Extending the model of internet standards adoption: A cross-country comparison of IPv6 adoption

Information & Management xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Information & Management journal homepage: www.elsevier.com/locate/im

Extending the model of internet standards adoption: A cross-country comparison of IPv6 adoption ⁎

Xuequn Wang , Sebastian Zander School of Engineering and Information Technology, Murdoch University, 90 South Street, Murdoch, WA, 6150, Australia

A R T I C L E I N F O

A B S T R A C T

Keywords: Internet standards Technology adoption IPv6 Top management support China Australia

Internet standards are very important for the effective operation of organizations. This study extends on the model of Internet standards adoption by examining the effect of organizational factors on adoption of Internet standards across different cultures. We propose attitude, top management support, and participation as important organizational factors influencing Internet standards adoption. Surveys were conducted in Australia and China, and the results show that organizational factors indeed play an important role in Internet standards adoption. Moreover, the effects of organizational factors are significantly different between the two countries. Discussion and implications of these findings are provided to conclude this study.

1. Introduction The Internet is ubiquitous these days. The vast majority of organizations in developed countries are connected to it, and developing countries are catching up fast. Standardized communication protocols are vital for organizations to seamlessly operate on the Internet. Arguably the most important standard is the Internet Protocol (IP), which is responsible for the logical addressing of all connected devices, so they can communicate with each other. However, the address space of IP version 4 (IPv4) has almost entirely been occupied. The remaining IPv4 address blocks were allocated to the five Regional Internet Registries (RIRs) by the Internet Corporation for Assigned Names and Numbers in 2011. Soon after, Microsoft offered US $7.5 million for a block of 666,624 IPv4 addresses [1]. As the demand for IP addresses has grown, an IPv4 address resale market has emerged. Brokers transfer IPv4 address from companies with an excess of addresses to those that need IPv4 addresses but can no longer obtain them from RIRs. Network Address Translation (NAT) helps to conserve IPv4 addresses, but it is only intended to be a short-term solution.1 To solve the issue of the IPv4 address shortage in the long term, IP version 6 (IPv6) was developed to provide a vast new address space. The enormous size of the Internet, with millions of connected devices owned by individuals and organizations, has encumbered the adoption

of new standards. Indeed, several new standards have been developed, but adoption has stalled; however, the transition from IPv4 to IPv6 is the most prominent example of a new Internet standard, for which adoption was initially slow but has accelerated in recent years [2]. To improve understanding of Internet standards adoption, Hovav and co-workers [3,4] propose a model of Internet standards adoption (ISA). Their work greatly furthers our understanding of the adoption of Internet standards such as IPv6. Despite recent progress, there are still two main limitations in the current literature. First, while Hovav et al.’s [4] ISA model is helpful to understand how usefulness and environmental conduciveness influence the adoption of Internet standards, little is known regarding how organizational factors influence ISA. The previous literature has suggested that organizational factors play an important role in technology adoption. For example, in the context of green information systems (IS), Wang et al. [5] argue that management leadership, green mindset, and past experiences positively influence the intention of organizations to adopt Green IS initiatives. Vykoukal [6] also finds that top management support is positively related to grid assimilation.2 However, few studies have examined how various organizational factors influence ISA. Second, previous studies on ISA in general and IPv6 adoption in particular tend to focus on a single country. For example, Hovav et al. [3] examine IPv6 adoption in South Korea, and several surveys have

⁎

Corresponding author. E-mail address: [email protected] (X. Wang). 1 NAT allows many devices with private IPv4 addresses, which can be reused in different private networks, to communicate with the public Internet through a NAT gateway. The gateway translates between private addresses and one or more public IP addresses assigned to the gateway. NAT is widely deployed in home networks and some consider it to be an alternative to IPv6. However, we believe that it is not really an alternative as it has a number of drawbacks compared with IPv6. For example, it causes problems with geolocation and geoblocking and it has stifled Internet protocol innovation (any new protocol has to be NAT-friendly). 2 Grid technology is a specific kind of Green IS initiatives that automatically adjusts the provision of hardware resources based on demand (Vykoukal [6]). http://dx.doi.org/10.1016/j.im.2017.10.005 Received 31 December 2016; Received in revised form 10 October 2017; Accepted 26 October 2017 0378-7206/ © 2017 Elsevier B.V. All rights reserved.

Please cite this article as: Wang, X., Information & Management (2017), http://dx.doi.org/10.1016/j.im.2017.10.005

Information & Management xxx (xxxx) xxx–xxx

X. Wang, S. Zander

applications that are perceived to have high value) may be important for ISA [16]. However, in their later study, Hovav et al. [3] fail to find any significant relationship between killer applications and ISA. The other perspective is the economic perspective, which deals with the economic value of the innovation to organizations [17]. This perspective examines the influence of the environment on adoption decisions, such as switching costs, resource concentration and power, government sponsorship, and normative pressure [3]. Resource concentration and power, and normative pressure have been found to significantly influence adoption decisions, while switching cost and government sponsorship have no significant effect [3]. By integrating these two perspectives, Hovav and co-workers [3,4,16] develop an integrative model of ISA. They argue that decisions to adopt Internet standards depend on both usefulness of features based on the diffusion of innovation perspective and environmental conduciveness based on the economic perspective. Their ISA model makes a significant contribution to our understanding of how organizations adopt Internet standards. However, the ISA model fails to consider the role of organizations in ISA. For example, Hovav and Schuff [16] argue that the dissemination of information through consortia is important in the adoption of Internet standards such as IPv6. Yet, the ISA model does not capture how organizations can obtain relevant information regarding Internet standards to facilitate their adoption decisions. From the previous literature, this study extends the ISA model by including organizational factors and by trying to understand the effects of these factors on ISA. The key variables considered in the present study include usefulness factors,4 environmental factors, and organizational factors (Table 1). The usefulness factors we included were relative advantage and complexity as they consistently relate to innovation adoption [15]. We also added the variable new features to deal with killer applications as this is argued to be vital in the context of Internet standards [3]. For environmental conduciveness, we include normative pressure as the main variable. Switching cost is excluded as no significant effect of this variable is detected in previous literature [3]. Resource concentration and power deals with the uneven allocation of important resources [3]. As the demand for IPv4 addresses is probably uneven between countries, resource concentration may not be an issue in certain contexts. For example, a country whose main industry is agriculture may not demand as many IPv4 addresses as another country focusing on Information Technology (IT) industries. Therefore, instead of using resource concentration and power, we select IPv4 running out as a control variable to capture the degree to which organizations are facing an IPv4 address shortage.

