), pp. 1660-1675.

Duan, Y., Cao, G. and Edwards, J.S. (2020) ‘Understanding the impact of business analytics on

innovation’. European Journal of Operational Research, 281 (3), pp. 673-686.

Eidizadeh, R., Salehzadeh, R. and Chitsaz Esfahani, A. (2017) ‘Analysing the role of business

intelligence, knowledge sharing and organisational innovation on gaining competitive advantage’.

Journal of Workplace Learning, 29 (4), pp. 250-267.

Fornell, C. and Larcker, D.F. (1981) ‘Evaluating structural equation models with unobservable variables

and measurement error’. Journal of Marketing Research, 18 (1), pp. 39-50.

Francis, D. and Bessant, J. (2005) ‘Targeting innovation and implications for capability development’.

Technovation, 25 (3), pp. 171-183.

33
Ghasemaghaei, M. (2019) ‘Does data analytics use improve firm decision making quality? The role of

knowledge sharing and data analytics competency’. Decision Support Systems, 120 14-24.

Ghasemaghaei, M. and Calic, G. (2020) ‘Assessing the impact of big data on firm innovation

performance: Big data is not always better data’. Journal of Business Research, 108 147-162.

Giménez-Figueroa, R., Martín-Rojas, R. and García-Morales, V.J. (2018) 'Business intelligence: An

innovative technological way to influence corporate entrepreneurship'. Entrepreneurship:

Development Tendencies and Empirical Approach. Rijeka: InTech, pp. 113-130.

Gorla, N., Somers, T.M. and Wong, B. (2010) ‘Organizational impact of system quality, information

quality, and service quality’. The Journal of Strategic Information Systems, 19 (3), pp. 207-228.

Gunasekaran, A. et al. (2017) ‘Big data and predictive analytics for supply chain and organizational

performance’. Journal of Business Research, 70 308-317.

Haas, M.R. and Hansen, M.T. (2007) ‘Different knowledge, different benefits: Toward a productivity

perspective on knowledge sharing in organizations’. Strategic Management Journal, 28 (11), pp.

1133-1153.

Herschel, R.T. and Jones, N.E. (2005) ‘Knowledge management and business intelligence: the

importance of integration’. Journal of Knowledge Management, 9 (4), pp. 45-55.

Hidalgo, A. and Albors, J. (2008) ‘Innovation management techniques and tools: a review from theory

and practice’. R&D Management, 38 (2), pp. 113-127.

Ho, L.A. and Kuo, T.H. (2013) ‘How system quality and incentive affect knowledge sharing’. Industrial

Management & Data Systems, 113 (7), pp. 1048-1063.

Hung, S.-Y. et al. (2010) ‘Critical factors of hospital adoption on CRM system: Organizational and

information system perspectives’. Decision Support Systems, 48 (4), pp. 592-603.

IşıK, Ö., Jones, M.C. and Sidorova, A. (2013) ‘Business intelligence success: The roles of BI

capabilities and decision environments’. Information & Management, 50 (1), pp. 13-23.

Jaklič, J., Grublješič, T. and Popovič, A. (2018) ‘The role of compatibility in predicting business

intelligence and analytics use intentions’. International Journal of Information Management, 43

305-318.

Jorgensen, T.D. et al. (2020) 'semTools: Useful tools for structural equation modeling'. R package

version 0.5-3, pp. doi.

Kang, M. and Lee, M.-J. (2017) ‘Absorptive capacity, knowledge sharing, and innovative behaviour of

34
 R&D employees’. Technology Analysis & Strategic Management, 29 (2), pp. 219-232.

Karimi, J. and Walter, Z. (2015) ‘The role of dynamic capabilities in responding to digital disruption: A

factor-based study of the newspaper industry’. Journal of Management Information Systems, 32 (1),

pp. 39-81.

Kranz, J. (2021) ‘Strategic innovation in IT outsourcing: Exploring the differential and interaction

effects of contractual and relational governance mechanisms’. The Journal of Strategic Information

Systems, 30 (1), pp. 101656.

Kulkarni, U.R., Robles-Flores, J.A. and Popovič, A. (2017) ‘Business intelligence capability: the effect

of top management and the mediating roles of user participation and analytical decision making

orientation’. Journal of the Association for Information Systems, 18 (7), pp. 516.

Kyriakou, N. and Loukis, E.N. (2019) ‘Do strategy, processes, personnel and technology affect firm’s

propensity to adopt cloud computing?’. Journal of Enterprise Information Management, 32 (3), pp.

517-534.

