• No results found

Individual and Organizational Antecedents of Technology Strategy Adoption in High Technology Firms

N/A
N/A
Protected

Academic year: 2023

Share "Individual and Organizational Antecedents of Technology Strategy Adoption in High Technology Firms"

Copied!
57
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Individual and Organizational Antecedents of Technology Strategy Adoption in High Technology

Firms

Supervisor Professor Robert Morgan

Date 30-june-2021

Study EPMS Digital Business

Author and student number Jasper Dijkstra 12000442 Author email dijkstra_jasper@outlook.com EBEC approval number EC 20210131110129

(2)

Acknowledgement

This study has been undertaken as final thesis of the master track Digital Business as part of

Executive Program for Management Studies by the University of Amsterdam/ Amsterdam Business School. It could not have been concluded without the support of Professor Robert Morgan as

supervisor, Drs. Laura Keessen, Program Manager at the Amsterdam Business School and Mr. Daniel Bosman.

This document is written by Jasper Dijkstra, student Digital Business, student number 12000442, who declares that the text and research in this document is original – except for the sources as mentioned and referred to within the document. The Amsterdam Business School is responsible solely for the supervision on the process and facilitation of access to the scientific articles used for this study and access to the Qualtrics platform. It is not responsible for any of the content.

(3)

Abstract

The implementation of new strategies does not always bring as much improvement as intended, resulting in technology strategies being replaced without having reached their full potential. Within high-technology firms the focus is in general on the technological aspect of a strategy. The purpose of this study is to better understand the relationship between technology implementation strategies and technology strategy adoption within a high-technology firm, to identify and provide insight on what behavioral aspects high-technology firms can focus on to positively affect the adoption of technology strategy. Based on previous studies, a conceptual model is defined in which several individual and organizational antecedents affect Perceived Ease of Use and Perceived Usefulness, which positively affect the adoption of Transactional Technology and Collaborative Technology. Data from over 150 users was generated following survey within a focal firm. Analyses showed support for the main hypotheses that Engagement—which can be influenced by the firm and included in the firm’s strategy—via Perceived Ease of Use and Perceived Usefulness positively affects the adoption of technology.

(4)

Contents

1. Tables and figures ... 5

2. Purpose and significance ... 6

Introduction ... 6

Research problem ... 7

3. Theory, conceptual model & hypotheses ... 8

Introduction ... 8

Theory ... 8

Conceptual Model ... 10

Hypotheses ... 13

4. Methodology ... 17

The focal firm ... 17

Measurement method ... 18

Survey design ... 21

5. Analysis & results ... 24

6. Discussion ... 29

7. Conclusion ... 31

Conclusions ... 31

Limitations ... 32

Implications for future research ... 32

Implications for the firm management ... 33

8. Reference List ... 34

Appendix 1: Survey ... 40

Appendix 2: Survey invitation ... 57

(5)

1. Tables and figures

Table 1: Hypotheses ... 17

Table 2: Scales for control variables ... 22

Table 3: Factor analysis of the scales used in this study. ... 24

Table 4: Regression analysis of the effects of Engagement, Open Mindedness, Learning Commitment, Self-Efficacy and the relevant interaction effects on Perceived Ease of Use and Perceived Usefulness. ... 26

Table 5: Regression analysis of the effects of Perceived Ease of Use and Perceived Usefulness on Transactional Adoption and Collaborative Adoption ... 27

Table 6: Hypothesis testing ... 28

Table 7: Instrument items for measuring Engagement ... 29

Table 8: the Self-Efficacy instrument and the Learning Commitment instrument. ... 30

Figure 1: Conceptual Model ... 12

(6)

2. Purpose and significance Introduction

During the early decades of Information Technology businesses had to develop their own applications, which was in general fitting the for that business available business processes and platforms. Not far behind the growth of Information Technology, businesses emerged which specialized in creating digital technology, not only custom made to a specific business, but also creating and selling cheaper digital solutions as “standard” packages, which in some cases could be partially customised to serve a wide range of different businesses. The larger the digital technology providers became, the less motivation for these businesses to provide customization options. The emergence of these standard technology providers created a choice for business, and therewith also created another challenge; a business could chose to pay a high price for custom designed technology, or pay less for tailored technology, and even less for off-the-shelf technology, with the result that scope and requirement definition – which requires a deep understanding of the business operations and technologies available- became more critical, and the selection of the technology could force businesses to modify their processes and environment -including employees- to fit the technology, which can easily lead to misfit of Information Technology within a business (Strong & Volkoff, 2010).

With Information Technology and the issue of alignment, also the number of studies and research regarding the subject grew. Multiple studies discuss that in order to benefit from the continuous growth of Information Technology, Information Technology should be strategized and the IT strategy should be aligned and integrated with the businesses other strategies, which will lead to more business profitability (Avison, Jones, Powell, & Wilson, 2004; Croteau & Bergeron, 2001; Henderson &

Venkatraman, 1999; Porter, 1989; Smaczny, 2001; Venkatesh, Morris, Davis, & Davis, 2003). The implementation and alignment of Information Technology is considered a strategical issue (Cândido

& Santos, 2015; Leonard, 2011). Avison et al write that “Alignment is seen to assist a firm in three ways: by maximizing return on IT investment, by helping to achieve competitive advantage through IS, and by providing direction and flexibility to react to new opportunities.” (Avison et al., 2004,

(7)

p.225). It is experienced that within high-technology firms, implementation and alignment of

Information Technology strategies so far remains a challenge, more due to due lack of organizational fit than on the technical alignment. Apart from the technical alignment, implementation of new technology faces less-tangible challenges. A large factor in the success of any software

implementation is the adoption and acceptance of the system by its users. One of the first successful models capturing the concept of user acceptance is the Technology Acceptance Model. In the Technology Acceptance Model, user acceptance is affected by the constructs of Perceived Ease of Use and Perceived Usefulness (Davis, 1989).