been conducted to understand the adoption of IPv6 in Australia [7], Croatia [8], and the US [9]. However, none of these studies examined the role of culture on the adoption of Internet standards such as IPv6. To fill those two gaps in the current literature, we aim to examine how organizational factors influence ISA across different countries, using the context of IPv6 adoption in Australia and China. Our study makes two importation contributions. First, our study extends the ISA model by including organizational factors. In particular, we propose that attitude toward Internet standards, top management support, and participation are important organizational factors that may influence ISA. These three variables are derived from DeSanctis and Poole’s [10] adaptive structuration theory (AST) and are adapted to the context of ISA. Attitude toward Internet standards is consistent with the degree to which members agree that structures should be appropriated. Top management support is consistent with members’ style of interaction and deals with management style to support ISA. Participation is consistent with members’ degree of knowledge and experience and focuses on how organizations seek information on Internet standards. AST is chosen because it is a useful theoretical lens through which to understand the process of technology adoption and has already been applied in different contexts, such as Green IS [5], and social media [11]. By examining the effects of organizational factors on ISA, our study aims to provide further insight into how organizations adopt Internet standards. Second, our study elucidates the role of culture on ISA. As McCoy [12] posit, as “globalization of businesses and systems continues to increase, our understanding about the adoption and use of IT needs to apply to other cultures,” we aim to understand how factors driving ISA vary across different cultures. Australia and China are selected because they are two economically important countries in the Asia-Pacific region with quite different cultures. Further, according to statistics by Google [13], both countries exhibited low but rising IPv6 deployment at the beginning of 2016.3 Thus, our study contributes to the current literature and provides valuable insights regarding how Internet standards, such as IPv6, are adopted across different cultures. The rest of the paper is organized as follows. We first review previous literature on ISA and describe the status of IPv6 adoption in Australia and China. Then, on the basis of findings from the literature, we develop research hypotheses and the research model. The research method and data analysis are then presented. Finally, the implications for theory and practice are discussed.

2. Literature review 2.1. Internet standards adoption

2.2. IPv6 adoption in Australia and China

Adoption of Internet standards is a complex scenario for organizations [3]. On one hand, organizations make their own decisions regarding whether to adopt certain Internet standards and to what extent. On the other hand, organizations may need to coordinate their adoption decisions with other organizations because they are closely related and interconnected. Therefore, there is an ongoing tension between autonomous decision-making and the need for interrelatedness in the context of ISA [4]. In regards to these two aspects, two perspectives have been proposed to understand ISA. The first perspective is diffusion of innovation, which focuses on the value of attributes of the innovation [14]. Among five innovation characteristics proposed by Rogers [14], only relative advantage and complexity consistently relate to innovation adoption [15]. According to Hovav and Schuff [16], as Internet standards represent infrastructure technology, they need new applications to derive relative advantage and drive adoption. Thus, killer applications (new

In the following paragraphs, we briefly introduce IP and then provide an overview of IPv6 adoption in the context of Australia and China. When IPv4 was developed in the late 1970s, 32-bit-long addresses were seen as sufficient to cope with future growth. However, in the late 1980s it became obvious that the Internet would run out of unused IPv4 addresses relatively soon, so IPv6 was developed. IPv6 has 128-bit-long addresses, making the address space so large that it is very unlikely to ever be exhausted. IPv6 was standardized in 1998, but it was not deployed by most organizations for many years, largely because (1) IPv6 is incompatible with IPv4 and requires a costly infrastructure upgrade and transition techniques that allow IPv4 and IPv6 to coexist and (2) IPv6 lacks attractive new features, apart from the large address space, 4 As the ISA model is developed by combining diffusion of innovation and the economic perspective, useful factors from the ISA model are from innovation characteristics proposed by Rogers [14]. Here, we use the name “useful factors” instead of “innovation characteristics” to be consistent with the terminology of the ISA model.

3 Note that on a per capita basis Australia has a much larger allocation of IPv4 addresses than China.

2

Information & Management xxx (xxxx) xxx–xxx

X. Wang, S. Zander

Table 1 Descriptions of key variables. Factors impacting on the adoption of IPv6 Usefulness New features Relative advantage Complexity Environmental conduciveness Normative pressure Organization Attitude Top management support Participation

Description

Source

Empirically tested?

Killer applications or integrated services introduced by Internet standards Competitive advantage opportunities (e.g., new markets, products, and services) created by Internet standards The effort required to implement Internet standard

Hovav and Schuff [16]; Hovav et al. [3]

Yes

Hovav et al. [4]

Yes

Hovav et al. [4]

Yes

Strong influences over ISA of other organizations from leading organizations or industry

Hovav et al. [3]

Yes

Overall perception toward Internet standards The degree to which top management regards IPv6 as important and provides resources for adoption Participate in forums or relevant activities to learn the technical features and the market values of Internet standards

Derived from DeSanctis and Poole [10] Derived from DeSanctis and Poole [10]

No No

Hovav and Schuff [16]; also, consistent with DeSanctis and Poole [10]

No

[21] proposes five dimensions of national culture: power distance, uncertainty avoidance, individualism/collectivism, masculinity/femininity, and long-/short-term orientation. Hofstede’s culture dimensions have been widely applied in various contexts, such as adoption of instant messaging [22], and health information seeking [23]. Among the five dimensions of national cultures proposed by Hofstede [21], individualism/collectivism is particularly relevant to understanding the differences in IPv6 adoption between Australia and China. Individualism/collectivism deals with the relationship between members and a group and refers to the degree to which the interests of members are prioritized over those of the group [21]. Members of an individualistic society, such as Australia, focus on achievement and productivity and thus emphasize their own interests rather than those of the group by Triandis [48]. In contrast, members of a collectivist society, such as China, are interdependent and place more emphasis on the interests of the group. In the present study, we apply the concept of individualism/collectivism to understand how organizations from Australia and China may deal with IPv6 adoption differently (Fig. 1). Below, we describe our hypotheses in more detail.

which makes the upgrade unappealing for first adopters. Australia has an allocation of about 47.6 million IPv4 addresses. With a population of roughly 22 million people, it is approximately 2160 IPv4 addresses per 1000 citizens. China has a much larger allocation of 330.3 million IPv4 addresses, but given that its population exceeds 1.3 billion people, this means there are only about 250 IPv4 addresses per 1000 citizens in China. Consequently, as China has far more citizens than allocated IPv4 addresses, the transition to IPv6 is very important in China. In August 2003, the State Council of China officially authorized the launch of the China Next Generation Internet (CNGI) project. The project was part of China’s five-year plan, and as part of the project, IPv6 backbone networks were extended to over 22 major cities and more than 270 access networks were connected to the backbone [18]. The Chinese Government also took a lead in planning and deploying IPv6 across the public-facing infrastructure (i.e., e-government). The CNGI initiative basically mandated IPv6 to the industry, and in 2014, it was reported that Chinese carriers claimed to have implemented full IPv6 deployment. The Ministry of Industry and Information Technology of China announced that “it will continue to promote the Internet sector’s adoption of IPv6 with an investment of over CNY20 billion” [19]. While in China the state mandate has driven IPv6 deployment in most Internet service providers (ISPs), in Australia adoption has been more muted. The Australian Government developed an initial strategy for implementing IPv6 in government agencies in 2007 (version 2, released in 2009) [20], and in 2013, it was determined that “the agencies were well advanced in their transition” and that “the majority of the work within agency systems is now completed”.5 In Australia, IPv6 trials were launched by several organizations early on, and the national research network (AARNET) deployed IPv6 in 2006.6 However, by late 2016, few ISPs had fully implemented IPv6 or offered it to their customers. It appears that the larger number of available IPv4 addresses per capita has made the transition less urgent in Australia.