Lane, P.J., Koka, B.R. and Pathak, S. (2006) ‘The reification of absorptive capacity: A critical review

and rejuvenation of the construct’. Academy of Management Review, 31 (4), pp. 833-863.

Liao, S.-H., Fei, W.-C. and Chen, C.-C. (2007) ‘Knowledge sharing, absorptive capacity, and innovation

capability: an empirical study of Taiwan's knowledge-intensive industries’. Journal of Information

Systems, 33 (3), pp. 340-359.

Lichtenthaler, U. and Lichtenthaler, E. (2009) ‘A capability‐based framework for open innovation:

Complementing absorptive capacity’. Journal of Management Studies, 46 (8), pp. 1315-1338.

Maoret, M., Tortoriello, M. and Iubatti, D. (2020) ‘Big fish, big pond? The joint effect of formal and

informal core/periphery positions on the generation of incremental innovations’. Organization

Science, 31 (6), pp. 1538-1559.

Melville, N., Kraemer, K. and Gurbaxani, V. (2004) ‘Information technology and organizational

performance: An integrative model of IT business value’. MIS Quarterly, 28 (2), pp. 283-322.

Mikalef, P. et al. (2019) ‘Big data analytics capabilities and innovation: the mediating role of dynamic

capabilities and moderating effect of the environment’. British Journal of Management, 30 (2), pp.

272-298.

Mikalef, P. et al. (2020) ‘The role of information governance in big data analytics driven innovation’.

Information & Management, 57 (7), pp. 1-15.

35
Mikalef, P., Pateli, A. and van de Wetering, R. (2016) 'IT flexibility and competitive performance: The

mediating role of IT-enabled dynamic capabilities'. Twenty-Fourth European Conference on

Information Systems (ECIS), 2016 Istanbul, Turkey.

Miroshnychenko, I. et al. (2020) ‘Absorptive capacity, strategic flexibility, and business model

innovation: Empirical evidence from Italian SMEs’. Journal of Business Research, 130 670-682.

Mola, L. et al. (2020) ‘Business intelligence system design and its consequences for knowledge sharing,

collaboration, and decision-making: an exploratory study’. International Journal of Technology and

Human Interaction, 11 (4), pp. 1-25.

Narayanan, A. (2012) ‘A review of eight software packages for structural equation modeling’. The

American Statistician, 66 (2), pp. 129-138.

Nelson, R.R., Todd, P.A. and Wixom, B.H. (2005) ‘Antecedents of information and system quality: an

empirical examination within the context of data warehousing’. Journal of Management

Information Systems, 21 (4), pp. 199-235.

Olszak, C.M. (2016) ‘Toward better understanding and use of Business Intelligence in organizations’.

Information Systems Management, 33 (2), pp. 105-123.

Pai, F.-Y., Chang, H.-F. and City, H. (2013) ‘The effects of knowledge sharing and absorption on

organizational innovation performance–A dynamic capabilities perspective’. Interdisciplinary

Journal of Information, Knowledge, and Management, 8 (0), pp. 83-97.

Park, Y., El Sawy, O.A. and Fiss, P. (2017) ‘The role of business intelligence and communication

technologies in organizational agility: A configurational approach’. Journal of the Association for

Information Systems, 18 (9), pp. 648 – 686.

Peters, M.D. et al. (2016) ‘Business intelligence systems use in performance measurement capabilities:

Implications for enhanced competitive advantage’. International Journal of Accounting

Information Systems, 21 1-17.

Philip, J. (2018) ‘An application of the dynamic knowledge creation model in big data’. Technology in

Society, 54 (0), pp. 120-127.

Popovič, A. et al. (2012) ‘Towards business intelligence systems success: Effects of maturity and

culture on analytical decision making’. Decision Support Systems, 54 729-739.

Popovič, A. et al. (2014) ‘How information-sharing values influence the use of information systems:

An investigation in the business intelligence systems context’. The Journal of Strategic Information

36
 Systems, 23 (4), pp. 270-283.

Prahalad, C.K. and Krishnan, M.S. (2002) ‘The dynamic synchronization of strategy and information

technology’. MIT Sloan management review, 43 (4), pp. 24.

Puklavec, B., Oliveira, T. and Popovič, A. (2018) ‘Understanding the determinants of business

intelligence system adoption stages: An empirical study of SMEs’. Industrial Management & Data

Systems, 118 (1), pp. 236-261.

Rai, A. and Tang, X. (2010) ‘Leveraging IT capabilities and competitive process capabilities for the

management of interorganizational relationship portfolios’. Information Systems Research, 21 (3),

pp. 516-542.