Research problem

The gap between the implementation of technology strategy and its users – a lack of adoption of technology strategy- is often a reason that implementations do not meet the full potential in the amount of time which was foreseen. Due to this, strategy formulation and implementation has become a market in which many consultancy firms are active (Payne, 1986; Van den Bosch, Frans A. J., Baaij,

& Volberda, 2005; McKenna, 2012). Within business, especially in rapid changing markets such as high-technology markets, this can result in strategies and software systems being phased out or replaced while the previous strategy implementation was never finished, leaving potential

unharvested. Due to the nature of the business within high-technology firms, the focus generally is on technology itself, and less on the human and organizational aspect (Dugal & Schroeder, 1995; Rosen, Schroeder, & Purinton, 1998; Haverila, 2013). Implementation of technology or strategy has been the subject of many studies, providing insights and guidelines on how on higher levels technology

strategy can be aligned with the business strategy, on how to implement and align technology with the organization and on how to implement a specific software system such as Enterprise Resource

Planning technology (Aladwani, 2001; Bingi, Sharma, & Godla, 1999), though no specific studies have been found providing insights on “low level” and insights on what practical actions on the soft side a firm can include in its technology implementation strategy. The purpose of this study is to identify and provide insight on what behavioral and soft aspects within high-tech business a firm can leverage to improve the adoption of technology strategy.

(8)

3. Theory, conceptual model & hypotheses Introduction

With the emergence of Information technology, the field of psychology became an important aspect in the use of Information Systems and technology strategy adoption. Many studies linked psychology and Information Systems, focusing on the human aspect; the use, -development, and success of Information Systems from a psychological point of view. The realization came that implementation of Information Systems and technology adoption requires more than just providing a specific tool but requires a strategy on how to implement and use the new tool. In the past decades multiple theories and models have been defined in order to predict the use and adoption of technology, as well as models on how strategy should be implemented (Davis, 1987; Ajzen, 1991; Okumus, 2001; Atkinson, 2006; Kazmi, 2008). This study is focusing on the implementation and adoption of technology strategy implementation within a high-technology firm.

Theory

In the 1980’s and 1990’s, the increase in growth of Information Technology also resulted in the development of studies and theories around the subject. In the article in which Henderson &

Venkatraman (1999) present their Strategic Alignment Model, they state that “IT is transcending its traditional "back office" role and is evolving toward a "strategic" role with the potential not only to support chosen business strategies, but also to shape new business strategies.” (Henderson &

Venkatraman, 1999, p.472). Henderson & Venkatraman reason that instead of just being a tool supporting the business, Information Technology should have a strategy and be part of the overall business strategy. Whilst most of the strategy studies published during the emergence of Information Technology were about the formulation of strategy and how Information Technology should be strategized, practice presented the resulting issue; after formulating strategy, the challenge is the implementation of strategy (Gupta & Govindarajan, 1984; Alexander, 1985; Atkinson, 2006;

Hrebiniak, 2006). This led to new studies focusing on the actual implementation and adoption of strategy. As a result, multiple successful theories on Technology Adoption and Acceptance on individual level have been published, providing models to predict individual intention to adopt technology, popular ones being the Theory of Planned Behavior and the Technology Acceptance

(9)

Model – and based on these, the Unified Theory of Acceptance and Use of Technology (Venkatesh et al., 2003). The Theory of Planned Behavior – also known as TPB (Ajzen, 1991) states that a user’s intention is the most important factor driving actual behavior, and that intention is a result from the different components attitude, subjective norms and perceived behavioral control. The Theory of Planned Behavior includes internal (individual) and external (environmental) factors influencing behavior. The Technology Acceptance Model – also known as TAM (Davis, 1989) states that the two constructs Perceived Usefulness (PU) & Perceived Ease of Use (PEU) effect the attitude towards using new (computer) technology, and that attitude influences actual usage. Both the Technology Acceptance Model and the Theory of Reasoned Behavior include internal factors -respectively Attitude Towards Using and Intention -influencing the outcome variables; actual system use and behavior. It can be discussed whether the attitude towards/ intention to adopt new technology is relevant if the actual adoption itself is the construct of interest, since earlier Technology Acceptance Model and Theory of Reasoned Behavior-based studies show a significant and positive relation between the intent and the actual use. In the Technology Acceptance Model, Perceived Usefulness &

Perceived Ease of Use are affected by external factors. Follow up studies of the Technology

Acceptance Model and the Theory of Reasoned Behavior extended the model with multiple individual or internal variables affecting Perceived Usefulness and Perceived Ease of Use, such as with

Experience, Age & Education (Burton-Jones & Hubona, 2006). This is valuable and useful theory, however, does not support factors which can be easily controlled or influenced directly by a firm or organization without adapting a business’s hiring policy. The Unified Theory of Acceptance and Use of Technology combines the Technology Acceptance Model and the Theory of Reasoned Behavior and besides constructs similar to Perceived Ease of Use (Effort Expectancy) and Perceived Usefulness (Performance Expectancy), attributes adoption to Social Influence and Facilitating Conditions. Here however, social influence and facilitating conditions are treated as how they are perceived by the users (Venkatesh et al., 2003). Based on the forementioned research and experience within high-tech businesses, it is hypothesized that multiple other internal and external constructs affect the Perceived Ease of Use and Perceived Usefulness, which at its turn directly affect the actual adoption of new technology strategies, of which a part can be affected by the business more strongly than the others;

(10)

Engagement, Open Mindedness, Learning Commitment and Self Efficacy, the latter which is described by its author as “People’s self-efficacy beliefs determine their level of motivation, as reflected in how much effort they will exert in an endeavor and how long they will persevere in the face of obstacles.” (Bandura, Freeman, & Lightsey, 1999).

Conceptual Model

From the purpose of this study to identify and understand the relation of individual- and

organizational antecedents with technology strategy adoption, it was decided to present the constructs of Engagement and Open Mindedness as independent variables in the conceptual model. Engagement, a combination of Buy-In and Championing Behavior, is a construct in the model which can be

influenced by the organization (Noble & Mokwa, 1999, p.63; Bassellier, Benbasat, & Reich, 2003), and support for any hypotheses involving engagement can be influenced by the organization will provide useful insights, which is the reason the place Engagement as independent variable within this study. Since in earlier research, the effect of the multiple internal constructs as mentioned in the previous paragraph on Perceived Ease of Use and Perceived Usefulness is shown, it is expected that adding these constructs in the model will improve the model by being able to isolate the effect of Engagement from other constructs influencing the Perceived Ease of Use and Perceived Usefulness (Ajzen, 1991; Ajzen, 2002; Venkatesh et al., 2003; Burton-Jones & Hubona, 2006) – and via those, the mediated effect on adoption. Though the constructs Self-Efficacy, Learning Commitment and Open Mindedness are different from each other as shown in earlier research, they are all

psychological constructs related to a person’s attitude and character and can be considered closely related. As the name of the construct Open Mindedness already suggests, it is a construct not related to any action and can be considered as a passive construct, while Self-Efficacy and Learning

