Master Thesis Empirical Analysis of the Impact of Transformational IT Leadership on Adaptive System Use

(1)

Master Thesis

Empirical Analysis of the Impact of Transformational IT Leadership on Adaptive System Use

“Adaptive System Use: User Revisions of IT-based Features and its Interplay with Transformational IT Leadership“ August 23, 2016 FILIP STRACH Student number: s2596350 Landstraat 4 9714 GR Groningen f.strach@student.rug.nl University of Groningen Faculty of Economics and Business

Supervisor dr. J.D. van der Bij

Co-Assessor dr. I. Maris-de Bresser

Word count: 18622

(2)

Abstract

This research puts emphasis on the use of IT features and how such use can be enabled. More precisely, this study investigates the interplay between adaptive system use and transformational IT leadership which so far has not been taken into account and as a consequence was identified as a gap in literature. This is done by adopting a study on adaptive system use (Sun, 2012) and expanding it around transformational IT leadership. Focusing on team members´ perceptions an empirical research study was conducted across international organizations, resulting in a sample size of 120 valid responses. The findings show that transformational IT leadership somewhat affects adaptive system use. More specifically, the findings suggest that transformational IT leadership does not thoroughly add in line to the already defined triggers – novel situations, discrepancies and deliberate initiative – when acting on its own. Though, when applied as moderating factor between the already existent triggers and adaptive system use or its relation towards adaptive system use being moderated itself, transformational IT leadership does exert impact on the given relationships and thus on post-adoptive system behaviours. However, in order to increase generalizability these factors should be further tested among a larger and broader sample. Moreover, future research could focus on additional triggering and moderating variables to identify further potential factors which influence adaptive system use. Finally, this research increases the understanding of the triggering factors on adaptive system use and its interplay with transformational IT leadership. As a result, managers can use the identified information as an instruction on how to affect adaptive system use positively to increase general performance and innovation by leading their employees more consciously.

(3)

Introduction

These days, a call for research which shall help understanding the dynamic relationship between individuals and their experience with technology applications is needed (Benbasat and Barki, 2007), as benefits of IT use arise when people perform their roles more efficiently and simultaneously more effectively (Peppard, Ward and Daniel, 2007). Given that IT-related innovations are expanding more than ever (Burnes, 2014; Cawsey, Deszca and Ingols, 2016) and at the same moment they are representing the biggest expense of capital across organizations (Stratopoulos and Lim, 2010), IT has evolved into one of organizations´ major backbones (Preittigun, Chantatub and Vatanasakdakul, 2012) causing organizations to be innovative in order to remain competitive (Lertpachin, Wingwon and Noithonglek, 2013). Resulting from this, organizations become to a greater extent dependent on IT (Wang, Butler, Hsieh and Hsu, 2008). At the same time, such IT investments do not always live up to their promise (Jasperson, Carter and Zmud, 2005), being costly and rather unsuccessful (Legris, Inghamm and Collerette, 2003). One of the reasons for this low success rate derives from actual underutilization of given IT (Jasperson et al., 2005). Consequently, research started to pay attention on post-adoptive system behaviours, focusing on the dynamic processes of individual IT use behaviours appearing through interaction among users, technologies and institutional properties (Nan, 2011).

Given the importance of IT use in organizations, adaptive system use (Sun, 2012) focuses on users´ feature revisions. More precisely, it is about the kind of features (what) and also about the way (how) these features are used. However, in order to be able to classify a user´s revision Sun (2012) argues that one has to consider one´s features in use which represent one´s basket of already known and familiar features. Doing so, adaptive system use leads to the two following dimensions of feature revision: revising the content of features in use and revising the spirit of features in use.

In order to advance such adaptation behaviours and ergo innovative IT patterns one needs leaders who can foster an environment of creativity and scope (Amabile, Conti, Coon, Lazenby and Herron, 1996; Mumford, Scott, Gaddis and Strange, 2002). Otherwise, innovation cannot arise. Since transformational leaders are leaders who are able to enthuse and motivate employees by addressing common sense to work (Shamir, 1991), they might challenge employees to think beyond their mind-sets which is leading to innovation (Bass and Avolio, 1997). Consistently, Daughtry and Finch (1997) draw the conclusion that transformational leaders are suitable for technological innovations. However, neither Podsakoff, MacKenzie and Bommer (1996) nor Bass (1985) took into consideration how transformational leaders influence their followers on their technology adaptation and use.