Ravichandran, T. and Rai, A. (1999) ‘Total quality management in information systems development:

key constructs and relationships’. Journal of Management Information Systems, 16 (3), pp. 119-

155.

Rikhardsson, P. and Yigitbasioglu, O. (2018) ‘Business intelligence & analytics in management

accounting research: Status and future focus’. International Journal of Accounting Information

Systems, 29 37-58.

Ringle, C.M., Sarstedt, M. and Straub, D.W. (2012) ‘Editor's Comments: A Critical Look at the Use of

PLS-SEM in" MIS Quarterly"’. MIS Quarterly, 36 (1), pp. iii-xiv.

Ritala, P. et al. (2015) ‘Knowledge sharing, knowledge leaking and relative innovation performance:

An empirical study’. Technovation, 35 (0), pp. 22-31.

Roberts, N. et al. (2012) ‘Absorptive capacity and information systems research: Review, synthesis,

and directions for future research’. MIS Quarterly, 36 (2), pp. 625-648.

Rosseel, Y. (2012) ‘lavaan: An R Package for Structural Equation Modeling’. Journal of Statistical

Software, 48 (2), pp. 1-36.

Rouhani, S. et al. (2018) ‘Business intelligence systems adoption model: an empirical investigation’.

Journal of Organizational and End User Computing, 30 (2), pp. 43-70.

Shollo, A. and Galliers, R.D. (2016) ‘Towards an understanding of the role of business intelligence

systems in organisational knowing’. Information Systems Journal, 26 (4), pp. 339-367.

Srinivasan, R. and Swink, M. (2018) ‘An investigation of visibility and flexibility as complements to

supply chain analytics: An organizational information processing theory perspective’. Production

and Operations Management, 27 (10), pp. 1849-1867.

37
Steininger, D.M. et al. (2022) ‘Dynamic capabilities in information systems research: A critical review,

synthesis of current knowledge, and recommendations for future research’. Journal of the

Association for Information Systems, 23 (2), pp. 447-490.

Svetina, D., Rutkowski, L. and Rutkowski, D. (2020) ‘Multiple-group invariance with categorical

outcomes using updated guidelines: an illustration using M plus and the lavaan/semtools packages’.

Structural Equation Modeling: A Multidisciplinary Journal, 27 (1), pp. 111-130.

Tang, T., Fang, E. and Qualls, W.J. (2020) ‘More is not necessarily better: An absorptive capacity

perspective on network effects in open source software development communities’. MIS Quarterly,

44 (4), pp. 1651-1678.

Taylor, W.A. and Wright, G.H. (2004) ‘Organizational readiness for successful knowledge sharing:

Challenges for public sector managers’. Information Resources Management Journal (IRMJ), 17

(2), pp. 22-37.

Teece, D.J., Pisano, G. and Shuen, A. (1997) ‘Dynamic capabilities and strategic management’.

Strategic Management Journal, 18 (7), pp. 509-533.

Trieu, V.H. (2017) ‘Getting value from Business Intelligence systems: A review and research agenda’.

Decision Support Systems, 93 111-124.

Wang, H. and Wang, S. (2008) ‘A knowledge management approach to data mining process for business

intelligence’. Industrial Management & Data Systems, 108 (5), pp. 622-634.

Warner, K.S. and Wäger, M. (2019) ‘Building dynamic capabilities for digital transformation: An

ongoing process of strategic renewal’. Long Range Planning, 52 (3), pp. 326-349.

Winter, S.G. (2003) ‘Understanding dynamic capabilities’. Strategic Management Journal, 24 (10), pp.

991-995.

Yeoh, W. and Koronios, A. (2010) ‘Critical success factors for business intelligence systems’. Journal

of Computer Information Systems, 50 (3), pp. 23-32.

Yeoh, W. and Popovič, A. (2016) ‘Extending the understanding of critical success factors for

implementing business intelligence systems’. Journal of the Association for Information Science

and Technology, 67 (1), pp. 134-147.

Zhou, K.Z. and Li, C.B. (2012) ‘How knowledge affects radical innovation: Knowledge base, market

knowledge acquisition, and internal knowledge sharing’. Strategic Management Journal, 33 (9),

pp. 1090-1102.