Commitment - not organizational commitment to learning (Sinkula, Baker, & Noordewier, 1997, p.309) - are related to actions; Self-Efficacy is about a person’s character trait and perception on undertaking an activity (Bandura et al., 1999; Chen, Gully, & Eden, 2001; Ajzen, 2002; Heslin &

Klehe, 2006). Though the construct of (individual) Learning Commitment is also inherent to a person’s character, it is a construct a person can directly influence, by for example scheduling time to spend on learning (Tsai, Yen, Huang, & Huang, 2007). Since Self-Efficacy and Learning

(11)

Commitment are shown to be related to perception but are more related to actions than Engagement and Open Mindedness are, it is chosen to place these constructs in the conceptual model as

moderating variables. Moderating effects of the more distinct variables Engagement and Open Mindedness on the Perceived Ease of Use and Perceived Usefulness. As taken over from the earlier mentioned models (Technology Acceptance Model, Theory of Reasoned Behavior and the Unified Theory of Acceptance and Use of Technology), also the effect of perception on adoption is included, being the actual result which this study is focusing on. This results in the following conceptual model:

(12)

Figure 1: Conceptual Model

(13)

Hypotheses

Within this research, the outcome variable adoption is split into two groups of technology strategy adoption: Transactional Adoption (TA) and Collaborative Adoption (CA). Transactional Adoption is the level of adoption of technology strategy providing a solution for the handover of information;

mono-directional or simplex; data created from one source being transferred towards somewhere else.

Collaborative Adoption is the level of adoption of technology strategy providing solutions for collaboration; simultaneous, multi-user creation and/or handling of data within the same period and environment. As based on the Technology Acceptance Model and experience within high-technology firms, the expectation is that Perceived Ease of Use positively affects Transactional Adoption (H1a) and that Perceived Ease of Use positively affects Collaborative Adoption (H1b); if a technology strategy is highly perceived as easy to use, the mental threshold for users to start experiencing &

adopting the strategy is lower, for as well as Transactional Adoption and Collaborative Adoption (Davis, 1989; Davis, Bagozzi, & Warshaw, 1989; Subramanian, 1994; Venkatesh & Davis, 1996). A significant difference of the effect of Perceived Ease of Use on Transactional adoption versus the effect of Perceived Ease of Use on Collaborative Adoption can be expected; for adoption of transactional technology strategy, the user itself, is not in the same way as with collaborative technology strategy, dependent on others to adopt the strategy; if the user itself is able to work with the transactional technology strategy, it is not directly relevant whether other users experience the same or a higher level of Perceived Ease of Use. For adoption of collaborative technology strategy, it can be reasoned that if a user has a high Perceived Ease of Use, but its peers do not, the group might decide (active or passive) that it will not use the strategy and therewith also influencing the specific user not to adopt the strategy (Venkatesh & Davis, 1996; Sarker, Valacich, & Sarker, 2005;

Kroenung, Eckhardt, & Kuhlenkasper, 2017). Perceived Usefulness is expected to positively affect Transactional Adoption (H2a) and Perceived Usefulness is expected to positively affect Collaborative Adoption (H2b); if a user perceives a high level of usefulness of a technology strategy, the user will likely expect that adopting the strategy will bring benefit to the user, such as reducing effort in a work process or improve quality. In line with the reasoning on the effect of Perceived Ease of Use on Transactional Adoption and Collaborative Adoption, here the difference of the effect of Perceived

(14)

Usefulness on Transactional Adoption and Collaborative Adoption is expected the other way around;

if a user perceives a relatively low usefulness of a technology strategy, but his or her peers have a higher perception of the strategy, a group might decide to adopt the strategy and pulling the specific user with a lower perception of the usefulness along, resulting in a relatively higher adoption rate of collaborative technology strategy (Brown, Dennis, & Venkatesh, 2010). The construct Engagement is expected to directly and positively affect Perceived Ease of Use (H3a) and Engagement is expected to directly and positively affect Perceived Usefulness (H3b); Engagement is the level of involvement and motivation an employee has -in this case to adopt technology strategy. Being a mixture of parts of the constructs Buy-In (commitment and a positive affect towards a strategy)- and Championing Behavior (Noble & Mokwa, 1999, p.63; Stuart, Mills, & Remus, 2009)- “the extent to which it is perceived that a strategy is being led through the implementation process by a specific individual.”-(Noble &

Mokwa, 1999, p.63), engagement is a construct which can actively be affected by the organization; if the firm creates and shows the benefit for the users, they are increasing Buy-In. If a key person involved in the implementation shows genuine championing behavior by taking the lead, encouraging

& motivating users to actively participate in the technology strategy implementation, this is creating engagement (Macey & Schneider, 2008). With a higher level of engagement users will spend more time on a new technology strategy and will likely become more familiar and comfortable with the new strategy and perceive the strategy as easy to use and useful more than users with low

engagement. Past studies have included the concept of Open Mindedness in relation to change, amongst other the studies of Baker & Sinkula (Baker & Sinkula, 1999; Sinkula et al., 1997, p.309), who state that organization’s learning capabilities depend on multiple factors, one being the

organizational and collective value Open Mindedness, and that Open Mindedness is a prerequisite or essential requirement for unlearning, and that unlearning is “at the heart of organizational change”

(Sinkula et al., 1997, p.309). However studies modeling the effect of Open Mindedness itself onto technology strategy adoption so far have not been found, while it is experienced and expected that individual Open Mindedness has a positive effect on Perceived Ease of Use (H4a) and Open Mindedness has a positive effect on Perceived Usefulness (H4b); it is expected that the higher the level of open mindedness, for an individual or a business, the less prejudice towards a new technology

(15)

strategy, the more faith in the decision on implementation by the decision makers – the user with a higher level open mind is more likely to trust that a new technology strategy which is being implemented will be easy to use and will be useful to its future users - and the higher the Perceived Ease of Use and Perceived Usefulness. The constructs Self-Efficacy and Learning Commitment are expected to moderate the effect of Engagement and Open Mindedness on Perceived Ease of Use and Perceived Usefulness. Self-Efficacy is expected to negatively moderate the effect of Engagement on Perceived Ease of Use (H5a); according to Bandura the level of Self-Efficacy of a person affects the amount of effort a person will put into overcoming endeavors (Bandura et al., 1999). If a person has a high level of Self-Efficacy, this person is likely to have put in more effort into overcoming past endeavors and has been more likely to succeed in overcoming them, influencing the persons attitude towards new endeavors and perceiving them as a challenge, not as an issue (Bandura et al., 1999;