(5)

so far there is no clear role of transformational IT leadership in terms of adaptive system use. More specifically, until now leadership in the context of adaptive system use is only specified by mandated use in the form of deliberate initiatives (Sun, 2012), comparable to transactional leadership theories which pay attention on the completion of orders and transactions (Bass, 1985; Burns, 1978). Having this in mind, transformational IT leadership can add to the field of adaptation behaviours since adaptive system use - dealing with system features and thus with technology on a daily basis – as part of an organization´s workaday life is dependent on technology-akin managers (Keen, 1991).

Therefore, the goal of this study is to investigate whether transformational IT leadership acts as an additional trigger of adaptive system use with respect to the original research done by Sun (2012). Second, this study wants to explore whether transformational IT leadership moderates the relationship between triggering variables and adaptive system use.

In other words, this research aspires to further explore adaptive system use by analysing user revisions of IT-based features and its interplay with transformational IT leadership. The goal is to better understand possible impacts transformational IT leaders can have on their employees, since that kind of influence can considerably effect general performance and innovativeness of organizations. Therefore, the following research question is posed:

RQ: Does transformational IT leadership influence adaptive system use and if so in which way?

(6)

Theoretical Background

This section will elaborate on the major building blocks which serve as the foundation for this research in order to generate a basic overview and understanding in the given context. First, adaptive system use and its corresponding concepts will be explained. Subsequently, triggers responsible for adaptive system use are introduced and defined, followed by the concept of transformational IT leadership and its possible ties with adaptive system use. Finally, facilitating conditions and personal innovativeness as contextual IT influences are presented. We build on the study of Sun (2012).

A conceptual overview of this research is presented in Figure 1.

Triggers Novel situations - New Task

- Changes in System Environments - Other People´s Use

Discrepancies

Deliberate Initiatives

Transformational IT Leadership

Adaptive System Use Transformational IT Leadership Facilitating Conditions Control Variables - Age - Gender - Tenure - Education H8a+ H8b+ H8c+ H8d+ H9d+ H9c- H9b+ H9a+ H7a+ H7b+ H7c+ H4+ H5+ H1+ H2+ H3+ H6+ Personal Innovativeness in IT

Figure 1. Conceptual model

Adaptive System Use

(7)

However, in order to understand the concept of adaptive system use it is necessary to review IT features in general since it is the individual who deals with them on a daily basis and can therefore be affected both positively or negatively (Benlian, 2012). Correspondingly, the general effectiveness of IT features is dependent on a user´s experience (Venkatesh and Davis, 1996). In line with this, in addition to his concept of adaptive system use Sun (2012) also introduces the concept of features in use. According to him, features in use should be treated “as the basket of system features that are ready to be used by a particular user to accomplish tasks“ (p.455). These features, established over a variety of used systems together create a so-called ecosystem which allows the user to interact with his/her environment.

In agreement with this, the concept of familiarity pockets, first introduced by Yamauchi and Swanson (2010) can be referenced. In consonance with them, a familiarity pocket “comprises work routines and components accumulated through situated interactive use of the system“ (p.200) and can be illustrated by the following: (1) a user does not define his/her familiarity pocket by the pure enumeration of routines, but also by the moves and/or actions which as reported by Pentland, Feldman and Becker (2009) actually cause the formation of a routine; (2) every single move of a user is interactive and deals as a link between the user and the system; (3) moves outside a user´s familiarity pocket affect its inner life in a way that it is commonly followed by moves within it.

(8)

Table 1

Dimensions and Sub-Dimensions of Adaptive System Use (Sun, 2012)

Revising the Content of Features in Use

A user´s revisions regarding what features are included in his/her Features in Use: what features are used.

Trying New Features Add new features to one´s Features in Use and thus expanding the scope of the Features in Use.

Feature Substituting Replacing features in the Features in Use with other features with similar functions.

Revising the Spirit of Features in Use

A user´s revisions regarding how features in his/her Features in Use are used.

Feature Combining Using features in Features in Use together for the first time. Feature Repurposing Using features in one´s Features in Use in a new way.

Revising the content of features in use. This dimension of adaptive system use is about what kind of features are used based on one´s features in use (Sun, 2012). Furthermore, it is divided into two more sub-dimensions, namely trying new features and feature substituting. Both dimensions can be labelled as being parts of IS-use related activity (Barki, Titah and Boffo, 2007) which is defined as a “set of behaviors individuals undertake concerning a specific task technology-individual-context“ (p.174), being associated with individual adaptation behaviours. According to Beaudry and Pinsonneault (2005), these behaviours represent self-modifying characteristics including learning activities as well as one´s interaction with given IT. Consequently, this set of behaviours deals with actions individuals take on when performing a task for which IT is needed (Goodhue, 1995; Goodhue and Thompson, 1995).