3.1. Usefulness and adoption decisions From previous literature, we identify three usefulness factors: (1) new features, (2) relative advantage, and (3) complexity. IPv6 is an infrastructure technology and thus needs to provide new features, such as killer applications, to increase user demand and create value [3]. The concept of new features is closely related to relative advantage, that is, the degree to which new features could generate new markets, products, and services to create competitive advantage [4]. Hovav and Schuff’s [16] case analysis shows that ISPs can benefit from the IPv6-related applications they offer. Indeed, if organizations perceive that IPv6 can bring new features and those new features can generate positive outcomes (e.g., meet user demand), they may be more likely to adopt IPv6. Conversely, organizations with an existing IPv4 infrastructure that needs to be upgraded may be less likely to adopt IPv6 if the process is too complex and/or difficult. Further, organizations from different cultures may respond to usefulness factors differently. As described above, an individualism society emphasizes achievement and productivity. In such a context, organizations may focus more on the benefits of new features IPv6 can bring and how they can generate competitive advantage. Thus, if the benefits associated with upgrading to IPv6 are not clear, organizations in those cultures may be less likely to adopt IPv6. We propose that the relationship between usefulness factors and IPv6 adoption may be stronger for those organizations from individualistic societies, such as Australia. Thus, our first hypotheses are

3. The research model Adopting Internet standards, such as IPv6, is a global issue. Organizations, regardless of their country of origin, need IP addresses to connect to the Internet. However, organizations from different cultural backgrounds may deal with IPv6 adoption differently. National culture refers to “the collective programming of the mind that distinguishes the members of one group or category of people from another” (). Hofstede 5 6

http://www.finance.gov.au/archive/agimo-archive/ipv6/. http://ipv6forum.com.au/timeline.php.

3

Information & Management xxx (xxxx) xxx–xxx

X. Wang, S. Zander

Fig. 1. Research Model.

H1. (a) New features are positively related to IPv6 adoption, and (b) this relationship is stronger in an individualistic society (Australia > China).

3.3. Organizational factors and adoption decisions For organizational factors, our study focuses on attitude, top management support, and participation.7 Attitude is defined as “positive or negative feelings (evaluative affect) about performing the target behavior” (). In the present study, we adapt the concept of attitude to refer to organizations’ overall perception of IPv6. For instance, Wang et al. [5] argue that organizations’ attitude toward Green IS positively influences their intention to adopt Green IS initiatives. Similarly, Molla et al. [26] propose that Green IT attitude (i.e., organizations’ attitude toward climate change and eco-sustainability and the role of IT) is an important component of Green IT readiness of organizations. In the context of IPv6, organizations with a positive attitude toward IPv6 are more likely to have a proactive mindset toward IPv6 and feel that IPv6 can generate positive outcomes (e.g., competitive advantage). Conversely, if organizations do not have a positive attitude, they may be reluctant to adopt IPv6. Culture may also influence the relationship between attitude and IPv6 adoption. Members of an individualist society tend to be more proactive when dealing with challenges, by focusing on actions and positive outcomes [27]. In such contexts, attitude may have a stronger effect, and organizations with a positive attitude may be more likely to adopt IPv6 proactively. On the contrary, collectivistic societies emphasize “collective coping,” that is, “the centered effort involving all members of a group to tackle the same problem” [28] p. 14. Therefore, organizations from a collectivistic society may rely more on authorities, such as governments, to solve the issue of IPv4 addresses running out. In other words, those organizations may wait for collective coping even if they have a positive attitude toward IPv6. Therefore, we posit that

H2. (a) Relative advantage is positively related to IPv6 adoption, and (b) this relationship is stronger in an individualistic society (Australia > China). H3. (a) Complexity is negatively related to IPv6 adoption, and (b) this relationship is stronger in an individualistic society (Australia > China).

3.2. Environmental conduciveness and adoption decisions For environmental conduciveness, we focus on the effect of normative pressure. Normative pressure refers to the influence of leading organizations or industries over the activities of other organizations [3]. Hovav et al. [3] argue that normative pressure from peer organizations or industries may direct other organizations to develop and implement IPv6. From the perspective of isomorphism, organizations function in regulated industries, and there are isomorphic forces exerted by other organizations and agencies on which these organizations depend [24]. In the context of IPv6, a norm of IPv6 adoption may spread across the industry, and any organizations that do not follow may lose their legitimacy. Further, normative pressure may have different effects in different cultures. In a collectivistic society, in which members are more interdependent and the interests of the group are emphasized, relationships between organizations may also be stronger. In such contexts, these stronger relationships may have a greater capacity to develop shared norms and make organizations behave in identity-appropriate ways [24]. As a result, organizations in collectivistic countries are more likely to be influenced by normative pressure to adopt IPv6. Indeed, Hovav et al. [3] argue that leading organizations from collectivistic societies may strongly influence the adoption of IPv6 by other organizations. Therefore, we hypothesize

H5. (a) Attitude is positively related to IPv6 adoption, and (b) this relationship is stronger in an individualistic society (Australia > China).

7 In the context of our study, attitude focuses on organizations’ overall perception toward IPv6 and reflects the degree to which organizations are willing to adopt IPv6. Conversely, top management support shows that managers are positive toward IPV6 and provide necessary resources for IPv6 adoption.

H4. (a) Normative pressure is positively related to IPv6 adoption, and (b) this relationship is stronger in a collectivistic society (China > Australia). 4

Information & Management xxx (xxxx) xxx–xxx

X. Wang, S. Zander

Top management support refers to the degree to which top managers regard IPv6 as important and provide resources for adoption. Previous studies have shown that top management support has a positive effect on technology adoption [29,30,6]. In the context of IPv6, a high level of top management support means that top managers understand the benefits associated with IPv6 and demonstrate their commitment and political support. Therefore, top management support is expected to have a positive effect on IPv6 adoption. The effect of top management support may differ across various cultural contexts. In a collectivistic society, authorities may be relied on to deal with collective issues. Thus, organizations may be more likely to adopt IPv6 when authorities, such as top managers, propose IPv6 as the approach to solve the issue of IPv4 addresses running out. Therefore, we suggest that

Table 2 Demographic characteristics of the organizations.

Industrya Agriculture, Forestry, and Fishing Mining Manufacturing Electricity, Gas, Water, and Waste Services Construction Wholesale Trade Retail Trade Transport, Postal, and Warehousing Information Media and Telecommunications Financial and Insurance Services Rental, Hiring, and Real Estate Services Professional, Scientific, and Technical Services Administrative and Support Services Public Administration and Safety Education and Training Health Care and Social Assistance Arts and Recreation Services Other Services

H6. (a) Top management support is positively related to IPv6 adoption, and (b) this relationship is stronger in a collectivistic society (China > Australia). Participation refers to the degree to which organizations participate in forums or professional activities to seek IPv6 information. Hovav and Schuff’s [16] case analysis indicates that the availability of information can influence organizations’ perception of risk, and reduce uncertainty surrounding IPv6 adoption. Thus, organizations that have little knowledge regarding the advantages of IPv6, implementation needs, and upgrade costs may show greater perceived risk and uncertainty toward IPv6. Indeed, increased risk perception may prevent organizations from adopting IPv6. However, through participation in forums and professional activities, organizations may not only understand the technical features of IPv6 but also its market potential (e.g., an ISP may gain a competitive advantage by adopting IPv6). As a result, they are more likely to adopt IPv6. Members of a collectivistic society tend to maintain harmony by seeking information and guidance [31]. Therefore, when dealing with a common issue, such as IPv4 addresses running out, organizations from a collectivistic society are more likely to seek out information and consult with others. In other words, organizations in collectivistic cultures probably rely more on authorities for information and advice; therefore, they are more likely to engage in various activities to seek information about IPv6 and rely on the obtained information to make their adoption decisions. To summarize, we suggest that H7. (a) Participation is positively related to IPv6 adoption, and (b) this relationship is stronger in a collectivistic society (China > Australia).