Chen et al., 2001; Heslin & Klehe, 2006). When a business is actively stimulating engagement for a new technology strategy, more than was done in previous implemented strategies, a person having overcome endeavors in the past might expect that the technology strategy being promoted more actively now will be more complicated and that that is the reason why the firm is attempting to stimulate engagement, hence the expectation that Self-Efficacy negatively moderates the effect of Engagement on Perceived Ease of Use. Via the same reasoning, Self-Efficacy is expected to positively moderate the effect of Engagement on Perceived Usefulness (H5b); users with a higher Self-Efficacy are likely to reason that the stimulation of engagement by the business is because the usefulness of the technology strategy is high compared to other strategy implementations, hence the moderation effect of Self-Efficacy on the effect of engagement on Perceived Usefulness is expected positive. Self- Efficacy is expected to positively moderate the effect of Open Mindedness (OM) on Perceived Ease of Use (H6a); for a person with a higher level of SE, the effect of Open Mindedness on Perceived Ease of Use is expected to be higher. For a user who has a high level of Open Mindedness, but a low level of Self- Efficacy, the Perceived Ease of Use is expected to be lower than for a user with high Self- Efficacy, since the user has relative more difficulties dealing with endeavors and challenges compared to its peers and will based on his/her experience likely expect that the technology strategy isn’t as easy to use for him or her than for peers (Heslin & Klehe, 2006). For Perceived Usefulness the same

(16)

applies, it is expected that Self-Efficacy positively moderates the effect of Open Mindedness on Perceived Usefulness (H6b); a user with higher Self-Efficacy has likely gained deeper experience with technology strategies than a user with low Self-Efficacy and therefrom gained better

understanding of the technology strategy, hence perceives more usefulness. The effect of Engagement on Perceived Ease of Use is expected to be positively moderated by Learning Commitment (H7a); a person who is highly committed to learning and has a high level of engagement will have a more positive attitude towards learning how to work with the new technology strategy, more than a person which is highly engaged but has lower learning commitment. Engagement provides to drive to adopt a new technology strategy, while the learning commitment is a tool to do so, hence the higher the Learning Commitment, the higher the effect of Engagement on Perceived Ease of Use (Tsai et al., 2007; Malik & Kanwal, 2018). Via the same reasoning, the effect of Engagement on Perceived Usefulness is expected to be positively moderated by Learning Commitment (H7b); high learning commitment will reinforce the positive effect of Engagement on Perceived Usefulness; Engagement leads to a person spending more time on a new technology strategy and Learning Commitment helps the person to better understand and master the new strategy and therewith will perceive more usefulness than a person with lower Learning Commitment would, hence the moderation effect of Learning Commitment on the effect of Engagement on Perceived Usefulness is expected to be

positive (Macey & Schneider, 2008). Learning Commitment is expected to have a positive moderation effect on the effect of Open Mindedness on Perceived Ease of Use (H8a); the effect of Open

Mindedness on Perceived Ease of Use will be higher for a person having a high learning commitment.

The Open Mindedness of a person makes that a person is open and unprejudiced towards a new technology strategy and is likely to take longer time to form an opinion about a new technology strategy (Sinkula et al., 1997, p.309; Hernández‐Mogollon, Cepeda‐Carrión, Cegarra‐Navarro, &

Leal‐Millán, 2010). A person with high Learning Commitment will take less time to learn how to use a new strategy so is more likely to have learnt how to use the new strategy before having formed an opinion than users with low Learning Commitment, hence are more likely to arrive at the point where they perceive the new strategy as easy. Similar, Learning Commitment is expected to have a positive moderation effect on the effect of Open Mindedness on Perceived Usefulness (H8b); a person with

(17)

high Open Mindedness will take longer to form an opinion on the usefulness and added value of a new technology strategy than a person with low Open Mindedness. High Learning Commitment will result in that as well persons with high Open Mindedness as persons with low Open Mindedness easier and quicker understand the strategy’s usefulness (Tsai et al., 2007; Malik & Kanwal, 2018).

Table 1: Hypotheses

H1a Perceived Ease of Use positively affects Transactional Adoption +

H1b Perceived Ease of Use positively affects Collaborative Adoption +

H2a Perceived Usefulness positively affects Transactional Adoption +

H2b Perceived Usefulness positively affects Collaborative Adoption +

H3a Engagement positively affects Perceived Ease of Use +

H3b Engagement positively affects Perceived Usefulness +

H4a Open Mindedness positively affects Perceived Ease of Use +

H4b Open Mindedness positively affects Perceived Usefulness +

H5a Self-Efficacy negatively interacts with the effect of Engagement on Perceived Ease of Use - H5b Self-Efficacy positively interacts with the effect of Engagement on Perceived Usefulness + H6a Self-Efficacy positively interacts with the effect of Open Mindedness on Perceived Ease of Use + H6b Self-Efficacy positively interacts with the effect of Open Mindedness on Perceived Usefulness + H7a Learning Commitment positively interacts with the effect of Engagement on Perceived Ease of Use + H7b Learning Commitment positively interacts with the effect of Engagement on Perceived Usefulness + H8a Learning Commitment positively interacts with the effect of Open Mindedness on Perceived Ease of Use + H8b Learning Commitment positively interacts with the effect of Open Mindedness on Perceived Usefulness +

4. Methodology The focal firm

To test these hypotheses, a focal firm was identified to generate data. The focal firm chosen is a Dutch entity of a worldwide operating high-technology Engineering, Procurement & Construction

contractor, having multiple operating & technology centers over the world, divided in Business Units, supplying industrial installations for energy conversion & petrochemical base products. A major part

(18)

of the firm’s employees has a technical background, and the firm heavily depends on the firms knowledge which is held and carried over by its employees. The local entities are set up as a matrix organization; the entities execute projects based on the firm’s own or third party technology, based on a strong operational backbone applying the know-how from the operational departments supported by the firms globally enrolled systems. Due to the specific knowledge and high-technology products of the firms’ portfolio, the firm is positioned in an oligopoly market, where for decades the primary focus was on technology, development of its existing technology and quality while there was not yet the immediate need for a strong focus on efficiency. Since the start of the energy transition however the market is changing and the market for the firm’s traditional products is slowly getting saturated.