As reported by Sun (2012), trying new features appears when a user expands his/her features in use due to a higher experienced level of system use and thereby can discover new features within the given IT (Hiltz and Turoff, 1981). This extended use as described by Saga and Zmud (1994) enables individuals to use more features within a technology to master more challenging tasks. Though this might lead to struggles in the beginning, over time users will adapt and use more and more features (Robey, Ross and Boudreau, 2002). Correspondingly, trying new features can be understood as a practice which extends one´s knowledge of features.

As noted above, for users it is also possible to substitute features. Sun (2012) states that the replacement of features can either happen physically or psychologically. In the first case old features are no longer accessible and hence have to be replaced, whereas in the latter case the reason lies only within the user. In this spirit, the user consciously decides not to use a set of features anymore and substitutes it, though physically still present.

(9)

repurposing.

Feature combining can be treated as a form of reinvention which in consonance with Rice and Rogers (1980) is “the degree to which an innovation is changed by the adopter in the process of adoption and implementation after its original development“ (p.501). In this case, innovation refers to a tool´s use (Rogers, Eveland and Klepper, 1977), hence to its features. Sun (2012) states that in order for users to combine features, they use features which are familiar to them to build new functionality. In other words, feature combining is about using several features in one´s features in use in conjunction for the first time. This relates to the concept of workarounds which are goal-driven adaptations, improvisations or any other kind of changes which affect one´s existing work system (Alter, 2015).

Feature repurposing however is about using already existent features in novel ways. That is, users might use certain features differently than expected by the developers, hence they extent the features on a voluntary basis (Jasperson et al., 2005). Similarly, Ahuja and Thatcher (2005) argue that such trying to innovate can evolve to successful innovation and in the end leads to an optimization of task performance or organizational processes. Sun (2012) complements that not all of the features in one´s features in use are revisable because of their simplicity and their recognition value and thus adaptive system use may be only performed partly.

Triggers of Adaptive System Use

Based on the research of Louis and Sutton (1991) about switching cognitive gears which introduces three types of triggers encouraging people to reconsider their way of thinking, namely novel situations, discrepancies and deliberate initiatives, Sun (2012) posits that these triggers are the precondition for adaptive system use. The reason for this purpose is that these conditions can cause active thinking which according to Sun (2012) on the other hand is required for active use (Jasperson et al, 2005).

Table 2

Organizational situations provoking a switch in cognitive modes from habits of mind to active thinking (based on Louis and Sutton, 1991)

Condition Individual Group/Organization

Novel Situations (Novelty) Role transitions - entry - promotion - transfer Merger/acquisition Technological change Organizational birth

Discrepancies (Discrepancy) Performance review Role transitions (job loss)

Organizational decline Criminal incidents Deliberate initiatives (Deliberate requests

for active thinking)

Career planning Assessment centre

(10)

Novel situations. As described by Louis and Sutton (1991), novel situations can be divided into two categories, namely the individual and the group/organization level. Hence this study mainly focuses on the individual level when referring to adaptive system use, only the individual level presented by Louis and Sutton (1991) is taken into account. According to them, on the individual level novel situations deal with role transitions and can be understood as “shifts in one´s role, or major reorientations to a currently held role“ (pp.66-65). Moreover, they state that also technological innovations can act as novel situations, which is in line with Benamati, Lederer and Singh (1997) who refer to new technology-affiliated environments. Further, Jasperson et al. (2005) relate to “new features into the IT application“ (p.536), whereas McKersie and Walton (1991) introduce novel ways of how to handle tasks. Based on this, Sun (2012) comes up with three sub-categories of novel situations, which are (1) new tasks; (2) changes in system environment; and (3) other people´s use.

Starting with the first one, Sun (2012) refers to modifications related to tasks at work which can cause triggering effects (Jasperson et al., 2005). An example of this would be when a person uses PowerPoint for the first time to prepare a digital presentation. Followed by this, when referring to changes in system environments Sun (2012) recalls the example of new technologies which replace older ones which likewise can be of a triggering nature. Finally, by other people´s use one´s own experience with a task/feature compared to someone else´s can cause a novel situation. In this case, observation plays an important role in a way that people can learn from their colleagues (Bandura, 1977; Compeau and Higgins, 1995).