4. Methodology In this section, we first describe our data collection procedure and the characteristics of our sample. We then describe our measures and assess their validity. Finally, we present our results of modeling testing with ordinal regression.

China (N = 148)

Phase I (N = 114)

Phase I (N = 111)

Phase II (N = 59)

Phase II (N = 37)

0.9 0.9 10.5 1.8

5.1 5.1

19.8 0.9

3.5 7.0 14.9 1.8

1.7

1.8 0.9 3.6 3.6

2.7

28.1

37.3

40.5

51.4

7.0

1.7

4.5

2.7

18.9

1.7 11.4

20.3

14.4

1.8

1.7

1.8

3.4

0

8.5 5.1

5.4 1.8

3.5 7.9

1.7 6.8

10.8

2.7 10.8

Financial status: Profit

98.2

74.6

95.5

86.5

Number of Employees 1–100 101–500 501–1000 1001–2000 2001–10,000 10,001–25,000 25,001 or above

5.3 16.7 40.4 24.6 10.5 2.6

54.2 15.3 3.4 6.8 15.3 1.7 3.4

7.2 20.7 34.2 17.1 13.5 4.5 2.7

18.9 21.6 13.5 8.1 21.6 10.8 5.4

Size of customer base 1–1000 1001–10,000 10,001–100,000 100,001–500,000 500,001–1,000,000 > 1,000,000

14.9 8.8 24.6 34.2 11.4 6.1

35.6 18.6 20.3 15.3 10.2

13.5 35.1 25.2 11.7 6.3 8.1

18.9 16.2 10.8 10.8 10.8 32.4

Number of IT Staff Less than 10 10–50 51–100 More than 100

12.3 42.1 30.7 14.9

44.1 25.4 13.6 16.9

7.2 36.9 40.5 15.3

8.1 35.1 13.5 43.2

a

4.1. Data collection

Australia (N = 173)

2006 Australian and New Zealand Standard Industrial Classification (ANZSIC).

checked to detect potential data records from same organizations. Two responses may come from the same organization if they (1) had the same domain name, (2) had the same IP address, or (3) had similar answers regarding the demographics of their organization and deployment of IPv6. After identifying those potential data records, the two authors then discussed and decided whether to remove a particular data record for each case. From the multiple records deemed to come from the same organization, only one random record was kept. In total, we received 288 complete responses (Australia 138; China 150) from Phase I and 98 (Australia 60; China 38) from Phase II. Table 2 presents the demographics of organizations, and Table 3 presents the demographics of participants in both phases. We also compare our sample with previous studies in Table 4. Similar to Dell [7], our Australian samples were from a variety of industries based on

To test our model, we conducted surveys among organizations in Australia and China. In Phase I (early 2016), survey companies were hired to conduct data collection. Because this study examines the status of IP adoption within organizations, participants would only qualify for the survey if they (1) worked in the area of IT, (2) knew what IPv6 is, and (3) understood the current status of IPv6 deployment and the IPv6 strategy in their organization. In Phase II (mid-2016), an invitation to the survey was sent by APNIC8 to its members. The similarities between responses for each country were then 8 Asia-Pacific Network Information Canter (APNIC) is the RIR that administers IP addresses in the Asia-Pacific Region.

5

Information & Management xxx (xxxx) xxx–xxx

X. Wang, S. Zander

theory of reasoned action [25] to reflect organizations’ overall attitude toward IPv6; the items for participation were adapted from Ozgen and Baron’s [37] participation in professional forums. Newly developed items were used to measure new features following Churchill’s [38] procedures. Unlike Hovav et al. [3], we used items to measure overall perception of new features provided by IPv6. All items were measured on five-point Likert scales from strongly disagree to strongly agree. See Appendix A for details on the measurement.

Table 3 Demographic characteristics of the participants.

Job level CIO or other top-level manager Middle-level IT manager Low-level IT manager Senior IT administrator/ technician IT administrator/technician Decision-making authority Final Significant decision making or influence Minimal decision making or influence None

Australia (N = 173)

China (N = 148)

Phase I (N = 114)

Phase II (N = 59)

Phase I (N = 111)

Phase II (N = 37)

40.4

33.9

64.9

24.3

48.2 2.6 4.4

20.3 1.7 33.9

18.0 3.6 9.9

24.3 18.9 21.6

4.4

10.2

3.6

10.8

64.0 32.5

35.6 50.8

71.2 26.1

8.1 62.2

3.5

10.2

2.7

21.6

0

3.4

0

8.1

Gender: Female

21.1

3.4

18.9

10.8

Age 18–19 20–29 30–39 40–49 50–59 60 or above

0.9 20.2 29.8 39.5 9.6 0

0 13.6 35.6 27.1 20.3 3.4

0 15.3 45.9 34.2 4.5 0

0 21.6 48.6 27.0 0 2.7

4.2.3. Control variables Following Hovav et al. [3], we included several control variables that may influence IPv6 adoption. First, a five-point scale measuring “maintaining a leadership position” was included as organizations are more likely to adopt new technical standards if they regard themselves as leaders in the field. Second, telecommunication companies may adopt new technical Internet standards earlier than companies in other industries. Therefore, we included an industry dummy variable: “1” for companies from the telecommunication industry and “0” for all other industries. Third, organization size was included to account for the effect of organizational resources. It was measured on a seven-point scale ranging from less than 100 employees to more than 25,000 employees. Lastly, organizations are more likely to adopt IPv6 if their IPv4 addresses are running out. Therefore, we also measured, on a five-point scale, the degree to which the organization’s IPv4 addresses were running out. A pooled sample was used to assess the validity of our measures. Please refer to Appendix A for measures of Cronbach’s alpha, composite reliability (CR), average variance extracted (AVE), and factor loading. Specifically, Cronbach’s alpha for most of the variables was above 0.70.10 Confirmatory factor analysis was conducted to determine the model fit. The results showed that a seven-factor model produced χ2 (98) = 204.44, p < 0.000, SRMR = 0.040, RMSEA = 0.058, CFI = 0.97, and TLI = 0.96, indicating a good model fit. The results of standardized residual co-variances and modification index values indicated no conspicuously significant changes to the model. All factor loadings were significant and above 0.60, and the AVE for each factor was over 0.50, which indicates good convergent validity. Moreover, correlations among key variables were all below 0.85 [40] and the square root of their AVE exceeded all correlations between those variables and all other variables (Table 5), indicating good discriminant validity. Further, the variance inflation factors for all variables did not exceed 5, indicating that multi-collinearity was not a serious problem [41].

the ANZSIC industrial classification. Also consistent with Hovav et al.’s [3] South Korea sample, a relatively large percentage of participants in our Australian and Chinese samples were from the telecommunication industry. As Hovav et al. [3] suggest, studies examining the adoption of an IT-related infrastructure technology would be expected to have a large portion of the sample from the telecommunication industry. Because the sample size from Phase II was small, we decided to pool the Phase I and Phase II samples into one Australian sample and one Chinese sample.9 In total, 79.1% of the participants’ organizations had implemented IPv6 to some extent. For those that had not implemented IPv6, 34.3% were planning to implement IPv6 in the near future. As all variables were collected in one survey, we applied two approaches to examining the common method bias (CMB) [32]. First, a Harmon single factor test was conducted, and no single dominant factor explained majority of the variance. Second, the marker-variable technique was used [33]. After adjustment, all significant correlations between our key variables remained significant. Therefore, CMB is probably not a serious concern for this dataset.