The global firm -in the form it was in 2020 over 40.000 employees, and in its new specialized split-off around 20.000 employees- started diverging their product lines along with the energy transition, in which’ market more other players are successful. Since then, the strategy changed towards

optimalization of the business processes for existing and new technology and increasing flexibility to be able to align with the market demands quickly and efficiently, by leveraging digital technology with optimal quality, costs and efficiency. The alignment with the market and creation of new

technologies is highly dependent on the implementation of digital technologies and the adoption by all stakeholders; by alignment of the systems in the internal domain with systems in the external domain – being clients, competitors, suppliers and other technology centers of the firm- the firm is - and will be even better able to efficiently combine technology from different sources with inclusion of all relevant metadata, to provide clients a full package of customized high-technology industrial installations with all the related data included in a single package including 3D (information model), 4D (construction sequencing), 5D (cost and cashflow) and 6D (project lifecycle) models of its technology. During the execution phase of the firm’s projects, where in the old days data used to be transmitted between stakeholders via extracts of the data in the form of documentation, data nowadays is shared via access to a single source of truth, regardless of any physical distance of the stakeholders.

Measurement method

The goal was to collect and analyze sufficient quantitative data on a technology strategy to confirm or reject the hypotheses as per the conceptual model. For measuring the psychological constructs which

(19)

are part of the conceptual model as independent-, moderating- and mediating variables there was no other feasible measurement method besides asking users via a structured direct survey (Coderre, Mathieu, & St-Laurent, 2004; Rindfleisch, Malter, Ganesan, & Moorman, 2008; Malhotra, Nunan, &

Birks, 2017). For the adoption level, besides asking the survey respondents, measuring could be done by using direct data on software usage. In all cases, a measurement method needed to be selected fitting the firm’s setup and systems. After analysis of the technologies available, assessing the expected reachable number of users within the firm’s Dutch entity and the entire Business Unit, the use of the technologies and access to data and resources, multiple potential technology subjects were selected. The first option would be to measure adoption of an engineering technology, such as one of the 3D design-, simulation- and calculation systems. Within the Dutch entity, technologies from a few providers are being used, depending on client requirements. Here however there was not a single engineering technology available which is used by over 100 employees, so data would have to be collected throughout the Business Unit in which other entities use the same technologies. The selection of the engineering tools up to a certain degree is made by local entities’ management, and the available engineering technology strategies have been implemented under different circumstances in different periods with different implementation programs, likely resulting in different levels of adoption. A positive aspect about this option would be that actual adoption levels could be measured from the systems, such as the number of users of a program within a certain period, corrected for the number of users of the departments using this specific technology per different entity of the firm. A negative aspect in this option is that the difference in circumstances will influence the results

decreasing the chance of significant outcomes of the data analysis. An alternative option would be to measure adoption of a technology strategy which is decided on and enrolled by the firm’s

headquarters. This mostly regards intra-firm developed technologies for daily operations supporting project execution, specifically designed for control and reporting such as hour booking, cost control, procurement of direct- and indirect goods and services, and financial reporting, serving the local entities business as well as headquarters strategies. An aspect of this option is that for the mentioned activities, in general there is no alternative technology available or allowed within the firm,

theoretically resulting in an adoption rate of 100% - since for example hour booking for all employees

(20)

is done within the same Enterprise Resource Planning system. In case that this option would have been selected, an option would be to ask the survey respondents whether they would choose an alternative technology if available. This would require that the alternative technology would be a technology the respondents would be familiar with from the past, or that the alternative technology would be fictional and described into a high level of detail, which would ask a lot of time and effort for the respondents to judge accurately, and still would have a high chance to lead to less accurate results than other measurement methods. A third option, which is eventually chosen, is to measure the adoption of a supportive, not job-related technology strategy, which can be used by all employees and for which alternatives were available. A suitable technology strategy subject turned out to be the Microsoft Office 365 suite. This technology is used by sufficient employees with the Dutch entity;

hence the research could be limited to the Dutch entity of the firm – from here on the focal firm being the Dutch entity. During the selection of the technology strategy subject, several factors were

reasoned to influence the adoption level. One being the influence of COVID-19 and the requirement to work from home. Employees were not physically able to visit each other and talk face to face in real life, hence direct live communication was done via other means such as via videocall

applications. Though several other videocall services are freely available, such as Zoom or Cisco WebEx, with Microsoft Office 365 being the firm’s standard office application suit, including the live videocall application Teams, Teams was the only officially supported videocall option. COVID-19 and the working-from-home requirement drastically increased the adoption of Microsoft Teams, despite that other officially supported verbal communication methods were still available, being (the firm’s supplied) mobile phone and the virtual desk phone. Another aspect on measuring adoption of Microsoft Office 365 suite is that for some applications an alternative within the suite is available, and for other applications there is not; if an employee decides to convey a written message, there are several options available within the Office 365 suite; a user can convey a written text via email (Outlook) or via chat message in Teams, but also by sharing a Word document. For some applications more extensive or more powerful alternatives are available but are experienced to be less easy to use and maybe “overkill” for the purpose, for example Microsoft Access, which is specifically designed to create databases, the most common application used for databases within the Office 365 suite is

(21)

expected to be Excel. For other purposes, no alternatives are available, for example to share a calculation, within the Office 365 suite, the obvious application to use is Excel, Which applications were at that time officially part of the Office 365 suite was not clear; this seemed dynamic because of new applications being developed by Microsoft and applications being phased out – such as Skype- and for commercial reasons the Microsoft Office 365 suite can be acquired with different

subscriptions, from only the basic applications to the full Office 365 suite (Microsoft.com, 2021). As shown in the conceptual model and the hypotheses, the theory differentiates between Collaborative Adoption and Transactional Adoption, since those technologies have a different purpose. The variable Collaborative Adoption covers the firm-widely used collaboration applications OneNote, OneDrive and SharePoint, while the variable Transactional Adoption covers Outlook, Teams, Word and Excel.