38
APPENDIX

The questionnaire questions

A. Personal and firm information

Dimension Category Dimension Category
Heavy industry
Female
manufacturing
Gender
Sales and processing
Male
industry
Administrators Catering and tourism
Business Intelligence
Transport and logistics
and Analytics
You are engaged in a
Information System Your company is Light industry
department related to
Management engaged in manufacturing
Other traditional
Other staffs
services
Less than 100
Internet services
employees
Information
100-500 employees
Number of staffs technology services
500-1000 employees Financial services
More than 1,000
Less than 50 million
employees
Between 50 million &
State-owned enterprise
100 million
Revenue of your
Between 100 million
Ownership of your Private enterprise company (RMB)
& 1 billion
company
Between 1 billion &
Joint venture
10 billion
Foreign-owned More than 100 billion

B. Measurement Items
Construct
Measurement Items (7-point scale) Literature
names
In my company, our business intelligence system is: Nelson, Todd and
easy to use to help with data analysis and management Wixom (2005);Yeoh
System
decisions. and Popovič
Quality of
able to integrate different business analytics components (2016);Kulkarni,
BI
and has the flexibility to switch between them. Robles-Flores and
reliable and has a low probability of software and Popovič (2017);Ahn

39
 hardware failure. and Sura (2020)
able to respond to our needs and complete business
analysis and visualization tasks in a short period.
In our company, the data and information provided by
business intelligence are:
accurate and objective and the probability of misleading
information is low.
Information complete, and the probability of missing information is
Quality of low.
BI consistent, and the probability of inconsistent information
output is low.
is available on time and can improve our efficiency.
usable, and information compatibility and
comprehensibility are high.
In our company:
I often share my knowledge, skills, and work information
with other employees.
other employees often share their knowledge, skills, and
work information with me. Kang and Lee
Knowledge
our company often holds inter-departmental meetings to (2017);Mola et al.
Sharing
exchange knowledge and information compared with (2020)
competitors.
we often hold inter-departmental meetings to exchange
and learn the company's work skills compared with
competitors.
Compared with competitors:
our company motivates employees to acquire new
knowledge and skills.
our company provides employees with opportunities to Rouhani et al.
Absorption
learn new knowledge and new skills. (2018);Božič and
Capability
our company's employees can acquire new knowledge and Dimovski (2019b)
skills.
our company’s employees can put their newly acquired
knowledge and skills into work and promote innovation.
Compared with competitors:
our company is efficient to produce new products and
Cui et al.
Innovation services.
(2015);Miroshnychenko
Performance our company expansion of new markets timely.
et al. (2020)
our company takes less time for innovation.
our company takes less financial cost for innovation.

40
Table 1 Sample demographics
Sample characteristics Category Obs. Percentage

Less than 100 employees 65 18.47%

100-500 employees 46 13.07%

Firm size
500-1000 employees 85 24.15%

More than 1,000 employees 156 44.32%

Less than 50 million 39 11.08%

Between 50 million & 100 million 45 12.78%

Sales revenue Between 100 million & 1 billion 47 13.35%

Between 1 billion & 10 billion 99 28.13%

More than 100 billion 122 34.66%

State-owned enterprise 106 30.11%

Private enterprise 105 29.83%

Ownership
Joint venture 69 19.60%

Foreign-owned 72 20.45%

Heavy industry manufacturing 39 11.08%

Sales and processing industry 41 11.65%

Catering and tourism 32 9.09%

Transport and logistics 27 7.67%

Industry Light industry manufacturing 14 3.98%

Other traditional services 15 4.26%

Internet services 64 18.18%

Information technology services 59 16.76%

Financial services 61 17.33%

High level of information technology 184 52.27%

Level of information
Low level of information technology 168 47.73%
Table 2 Model competition results of CFA
Model χ2 df χ2/df Δχ2 RMSEA CFI

Five-dimension model 430.728 179 2.406 0.063 0.953

Four-dimension model 1442.59 183 7.883 1011.8622*** 0.140 0.766

Three-dimension model 1939.862 186 10.429 497.2722*** 0.164 0.674

Two-dimension model 2738.974 188 14.569 799.1119*** 0.196 0.526

Single-factor model 3301.539 189 17.468 562.5649*** 0.216 0.421

Note: *Significant at the 0.05 level; **Significant at the 0.01 level; ***Significant at the 0.001 level.
Table 3 Factor Loading, Cronbach'α, and CR
Factor Factor loading Cronbach'α CR

BISQ1 0.776

BISQ2 0.792
0.890 0.893
BISQ3 0.907

BISQ4 0.809

BIIQ1 0.734

BIIQ2 0.723

BIIQ3 0.757 0.854 0.855

BIIQ4 0.665

BIIQ5 0.801

KS1 0.837

KS2 0.919
0.923 0.925
KS3 0.835

KS4 0.885

AC1 0.817

AC2 0.906
0.922 0.925
AC3 0.847

AC4 0.902

IP1 0.819

IP2 0.864
0.929 0.930
IP3 0.913

IP4 0.906

Note: BISQ: system quality of BI; BIIQ: information quality of BI; KS: knowledge sharing; AC:

absorptive capability; IP: innovation performance.