5. Results To validate our hypotheses, we ran three models with ordinal regression for the Australian and Chinese data, respectively (Table 6). In Model 1, we considered only the control variables. In both datasets, none of control variables were significantly related to the adoption of IPv6. In Model 2, we included the variables of usefulness and environmental conduciveness. For the Australian dataset, only complexity was significantly related to IPv6 adoption, so only H3a was supported. For the Chinese dataset, complexity and normative pressure were both strongly associated with IPv6 adoption, supporting H3a and H4a. Finally, in Model 3, the variables related to organizations were added. For the Australian dataset, only top management support was significantly related to IPv6 adoption, so H6a was supported. For the Chinese dataset, top management support and participation were both strongly associated with IPv6 adoption, supporting H6a and H7a.

4.2. Measures 4.2.1. Dependent variable Following Hovav et al. [3], the dependent variable of adoption of IPv6 was measured by a four-point ordinal scale ranging from we have not adopted IPv6 to we have already implemented IPv6. 4.2.2. Independent variable The items for relative advantage and complexity were adapted from Ramamurthy et al. [34] and Wang et al. [35]; the items for normative pressure were adapted from Hovav et al. [3]; the items for top management support were adapted from Soliman and Janz [36] and Wang et al. [35]; the items for attitude toward IPv6 were adapted from the 9 We performed Box's M-test and the result was significant (p < 0.001). This is probably due to the fact that the characteristics of the two samples are not completely consistent. We admit that this is a limitation of our study and our results should be interpreted cautiously.

10 Two exceptions are from complexity and normative pressure, whose alphas are very close to 0.70. According to Robinson et al. [39], levels of 0.60 can be used in exploratory research. Nevertheless, it is still a limitation of our study.

6

Information & Management xxx (xxxx) xxx–xxx

X. Wang, S. Zander

Table 4 Comparison of demographic characteristics of the organizations between studies. Australia (N = 173)

Industry Agriculture, Forestry, and Fishing Mining Manufacturing Electricity, Gas, Water, and Waste Services Construction Wholesale Trade Retail Trade Accommodation and Food Services Transport, Postal, and Warehousing Information Media and Telecommunications Financial and Insurance Services Rental, Hiring, and Real Estate Services Professional, Scientific, and Technical Services Administrative and Support Services Public Administration and Safety Education and Training Health Care and Social Assistance Arts and Recreation Services Other Services Number of Employees 1–100 101–500 501–1000 1001–2000 2001–10,000 10,001–25,000 > 25000 a

China (N = 148)

Dell’ [7] Australian Sample (N = 180)

Hovav et al.’s [3] South Korea Sample (N = 84)

2 10 5 4 1 6 1 4 4 4 4

20

27 18 7 1 4

17 5

0.7 0.6 8.7 2.9 2.3 4.6 9.8

14.9 0.7 1.4 0.7 2.7

1.7 31.2 5.2 0.6 14.5 1.7 1.2 5.2 6.9 0.6 2.3

3.4 43.2 4.1

6.8 1.4 0.7 2.7

22.0 16.2 27.7 18.5 12.1 2.3 1.2

10.1 20.9 29.1 14.9 15.5 6.1 3.4

39

15.5 1.4

18.5a

8

35 24 11 11 15

20 22 37 7 7

8.5% is from “service” industry and 8% is from “other” industry [3].

Table 5 Correlations, means, and standard deviations (S.D.) of key variables (square root of AVEs on diagonal). 1

2

3

4

5

6

7

8

9

10

11

12

1 Leadership 2 Industry dummy 3 Firm size 4 IPv4 running out 5 New features 6 Relative Advantage 7 Complexity 8 Normative pressure 9 Top management support 10 Attitude toward IPv6 11 Participation 12 Adoption of IPv6

0.01 −0.01 0.36** 0.65** 0.27** 0.21** 0.67** 0.43** 0.49** 0.36** 0.09

−0.10 0.15** 0.04 −0.03 −0.07 −0.06 0.00 0.03 −0.01 0.11

−0.19** 0.08 0.03 0.02 0.10 0.15** −0.04 0.27 0.11

0.32** 0.09 0.14* 0.37** 0.16** 0.46** 0.12* 0.02

0.84 0.24** 0.15** 0.67** 0.41** 0.53** 0.31** 0.13*

0.85 −0.16** 0.10 0.27** 0.30** 0.15** 0.16**

0.71 0.33** 0.10 0.22** 0.12* −0.18**

0.72 0.41** 0.45** 0.41** 0.11

0.88 0.41** 0.72** 0.51**

0.85 0.29** 0.14**

0.93 0.43**

-

Mean S.D.

3.90 0.89

0.37 0.48

3.14 1.50

3.98 0.90

3.90 0.81

2.89 1.05

3.75 0.80

3.81 0.74

3.11 1.26

4.03 0.83

2.69 1.39

2.73 1.20

* p < 0.05 (two-tailed). ** p < 0.01 (two-tailed).

Australia, and factors regarding organizations also better predict IPv6 adoption than factors regarding usefulness and environmental conduciveness. Finally, cross-culture comparisons were conducted. Specifically, the formula of Keil et al. [42] was used to assess the statistical difference in path coefficients between two countries:

We also compared the Model Fitting Information, which included the −2 log-likelihood values for the baseline model (the “Intercept Only” model: a model, without any explanatory variables) and a specific model (the model with certain explanatory variables). Chi-square tests were run to examine the difference between the −2 log-likelihood for the two models. For the Australian dataset, the chi-square tests of Models 1 and 2 were not significant (p > 0.05), whereas the test of Model 3 was significant (p < 0.001). These results indicated that control variables and factors regarding usefulness and environmental conduciveness did not significantly improve the prediction of IPv6 adoption, while factors regarding organizations better predicted IPv6 adoption. For the Chinese dataset, the chi-square test of Model 1 was not significant (p > 0.05), whereas the tests of Models 2 and 3 were both significant (p < 0.01 and p < 0.001, respectively). These results indicated that our model better predicts IPv6 adoption in China than in

Path coefficient Group1 − Path coefficent Group2

t= ⎡ ⎣

(m−1)2 (m+ n− 2)

2 × SEGroup1 +

(n−1)2 (m+ n− 2)

2 ⎤×⎡ × SEGroup2 ⎦ ⎣

1 m

+

1 n

⎤ ⎦

where m is the sample size for Group 1 and n is the sample size for Group 2. These findings are summarized in Table 7. The moderation in H6b was supported, showing that top management support and IPv6 adoption have a stronger association for Chinese 7

Information & Management xxx (xxxx) xxx–xxx

X. Wang, S. Zander

Table 6 Ordinal regression analysis for adoption of IPv6. Australia

Controls Leadership Industry dummy Firm size IPv4 running out

China

Model 1

Model 2

Model 3

Model 1

Model 2

Model 3

0.10 0.24 0.18 0.16

0.13 0.32 0.20 0.16

−0.34 0.23 0.04 0.01

0.30 0.62 0.08 −0.26

−0.23 0.83* 0.09 −0.44

−0.11 1.06* −0.08 −0.12

0.05 0.20

0.00 0.15

0.12 0.23

−0.64 −0.00

−0.53*

−0.79**

−0.76**

−0.79*

0.11

0.08

1.19**

0.50

Factors regarding usefulness New features Relative advantage Complexity Factors regarding environmental conduciveness Normative pressure Factors regarding organization Attitude toward IPv6 Top management support Participation 2

Pseudo R (Nagelkerke)

0.06 0.76

0.11 ***

1.25*** 0.50*

0.15 0.03

0.08

0.25

0.05

0.18

0.56

* p < 0.05 (two-tailed). ** p < 0.01(two-tailed). *** p < 0.001 (two-tailed).