Survey design

The data is collected via a structured direct survey distributed towards all employees within the focal firm on the Qualtrics platform, facilitated by the University of Amsterdam (Appendix 1: Survey). The population at that time was around 400 employees and by inviting all employees to participate in the survey, the sampling method was voluntary response sampling (Taherdoost, 2016; Malhotra et al., 2017). Several applications within the Office 365 suite were not expected to be used much within the firm and several applications are not available to all employees, though this does not mean users are not familiar with those applications from experience outside of the firm. In order not to bias the results, applications of which the use was doubtful have not been excluded from the survey (Groves et al., 2011; Malhotra et al., 2017). For the survey, the content was decided to cover a total of 18

applications as listed in the survey in Appendix 1. In the instruments used to measure the Independent Variables and moderators related to software implementation, is referred to the Microsoft Office 365 suite in general. The mediators, Perceived Ease of Use and Perceived Usefulness have not been measured for all 18 applications but for the Office 365 suite in general and for SharePoint and OneNote, to limit the requested input from the respondents (Malhotra et al., 2017). It was decided to include SharePoint and OneNote because of their nature and role within the focal firm; SharePoint is used for the focal firm’s Intranet environment and in general seemed not well known (pre-data collection) by the focal firm’s employees, while after recent updates from Microsoft, SharePoint

(22)

includes powerful capabilities and functionality based on the Microsoft Azure platform such as workflow automation, file storage & sharing and multi-user collaboration. OneNote itself is a relative old application but since the link with Azure backbone and OneDrive, it provides online synchronized notebook & collaboration functionality. The extent of adoption has been measured for all separate 18 applications. For the different constructs in the model validated instruments for the specific constructs have been selected, which have been slightly adapted to fit the research design, era and population.

For the independent variable Engagement, items from existing instruments for measuring Buy-In and Championing Behavior have been used (Noble & Mokwa, 1999, p.63). The construct Open

Mindedness was in its origin developed for Open Mindedness of a firm, though is here used to

measure Open Mindedness per person (Sinkula et al., 1997, p.309). Adoption was measured for all the Office 365 tools by asking the respondents to score all the 18 included applications from Office 365.

To limit the size of the survey and to decrease the risk of over-asking the respondents, where possible 7-point Likert scales have been used. It was chosen to use 7 Point Likert scales instead of 5 Point Likert scales because of the higher resolution the 7-point scales provide (Malhotra et al., 2017). The 7-point Likert scales have been defined with the descriptions: Strongly disagree – disagree –

somewhat disagree – agree nor disagree – somewhat agree – agree – strongly agree. Within Qualtrics the items with 7-point Likert scales have been coded as 1 to 7 and where applicable are reversed. For the collection of control data, the items age, education, firm tenure and IT expertise are selected, since they are expected to have an effect on the measured variables. Gender, being a control variable often used, has been excluded because within the high-tech environment of the focal firm, it is not expected to have any effect. The categories were defined as shown in Table 2.

Table 2: Scales for control variables

Age below 30, 30-39, 40-49, 50-59, 60 or above

Highest level of education High school – HBO/ university bachelor – Master – PHd Years working at the firm <5, 5<10, 10<15, 15<20, >20

Self-considered level in IT expertise; 1 to 10

(23)

The survey has been split into multiple blocks which are fixed in a flow; Introduction, Individual, Software, Office 365 and Demographics. A progress bar has been added for user convenience and stimulate respondents to finish the survey – knowing how much more to expect. Because of the fixed workflow users were not able to skip any questions, and with the auto-focus function enabled to text of the current item was shown clearly while other text on the page faded out. Since the firm provides its employees as standard with laptops, docking stations and monitors and the daily operations are mostly done online, it was expected that most respondents would fill in the survey on a large monitor, and there for the survey has not been limited in graphical size to fit on mobile phone, though where possible the survey has been optimized to also support mobile phone respondence (Manfreda, Bosnjak, Berzelak, Haas, & Vehovar, 2008; Malhotra et al., 2017).

The actual data collection window was overlapping the roll-out of a corporate organized program to inform and educate employees on the newest functionalities within applications of Microsoft Office 365 suite and the planning and execution of its first info-and training sessions, referred to as “Wave 1”. The invitation to participate in the survey has been sent via email to the focal firm’s all-employees mailing list (Appendix 2: Survey invitation). The invitation was sent as part of the before mentioned Microsoft Office 365 training program from a personal work email account, to make the invite more personal, hoping to increase the number of respondents compared to a less personal invite from the training program, or compared to an individual action hinting a lack of alignment within the firm. The invitation has also been shared on various displays within the office -since a part of the firm’s

employees was still allowed to work from the office- and on the Intranet main page. It was agreed with the team of the Microsoft Office 365 educational program on when a reminder was to be sent, in order not to be sent within a too close timeframe of any emails from the educational program’s team itself to all the focal firm’s Employees. After one full week of the survey being open and promoted, around 100 responses were recorded. At the agreed moment, a reminder has been sent. After an additional two weeks, the minimum number of preferred responses -over 150- was reached, and the survey had been closed for new entries – with the option for still ongoing entries to finish. A close-out email has been sent to thank all the employees for their input and attention.

(24)

5. Analysis & results

First the data was exported from Qualtrics into SPSS where face validity check has been done.

Incomplete records (from for example respondents who have not finished the survey completely, or respondents who started a new instance from a different computer) have been excluded from the analysis. Face validity did not show any other outliers. After face validity the resulting sample was N=158. To check whether the constructs have been measured as intended, principal component analysis with rotated component matrices (Varimax with Kaiser normalization) have been used for the independent, -mediating variables and the moderating variables. The Pearson correlation was

calculated to show the correlation of the separate items within the total construct scale.

Table 3: Factor analysis of the scales used in this study.