Table 4 Correlations matrix, AVE, square root of AVE
Latent Constructs BISQ BIIQ KS AC IP AVE

BISQ 0.824 0.679

BIIQ 0.127 0.737 0.543

KS 0.252 0.354 0.870 0.757

AC 0.404 0.255 0.614 0.870 0.756

IP 0.225 0.162 0.369 0.462 0.878 0.770

Note: The diagonal bold is the square root of AVE

Table 5 Model competition results

χ2 df Δχ2 RMSEA CFI GFI NFI

Model A 507.905 241 0.056 0.951 0.881 0.910

Model B 508.951 243 1.046 0.056 0.951 0.880 0.910

Model C 729.049 245 220.099*** 0.075 0.910 0.838 0.871

Note: Hypothetical model A is the partially mediating model; competitive model B is the completely

mediating model. competitive model C is the direct model. ***Significant at the 0.001 level.
Table 6 Summary of the SEM path coefficients
Structural path Model A Model B Conclusion

BISQ→IP 0.038 H1 not supported

BIIQ→IP 0.045 H2 not supported

BISQ→KS 0.218*** 0.218*** H3 supported

BIIQ→KS 0.332*** 0.331*** H4 supported

BISQ→AC 0.266*** 0.268*** H5 supported

BIIQ→AC 0.034 0.036 H6 not supported

KS→AC 0.540*** 0.540*** H7 supported

AC→IP 0.439*** 0.466*** H8 supported

size→IP 0.107 0.120

revenue→IP 0.036 0.024

ownership→IP -0.009 -0.012

industry→IP 0.003 -0.003

Notes: *Significant at the 0.05 level; **Significant at the 0.01 level; ***Significant at the 0.001 level.
Table 7 Summary of mediation results and hypotheses

95% Confidence 95% Confidence

Direct effect Indirect effect Conclusion
interval interval

BISQ→AC→IP 0.038 [-0.082; 0.157] 0.117*** [0.057; 0.177] H9a supported

BIIQ→AC→IP 0.045 [-0.072; 0.161] 0.015 [-0.028; 0.058] H9b not supported

0.038 (BISQ→IP)
BISQ→KS→AC→IP [-0.082; 0.157] 0.052** [0.017; 0.087] H10a partially
0.266*** (BISQ→AC)

0.045 (BIIQ→IP)
BIIQ→KS→AC→IP [-0.072; 0.161] 0.079*** [0.034; 0.123] H10b supported
0.034 (BIIQ→AC)

Notes: *Significant at the 0.05 level; **Significant at the 0.01 level; ***Significant at the 0.001 level.
Table 8 Summary of the multigroup SEM path coefficients
Industries with high potential IT Industries with low potential IT

capability capability
Structural path
95% Confidence 95% Confidence
Path coefficient Path coefficient
interval interval

BISQ→KS 0.271*** [0.130; 0.411] 0.173 [-0.009; 0.356]

BIIQ→KS 0.290*** [0.129; 0.451] 0.374*** [0.173; 0.574]

BISQ→AC 0.279** [0.112; 0.447] 0.247** [0.095; 0.399]

BIIQ→AC 0.053 [-0.076; 0.181] 0.002 [-0.165; 0.17]

KS→AC 0.513*** [0.344; 0.683] 0.579*** [0.402; 0.756]

AC→IP 0.482*** [0.324; 0.639] 0.442*** [0.291; 0.594]

size→IP 0.019 [-0.232; 0.27] 0.209* [0.006; 0.413]

income→IP 0.095 [-0.137; 0.328] -0.023 [-0.238; 0.192]

ownership→IP -0.050 [-0.188; 0.087] 0.020 [-0.156; 0.196]

BISQ→AC→IP 0.135** [0.051; 0.218] 0.109** [0.028; 0.191]

BIIQ→AC→IP 0.025 [-0.038; 0.088] 0.001 [-0.073; 0.075]

BISQ→KS→AC→IP 0.067* [0.015; 0.119] 0.044 [-0.007; 0.095]

BIIQ→KS→AC→IP 0.072* [0.012; 0.132] 0.096** [0.034; 0.158]

Notes: *Significant at the 0.05 level; **Significant at the 0.01 level; ***Significant at the 0.001 level.
 Figure 1 Research model.

Figure 2 Estimated relationships of the structural model

View publication stats