Table 7 Path coefficient differences between Australia and China. Hypotheses

H1b: H2b: H3b: H4b: H5b: H6b: H7b:

New features → IPv6 Adoption (A > C) Relative advantage → IPv6 Adoption (A > C) Complexity → IPv6 Adoption (A < C) Normative pressure → IPv6 Adoption (A < C) Attitude → IPv6 Adoption (A < C) Top management support → IPv6 Adoption (A < C) Participation → IPv6 Adoption (A < C)

Path Coefficient Australia

China

0.00 0.15 −0.79** 0.08 0.06 0.76*** 0.15

−0.64 −0.00 −0.79** 0.50 0.22 1.25*** 0.50*

Diff.

Sig

Hypotheses Supported?

0.64 0.15 −0.01 −0.43 −0.05 −0.49 −0.36

bns bns ns bns bns

No No No No No Yes Yes

***

sd

bns = both paths are not significant, ns = significant, sd = structurally different (on path is significant and the other is insignificant; A = Australia, C = China. * p < 0.05 (two-tailed). ** p < 0.01(two-tailed). *** p < 0.001 (two-tailed).

demand are not significantly related to IPv6 adoption. Our results are also consistent with the opinion of experts that “there are no compelling technical features or revenue levers in IPv6 that are driving new investments” [43]. Thus, the new features of IPv6 may not be perceived as very valuable. The most valuable feature, the larger address space, is only really valuable when an organization’s IPv4 addresses have been exhausted and there is sufficient IPv6 support. Additionally, it is possible that certain government policies (e.g., CNGI) reduce the competitive forces driving the “usefulness” of IPv6 [3], and the relative advantage of IPv6 is still not quite clear for organizations. In Model 2, normative pressure has a significant effect on IPv6 adoption for China, but not for Australia. This result is consistent with Hovav et al. [3], who argue that organizations have a greater tendency to follow the behaviors of leading organizations in collectivistic societies (e.g., China and South Korea). Another possible explanation is that in China there is a much stronger initiative from the government to deploy IPv6. For example, the CNGI initiative may exert strong pressure and require organizations to adopt IPv6. In such contexts, the CNGI initiative is consistent with mandatoriness of use from previous

organizations than Australian organizations. The moderation in H7b was also supported, indicating that participation had a stronger impact on IPv6 adoption for Chinese organizations than Australian organizations. The other moderation effects were not significant. 6. Discussions This study examines how usefulness, environmental, and organizational factors influence ISA in Australian and Chinese companies. We find that complexity and top management support significantly affected adoption of IPv6 (i.e., H3a and H6a are supported for both Australia and China). In addition, participation significantly influenced adoption in Chinese organizations (i.e., H7a is supported for China). Moreover, there are significant differences between two countries in the effect of top management support and participation (H6b and H7b are supported). Our results show that new features and relative advantage are not significantly related to IPv6 adoption. These results are similar to those of Hovav et al. [3], which show that killer applications and user 8

Information & Management xxx (xxxx) xxx–xxx

X. Wang, S. Zander

Our findings also have implications for the campaigns of organizations planning to adopt IPv6, especially for those operating in multiple countries. For example, top management support consistently improves IPv6 adoption in Australia and China. Therefore, organizations should ensure that their top managers provide sufficient support and resources, especially for those organizations from a collectivistic society. For organizations with a high level of top management support, IPv6 adoption may be more rapid, which may also positively influence future operations. For governments and regulators interested in driving adoption, it is important to bring top management on board, for example, through targeted information campaigns. Moreover, participation positively influences IPv6 adoption in China. Previous literature has shown the importance of normative pressure and suggested that governments should create an environment with strong normative pressure to facilitate IPv6 adoption by organizations [3]. Our study shows not only that normative pressure facilitates ISA but also that the effects of normative pressure are mediated by participation in a collectivistic society. Therefore, when organizations from collectivistic societies plan to adopt IPv6, they should participate in various IPv6-related activities to learn more about its technical features and business value. Hence, governments, in addition to creating an environment with normative pressure, should also hold forums and IPv6-related activities to provide information for organizations.

technology adoption literature [44]. However, the effect of normative pressure becomes insignificant in Model 3 after adding organizational variables. In our follow-up analyses, we find that the effect of normative pressure is mediated by participation. The result indicates that when feeling pressure from industry, Chinese organizations tend to participate in various IPv6-related activities to learn more about the technical features and market value of IPv6. Those organizations are thus more likely to adopt IPv6. Surprisingly, attitude is not significantly related to IPv6 adoption in Australian or Chinese organizations. In our follow-up analysis, we find that attitude becomes significant when top management support is excluded from our model for Chinese organizations, but not for Australian organizations. One possible explanation of this result is that a high level of top management support for IPv6 may be more likely to help organizations develop a positive attitude toward IPv6 in collectivistic cultures. In such contexts, the effect of attitude is superseded by that of top management support. The effect of complexity does not differ significantly between Australia and China, indicating that organizations from different cultural backgrounds consistently consider complexity of upgrading as an important factor for IPv6 adoption.11 6.1. Implications for theory Our study makes two significant theoretical implications. First, we extend Hovav et al.’s [4] ISA model by including relevant organizational factors. We argue that it is organizations that ultimately adopt an Internet standard. Therefore, it is critical to understand the role of organizational factors in ISA. From previous literature, we identify three relevant organizational factors: (1) attitude, (2) top management support, and (3) participation. Our results show that top management support is significantly associated with IPv6 adoption in both Australian and Chinese organizations. Further, participation is significantly related to IPv6 adoption in Chinese organizations. Lastly, our model including organizational factors performs significantly better than models without. Thus, it appears that organizational factors indeed play an important role in IPv6 adoption. Second, our study is among the first to examine the effect of culture on ISA. In the context of IPv6, our results show that participation and top management support have stronger effects in China. These results indicate that in a collectivistic society, such as China, top management support is more likely to lead organizations to adopt IPv6. Chinese organizations may also tend to seek authority for advice regarding the technical features and market value before they adopt IPv6. Therefore, although organizational factors play an important role in the process of ISA, the effects of these factors may differ in organizations from different cultures.