Item description Factor

loading Item-to- total scale correlation*

Engagement (Cronbach’s Alpha =.777)

I see a high level of alignment & support (“buy-in”) for the implementation .577 .710

• I feel there was a tremendous amount of support across the company. .620 .782

• One person in the organization definitely took charge of making this implementation happen. .588 .761

• The implementation had a champion to guide it through the implementation process. .725 .840 Open mindedness (Cronbach’s Alpha =.672)

I highly value open mindedness. .406 .619

I try to think outside the box. .750 .874

I try to be innovative in my work .673 .818

• My original ideas are highly valued in my organization 1

• I am not afraid to reflect critically on the shared assumptions we have about the way we do business. (‘R) 1

• I do not feel comfortable when my view of the world is questioned. (R) 1 Self-Efficacy (Cronbach’s Alpha =.868)

• I will be able to achieve most of the goals that I have set for myself. .471 .678

• When facing difficult tasks, I am certain that I will accomplish them. .663 .792

• In general, I think that I can obtain outcomes that are important to me. .579 .754

• I believe I can succeed at most any endeavor to which I set my mind. .626 .783

• I will be able to successfully overcome many challenges. .673 .807

• I am confident that I can perform effectively on many different tasks. .500 .723

• Even when things are tough, I can perform quite well. .496 .714

• Compared to other people, I can do most tasks very well. 1 Learning commitment (Cronbach’s Alpha =.844)

• I am willing to spend extra time taking part in the internal and external training courses provided

by the company. .578 .755

I am eager to learn more specific knowledge and skills to achieve the job goals. .682 .794

I believe that all the learning opportunities are advantageous to me. .762 .878

I believe that all the learning opportunities are advantageous to the company. .560 .783

To me, being able to learn constantly is very important. .592 .751

Perceived Usefulness Office 365 (Cronbach’s Alpha =.913)

• I think the Office 365 services improve my performance. .927

(25)

I think the Office 365 services improve my productivity. .932

• I think the Office 365 services are useful for my overall work. .911

Perceived Ease of Use Office 365 (Cronbach’s Alpha =.857)

I think learning how to use the Office 365 services is easy. .878

• Learning how to use the Office 365 services requires less mental efforts. .872

• I think the Office 365 services are easy. .902

*All items statistically significant at the .001 level

1 Item deleted from analysis during scale purification process

The rotated component matrices showed that a few items had loadings on other more than the desired component. For Self-Efficacy one item and for Open Mindedness three items have been excluded from further analysis as shown in Table 3. Perceived Usefulness and Perceived Ease of Use were measured on OneNote, SharePoint, and Office 365 as a total, though in the analysis only the

mediators measured on Office 365 have been used. After the factor analysis, regression analysis has been done with 2-way ANOVA. To deal with the multiple independent variables & moderators, several models have been constructed and mean-centering has been used to be able to calculate the interaction effects, as per example done by Atuahene-Gima (Atuahene-Gima, 2005).

First 6 different models were defined to determine the effect of the independent variables and the moderating variables on the mediating variables. Both mediators Perceived Ease of Use and Perceived Usefulness were considered as dependent variables and for each 3 models have been used; in the first model Analysis of Variance (ANOVA) is done for only the controls’ effect on each mediator. On the second model ANOVA is done on the controls + the independent variables and moderators’ direct effect on the Perceived Ease of Use and Perceived Usefulness. Herein the moderators Self Efficacy and Learning Commitment have been treated as independent variables. In the ANOVA for the third model the interaction effect of the moderators has been added. For the interaction effect mean centering has been applied; new variables have been created in SPPS, being the centered versions of the independent variables and the moderators. From the centered versions of the independent and moderation variables, interaction variables have been created by multiplying them. In the third model these interaction variables have been added. For each model the Sum of Squares, R2, the F values and the Degrees of Freedom have been extracted and are included in Table 4.

(26)

Table 4: Regression analysis of the effects of Engagement, Open Mindedness, Learning Commitment, Self-Efficacy and the relevant interaction effects on Perceived Ease of Use and Perceived Usefulness.

Perceived Ease of Use Perceived Usefulness

Variables Hypotheses Model 1 Model 2 Model 3 Model 1 Model 2 Model 3

Control variables

Constant 4.976

(9.592)*** 1.997

(2.243)* 2.490

(2.645)** 5.199

(8.889)*** 2.016

(2.035)* 2.453 (2.347)*

Age -.340

(-4.331)*** -.362

(-4.840)*** -.356

(-4.662)*** -.051

(-.576) -.080

(-.958) -.087 (-1.022)

Education .003

(.027) -.043

(-.433) -.062

(-.616) -.021

(-.175) -.088

(-.804) -.113 (-1.019)

Tenure -.021

(-.366) .053

(.948) .049

(.867) -.170

(-2.676)** -.079

(-1.267) -.080 (-1.271)

IT Skills .188

(3.398)*** .129

(2.316)* .142

(2.507)* .105

(1.688)† .056

(.902) .064 (1.020) Independent variables

Engagement H3a, H3b .202

(2.873)** .230

(3.149)** .303

(3.878)*** .337 (4.162)***

Open Mindedness H4a, H4b .241

(1.168) .228

(1.509) .284

(1.716)† .287 (1.706)†

Learning Commitment .014

(.131) -.025

(-.219) .076

(.645) .012 (.094)

Self-Efficacy .175

(1.477) .121

(.900) .017

(.127) -.009 (-.060) Relevant interaction

effects

Engagement x Self Efficacy H5a, H5b -.019

(-.150) .024

(.176) Engagement x Learning

Commitment H7a, H7b -.076

(-.850) -.156

(-1.574) Open Mindedness x Self

Efficacy H6a, H6b -.178

(-.909) -.131

(-.604) Open Mindedness x

Learning Commitment H8a, H8b -.011

(-.061) .057

(.286)

R2 .265 .370 .384 .121 .265 .285

SS 40.989 57.192 59.349 19.846 43.608 46.767

F value 13.435*** 10.638*** 7.318*** 5.118*** 6.550*** 4.676***

DoF 4/153 8/153 12/153 4/153 8/153 12/153

† p≤ .10; * p≤.05; ** p≤.01; *** p≤0.001

The same method has been applied for the effect of the mediating variables on the dependent variables – the extent of adoption. The communalities from the factor analysis with the rotated component matrices showed that within the different Office 365 applications for which extent of adoption has been measured, there are multiple variables (Office 365 applications) which have a significant loading on the same components. 2 Groups of applications which loadings on the same component have been identified and selected: 1) Outlook, Teams, Word & Excel 2) OneDrive, SharePoint & OneNote. The first group being transactional applications and the second being

(27)

collaborative applications. Hence the original dependent variable extent of adoption has been split into Transactional adoption and Collaborative adoption. The new variables have been created within SPSS and are further used in the following regression analysis. For the regression analysis, 4 models have been created; for each dependent variable the first model of the control variables’ effect on the dependent variables, and a second model where the mediating variables have been added as

independent variables. Again 2-way ANOVA has been done for each model and the R2, Sum of Squares, F-value and Degrees of Freedom have been extracted. The results are shown in Table 5.

Table 5: Regression analysis of the effects of Perceived Ease of Use and Perceived Usefulness on Transactional Adoption and Collaborative Adoption

Transactional Adoption Collaborative Adoption

Variables Hypotheses Model 1 Model 2 Model 1 Model 2

Control variables

Constant 7.081

(23.686)*** 6.323

(11.558)*** 1.802

(2.245)* -2.741 (-1.973)*

Age -.062

(-1.363) -.063

(-1.304) -.165

(-1.359) -.087 (-.706)

Education -.110

(-1.836)† -.118

(-1.999)* .221

(1.374) .182 (1.207)

Tenure -.001

(-.030) .032

(.925) .008

(.090) .109 (1.262)

IT Skills -.005

(-.162) -.013

(-.371) .209

(2.441)* .115 (1.339)

Engagement .065

(1.447) .102

(.897)

Open Mindedness -.005

(-.060) .268

(1.167)

Learning Commitment .020

(.310) .216

(1.330)

Self-Efficacy -.023

(-.320) -.265

(-1.453) Independent variables

Perceived Ease of Use H1a, H1b -.018

(-.355) .273

(2.078)*

Perceived Usefulness H2a, H2b .116

(2.502)* .288

(2.443)*

Relevant interaction effects None

R2 .035 .117 .084 .249

SS 1.387 4.565 24.885 74.010

F value 1.369 1.889* 3.407** 4.747***

DoF 4/153 10/153 4/153 10/153

† p≤ .10; * p≤.05; ** p≤.01; *** p≤0.001

(28)

From the regression analysis results from Table 4 and Table 5 the support of the hypotheses as defined in Table 1 is tested and shown below in Table 6: Hypothesis testing, where the hypotheses with a significant outcome are supported.

Table 6: Hypothesis testing

Hypothesis Hypothesis test

H1a Perceived Ease of Use positively affects Transactional Adoption Not supported

H1b Perceived Ease of Use positively affects Collaborative Adoption Supported H2a Perceived Usefulness positively affects Transactional Adoption Supported H2b Perceived Usefulness positively affects Collaborative Adoption Supported

H3a Engagement positively affects Perceived Ease of Use Supported

H3b Engagement positively affects Perceived Usefulness Supported

H4a Open Mindedness positively affects Perceived Ease of Use Not supported

H4b Open Mindedness positively affects Perceived Usefulness Supported

H5a Self-Efficacy negatively interacts with the effect of Engagement on Perceived Ease of Use Not supported H5b Self-Efficacy positively interacts with the effect of Engagement on Perceived Usefulness Not supported H6a Self-Efficacy positively interacts with the effect of Open Mindedness on Perceived Ease

of Use

Not supported

H6b Self-Efficacy positively interacts with the effect of Open Mindedness on Perceived Usefulness

Not supported

H7a Learning Commitment positively interacts with the effect of Engagement on Perceived Ease of Use

Not supported

H7b Learning Commitment positively interacts with the effect of Engagement on Perceived Usefulness

Not supported

H8a Learning Commitment positively interacts with the effect of Open Mindedness on Perceived Ease of Use

Not supported

H8b Learning Commitment positively interacts with the effect of Open Mindedness on Perceived Usefulness

Not supported

(29)

6. Discussion

The regression analysis shows that within this research a factor of .23 of the Perceived Ease of Use and .337 of the Perceived Usefulness can be explained by the construct Engagement (hypotheses H3a

and H3b). The regression analysis also shows that a factor .273 of Collaborative Adoption (Adoption of collaborative technology strategy) can be explained by the Perceived Ease of Use, and a factor of .288 can be explained by Perceived Usefulness (hypotheses H1b and H2b). For Transactional Adoption (Adoption of transactional technology strategy), the regression analysis shows that a factor of .116 can be explained by Perceived Usefulness (hypothesis H2a). For the effect of Perceived Ease of Use on Transactional adoption no significant results were obtained (hypothesis H1a). The instrument items used to measure the construct Engagement, shown in Table 7 provide more insight.

Table 7: Instrument items for measuring Engagement

1 I see a high level of alignment & support (“buy-in) for the implementation.

2 I feel there weas a tremendous amount of support across the company.

3 One person in the organization definitely took charge of making this implementation happen.

4 The implementation had a champion to guide it through the implementation process.

Summarized, this implies that with technology strategy implementation, visible alignment & support within the firm in combination with one person taking charge of the implementation showing championing behavior, positively affects the Perceived Ease of Use and the Perceived Usefulness, contributing to a higher adoption rate of Transactional and Collaborative technology strategy adoption. This is a construct a firm can actively influence, which hence is worth considering to actively include in implementation strategies. As the wording in the instrument already shows, a key element is what is visible towards the users within the firm; when alignment and support for

implementation of a technology strategy is available, the level of visibility towards the firm’s

employees or the future user of the technology influences the eventual adoption level. In this situation strategy and marketing come together, for which existing research is available. For Perceived

Usefulness, a factor of .287 can be explained by the independent variable Open Mindedness

(hypothesis H4b). For the effect of Open Mindedness on Perceived Ease of Use no significant results were obtained (hypothesis H4a). That Open Mindedness has a positive effect on Perceived Usefulness

Referenties

GERELATEERDE DOCUMENTEN

Conceptual model of cultural dimensions and radical innovation adoption Power distance Individualism Masculinity Uncertainty avoidance High, low-context Radical innovation

The model shows an adjusted R-square of 0.749, which means that 74,9% of the variance in the dependent variable Intention to Use, can be explained using the

Since the establishment of the state of Israel in 1948, Israeli media and the government have been close collaborators working together to gain control of the

The paper reports on a comparative study, conducted in 2013-14, of four higher education systems (the Netherlands, Flanders, North Rhine-Westphalia and Ireland),

The information disclosure of internal control can form a balance mechanism which can reduce the possibility of earnings management and promote the managers to

waveform and frequency are selected, the DEM simulations of dynamic probing are performed to obtain the P- and S-wave velocities of the granular material at various stress states

The implication of the second feature is that in case of direct regu- lation, the policy maker cannot exempt firms with high adoption costs from regulation, while in case of a

A novel finding from the present work was that the acquisition of the novel discrete sequence production task practiced on Day 2 was better for individuals that had prior random