6.3. Limitations and opportunities for future research Our study had several limitations. First, although different approaches were used to collect data, and our samples from both countries covered a variety of industries, our samples may still be biased. In addition, only Australia and China were selected to examine the effect of culture. Although these two countries have different orientation of individualism/collectivism, it is possible that other factors (e.g., political or economic) may also influence the differences in IPv6 adoption between the two countries. Future studies could extend our study by collecting variables measuring different aspects (e.g., different political contexts) to examine whether these factors indeed have significant moderation effects. Future studies could also test our model with additional countries to check whether the conclusions still hold. Our study demonstrates the importance of organizational factors for ISA. Future studies could consider additional theoretical frameworks to examine how other organizational factors may influence ISA. Lastly, our study only considers adoption of IPv6. Future studies could examine the adoption of other Internet standards; for example, videoconferencing. There are several associated signaling protocols, such as H.323 (defined by ITU), SIP (defined by IETF), and H.248 (Megaco)/MGCP (collaboration of ITU and IETF). It would be interesting to examine how usefulness, environmental, and organizational factors influence an organization’s adoption of different videoconferencing protocols. As videoconferencing is directly used by end users, usefulness factors may have a stronger effect in such contexts.

6.2. Implications for practice Our study also has important implications for practice. First, governments promoting the adoption of IPv6 should take note that complexity consistently influences IPv6 adoption in Australia and China. Therefore, in addition to investing in related technology and test beds, governments should also provide training and technical support to help organizations upgrade their IT infrastructures and thus reduce the complexity of the process. Further, a well-designed information campaign could help to reduce the perception of complexity.

7. Conclusion Internet protocol standards such as IPv6 are crucial for organizations to operate efficiently. The present study extends on the previous literature by examining the effects of organizational factors on ISA in different cultures. Using the context of IPv6 adoption in Australia and China, our study shows that organizational factors, such as top management support, indeed play an important role in ISA. Further, the effects of organizational factors differ significantly between Australia and China. Future studies are needed to examine the effects of additional organizational factors and if the results still hold for the adoption of other Internet standards.

11 Although previous literature has not explicitly examined the effect of national culture on the relationship between complexity and adoption, it has shown that complexity has a negative effect on technology adoption in various cultures, such as cloud computing by enterprises in India [45], communication technology adoption by hotels in South Africa [46], and data warehousing adoption by firms from the US [47]. Nevertheless, the effect of complexity on technology adoption may depend on the culture, context, and technology examined.

9

Information & Management xxx (xxxx) xxx–xxx

X. Wang, S. Zander

[13] Google, Google IPv6 Statistics. (2016), (2016) http://www.google.com/intl/en/ ipv6/statistics/ , Retrieved 18 November 2016. [14] E.M. Rogers, Diffusion of Innovations, 2nd ed., The Free Press, New York, NY, 1983. [15] L.G. Tornatzky, K.J. Klein, Innovation characteristics and adoption–implementation: a meta-analysis of findings, IEEE Trans. Eng. Manage. 29 (1) (1982) 28–45. [16] A. Hovav, D. Schuff, Global diffusion of the internet V – the changing dynamic of the internet: early and late adopters of the IPv6 standards, Commun. Assoc. Info. Syst. 15 (2005) 242–262. [17] P. Dell, Two economic perspectives on the IPv6 transition, Info 12 (4) (2010) 3–14. [18] J. Wu, J.H. Wang, J. Yang, CNGI-CERNET2: an IPv6 deployment in China, ACM SIGCOMM Comput. Commun. Rev. 41 (2) (2011) 48–52. [19] Chinatechnews, Chinese Government Will Invest CNY20 Billion To Promote IPv6, (2014) https://www.chinatechnews.com/2014/04/30/20482-chinesegovernment-will-invest-cny20-billion-to-promote-ipv6 . Accessed 28 November 2016. [20] Information Management Office, A Strategy for the Implementation of IPv6 in Australian Government Agencies, (2009) https://www.finance.gov.au/sites/ default/files/Endorsed_Strategy_for_the_Transition_to_IPv6_for_Australian_ Government_agencies.pdf . Accessed 28 Nocember 2016). [21] G. Hofstede, Culture's Consequences: Comparing Values, Behaviors, Institutions and Organizations Across Nations, Sage, Beverly Hills, CA, 2001. [22] D. Li, P.Y.K. Chau, C.V. Slyke, A comparative study of individual acceptance of instant messaging in the US and China: a structural equation modeling approach, Commun. Assoc. Info. Syst. 26 (2010) 85–106. [23] Y. Li, X. Wang, X. Lin, M. Hajli, Seeking and sharing health information on social media: a net valence model and cross-cultural comparison, Technol. Forecasting Soc. Change (2017) in press. [24] S. Gosain, Enterprise information systems as objects and carriers of institutional forces: the new iron cage? J. Assoc. Info. Syst. 5 (4) (2004) 151–182. [25] M. Fishbein, I. Ajzen, Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research, Addison-Wesley, Reading, MA, 1975. [26] A. Molla, V. Cooper, S. Pittayachawan, The green IT readiness (G-Readiness) of organizations: an exploratory analysis of a construct and instrument, Commun. Assoc. Info. Syst. 29 (2011) 67–96. [27] A.J. Elliot, C. Sedikides, K. Murayama, A. Tanaka, T.M. Thrash, R.R. Mapes, Crosscultural generality and specificity in self-regulation: avoidance personal goals and multiple aspects of well-being in the United States and Japan, Emotion 12 (5) (2012) 1031–1041. [28] P.T.P. Wong, L.C.J. Wong, C. Scott, Beyond stress and coping: the positive psychology of transformation, in: P.T.P. Wong, L.C.J. Wong (Eds.), Handbook of Multicultural Perspectives on Stress and Coping, Springer, New York, NY, 2006, pp. 1–26. [29] H.M. Lai, I.C. Lin, L.T. Tseng, High-level managers’ considerations for RFID adoption in hospitals: an empirical study in Taiwan, J. Med. Syst. 38 (3) (2014) 1–17. [30] C.H. Lin, I.C. Lin, J.S. Roan, J.S. Yeh, Critical factors influencing hospitals' adoption of HL7 version 2 standards: an empirical investigation, J. Med. Syst. 36 (3) (2012) 1183–1192. [31] Y. Chen, F.M. Zahedi, Individuals’ internet security perception and behaviors: polycontextual contrasts between the United States and China, MIS Q. 40 (1) (2016) 205–222. [32] P.M. Podsakoff, S. MacKenzie, J. Lee, N.P. Podsakoff, Common method biases in behavioral research: a critical review of the literature and recommended remedies, J. Appl. Psychol. 88 (5) (2003) 879–903. [33] M.K. Lindell, D.J. Whitney, Accounting for common method variance in cross-sectional research designs, J. Appl. Psychol. 86 (1) (2001) 114–121. [34] K. Ramamurthy, G. Premkumar, M.R. Crum, Organizational and interorganizational determinants of EDI diffusion and organizational performance: a causal model, J. Organ. Comput. Electron. Comm. 9 (4) (1999) 253–285. [35] Y. Wang, H. Li, C. Li, D. Zhang, Factors affecting hotels' adoption of mobile reservation systems: a technology-organization-environment framework, Tour. Manage. 53 (2016) 163–172. [36] K.S. Soliman, B.D. Janz, An exploratory study to identify the critical factors affecting the decision to establish Internet-based interorganizational information systems, Info. Manage. 41 (6) (2004) 697–706. [37] E. Ozgen, R.A. Baron, Social sources of information in opportunity recognition: effects of mentors, industry networks, and professional forums, J. Bus. Venturing 22 (2) (2007) 174–192. [38] G. Churchill, A paradigm for developing better measures of marketing constructs, J. Market. Res. 16 (1) (1979) 64–73. [39] J.P. Robinson, P.R. Shaver, L.S. Wrightsman, Criteria for scale selection and evaluation, in: J.P. Robinson, P.R. Shaver, L.S. Wrightsman (Eds.), Measures of Personality and Social Psychological Attitudes, Academic Press, San Diego, CA, 1991. [40] T.A. Brown, Confirmatory Factor Analysis for Applied Research, The Guilford Press, New York, NY, 2006. [41] P.A. Rogerson, Statistical Methods for Geography, Sage, London, UK, 2001. [42] M. Keil, B.C.Y. Tan, K. Wei, T. Saarinen, V. Tuunainen, A. Wassenaar, A crossCultural study on escalation of commitment behavior in software projects, MIS Q. 24 (2) (2000) 299–324. [43] G. Huston, IPv6: Extinction, Evolution or Revolution? (2006) http://www.circleid. com/posts/ipv6_extinction_evolution_or_revolution . Assessed 28 November 2016). [44] V. Venkatesh, M. Morris, G. Davis, F. Davis, User acceptance of information technology: toward a unified view, MIS Q. 27 (3) (2003) 425–478. [45] H. Gangwar, H. Date, R. Ramaswamy, Understanding determinants of cloud computing adoption using an integrated TAM-TOE model, J. Enterp. Info. Manage. 28 (1) (2015) 107–130.

Acknowledgements This work was supported by a grant from the Information Society Innovation Fund (ISIF Asia). We thank George Michaelson, Sylvia Cadena and Siena Perry from APNIC for supporting this work. Appendix A. Measurement New Features (α = 0.83; CR = 0.83; AVE = 0.71) IPv6 provides new features (0.84). Innovative IPv6 applications have great potential (0.85). Relative Advantage (α = 0.89; CR = 0.89; AVE = 0.73) IPv6 has value and links to business drivers (0.80). IPv6 provides benefits to my organization (0.89). There is a strong return on investment to deploy IPv6 (0.87). Complexity (α = 0.69; CR = 0.68; AVE = 0.51) Conversion of existing applications can be the obstacle for IPv6 deployment (0.75). Complexity of infrastructure upgrade can be the obstacle for IPv6 deployment (0.68). Normative Pressure (α = 0.67; CR = 0.68; AVE = 0.52) We adopt IPv6 because of regulatory or industry compliance (0.64). We adopt IPv6 because of industry pressure and global uptake of IPv6 (0.79). Attitude (α = 0.85; CR = 0.83; AVE = 0.72) It will be necessary to transition to IPv6 in the foreseeable future (0.81). It will be essential to transition to IPv6 in the foreseeable future (0.92). Top Management Support (α = 0.93; CR = 0.93; AVE = 0.77) Top management discusses IPv6 as a priority (0.92). Our CIO (or equivalent) plays a leading role in IPv6 planning/deployment (0.84). We have earmarked a budget and other resources for IPv6 (0.89). Our organization demonstrates adequate readiness for IPv6 (0.86). Participation (α = 0.92; CR = 0.92; AVE = 0.86) My organization participates in international fora where IPv6 is discussed and best practices shared (0.92). My organization participates in activities at the national level with industry and academia to foster IPv6 adoption (e.g., World IPv6 Day) (0.93). References [1] BBC, Microsoft Spends $7.5 m on Net Addresses, (2011) http://www.bbc.co.uk/ news/technology-12859585 . Accessed 28 November 2016. [2] M. Nikkhah, R. Guérin, Migrating the internet to IPv6: an exploration of the when and why, IEEE/ACM Trans. Netw. 24 (4) (2016) 2291–2304. [3] A. Hovav, M. Hemmert, Y.J. Kim, Determinants of internet standards adoption: the case of South Korea, Res. Policy 40 (2) (2011) 253–262. [4] A. Hovav, R. Patnayakuni, D. Schuff, A model of Internet standards adoption: the case of IPv6, Inf. Syst. J. 14 (3) (2004) 265–294. [5] X. Wang, S. Brooks, S. Sarker, Understanding green IS initiatives: a multi-theoretical framework, Commun. Assoc. Info. Syst. 37 (2015) 670–704. [6] J. Vykoukal, Grid technology as green IT strategy? Empirical results from the financial services industry, Proceedings of European Conference on Information Systems, Pretoria, South Africa, 2010. [7] P. Dell, Australian IPv6 readiness: results of a national survey, J. Res. Pract. Info. Technol. 44 (1) (2012) 3–15. [8] O. Dobrijevic, V. Svedek, M. Matijasevic, IPv6 deployment and transition plans in Croatia: evaluation results and analysis, Proceedings of 20th International Conference on Software, Telecommunications and Computer Networks (2012) 1–7. [9] J. Pickard, A.Y. Patrick, A. Robinson, Analysis of enterprise IPv6 readiness, Proceedings of the IEEE SoutheastCon, Fort Lauderdale, Florida, 2015, pp. 1–7. [10] G. DeSanctis, M.S. Poole, Capturing the complexity in advanced technology use: adaptive structuration theory, Organ. Sci. 5 (2) (1994) 121–147. [11] J.K. Sinclaire, C.E. Vogus, Adoption of social networking sites: an exploratory adaptive structuration perspective for global organizations, Info. Technol. Manage. 12 (4) (2011) 293–314. [12] S. McCoy, D.F. Galletta, W.R. King, Integrating national cultre into IS research: the need for current individual-level measures, Commun. Assoc. Info. Syst. 15 (2005) 211–224.

10

Information & Management xxx (xxxx) xxx–xxx

X. Wang, S. Zander

Social Change, Communications of the Association for Information Systems, Journal of Organizational Computing and Electronic Commerce, Behaviour & Information Technology, Journal of Computer Information Systems, and Journal of Knowledge Management.

[46] N. Mndzebele, The effects of relative advantage, compatibility and complexity in the adoption of EC in the hotel industry, Int. J. Comput. Commun. Eng. 2 (4) (2013) 473–476. [47] K. Ramamurthy, A. Sen, A.P. Sinha, An empirical investigation of the key determinants of data warehouse adoption, Decis. Support Syst. 44 (4) (2008) 817–841. [48] H.C. Triandis, R. Bontempo, M.J. Villareal, M. Asai, N. Lucca, Individualism and collectivism: Cross-cultural perspectives on self-ingroup relationships, J. Pers. Social Psychol. 54 (2) (1988) 323–338.

Sebastian Zander received the Dipl.-Ing. degree in applied computer science from Technical University Berlin, Germany in 1999 and the PhD degree from Swinburne University of Technology, Melbourne, Australia in 2010. From 1999 to 2004 he worked as researcher and project manager for Fraunhofer FOKUS in Germany and from 2010 to 2015 he was a Research Fellow at Swinburne University. Since 2015 he is a Lecturer at Murdoch University, Perth, Australia. He co-authored Information Hiding in Communication Networks (Wiley, 2016). His research interests include the IPv4 to IPv6 transition, network measurement, traffic classification, covert channels and network security. Sebastian Zander is a member of the IEEE and the Australian Computer Society (ACS).

Xuequn (Alex) Wang is a Lecturer in Murdoch University. He received his Ph.D. in Information Systems from Washington State University. His research interests include social media, online communities, knowledge management, and human-computer interaction. His research has appeared (or is forthcoming) in Technological Forecasting and

11