Discrepancies. Also when talking about discrepancies, Louis and Sutton (1991) divide it into two levels and as in the prior example, only the individual level is of note. Here they present performance reviews and role transitions, for instance due to job loss, as suitable examples. According to Wong and Weiner (1981), discrepancies are the results of not recognizing one´s experience in an already present cognitive schema. What is more, Sun (2012) compares discrepancies with misalignments mentioned in IS literature, which in agreement with Lyytinen and Newman (2008) can be labelled as gaps. These gaps are considered to be system contingencies which can reduce performance if not recognized. Furthermore, Plowman, Baker, Beck, Kulkarni, Solansky and Travis (2007) define misalignments as the outcomes of changes which are of steady and innocent nature in one component causing a pushing reversal point. An example of a discrepancy could be the scanner not scanning or a smartphone not being able to perform its features. In other words, discrepancies can be regarded as not fulfilled expectations.

(11)

Hastie (1984) and Schön (1983) argue that people react to external causes which in return act as antecedents for active thinking. This can be seen as the opposite of voluntary usage which according to Agarwal and Prasad (1997) is a user´s perception to which extent an adoption decision is non-mandatory. In line with this, Hartwick and Barki (1994) state that although there is the organizational requirement of mandatory use, usage intentions can vary because of a user´s individual willingness to adopt to new situations. This can result from one´s missing perceived usefulness which according to Davis (1989) is the “degree to which a person believes that using a particular system would enhance his or her job performance“ (p.320). When imagining a suitable scenario regarding adaptive system use, an example for deliberate initiatives would be when a person is asked to use new features to perform a task though the same task could be accomplished using the old and familiar features.

Table 3

Types of Triggers (Sun, 2012)

Trigger Definition

Novel Situations Situations where a person encounters things that are unfamiliar, previously unknown, unique, or that appear to be out of ordinary.

Discrepancies A discrepancy represents situations where an unexpected failure, a disruption, or a significant difference exists between expectations and the reality.

Deliberative initiatives The initiatives one takes in response to a request for an increased level of attention, when asked to think, or while being explicitly questioned.

As explained above, Sun (2012) posits that all of the three presented triggers are positively associated with adaptive system use since each of them builds up on cognitive scripts which advance individual behaviour (Bargh, 1989). Or as said by Ashcraft (1998), cognitive processing is about both the mental processes and mental activity of applying processes. Otherwise, de Guinea and Webster (2013) propose that when people engage with available IT and its features only on a regular basis, they fall into automatic patterns which are free of emotions, thoughts and any continuative behaviours. That is, people do not think about what they are doing and why they are doing it because their familiar IT environment behaves as usual (Waller, Johnston and Milton, 2007). Therefore, this research proposes that the presented triggers will enhance adaptive system use because each of them can break existing patterns and enhance new ways of IT feature interaction. This leads to the following hypotheses (Sun, 2012):

H1: Novel Situations (in terms of new tasks / changes in system environment / other people´s use) are positively associated with adaptive system use.

H2: Discrepancies are positively associated with adaptive system use.

(12)

Furthermore, as triggers can be transformed into other triggers (Louis and Sutton, 1991), Sun (2012) posits that both novel situations and deliberate initiatives can indirectly influence adaptive system use by engendering discrepancies. This is done through feedback mechanism (Beaudry and Pinsonneault, 2005; Jasperson et al., 2005) which affects adaptive system use´s cyclical process through early and recurring interaction with novel situations or deliberate initiatives consequently causing discrepancies. Therefore, the following hypotheses are drawn (Sun, 2012):

H4: Novel situations are positively associated with discrepancies.

H5: Deliberate initiatives are positively associated with discrepancies.

Transformational IT Leadership

Nowadays, where IT play a crucial role in all kind of businesses and moreover are being seen as essential for most organizations in order to compete (Carlo, Lyytinen and Rose, 2012; Houghton, Dawley and DiLiello, 2012), the question most prominent is how organizational leaders are making use of IT in order to drive their employees towards change. Having this in mind, the concept of Transformational IT leadership arises, since according to Yurov and Potter (2006) IT leaders displaying transformational leadership qualities can have impact on their followers´ system support and enhancement. However, when referring to this construct, it is essential to realize that this term is a rather new stream in literature and hence has not been fully researched yet. For that reason, in order to understand the importance of transformational IT leadership one first has to take a step back and focus on transformational leadership theories (Bass, 1985, 1999; Podsakoff, 1990) on which it is based and later link it to IT. Furthermore, since the focus of this research is pointed at organizations, the followers of (transformational IT) leaders are here referred to as employees.

Transformational leadership theories have their origin in the field of transactional leadership, thus going further and putting more emphasis on the human interaction between leaders and employees (Podsakoff, 1990). Contrary to that, the relationship between transactional leaders and employees is considered to be purely grounded in rewarding and intervening in case organizational standards are not met and can therefore be seen as extrinsically motivated (Bass, 1999).

(13)

in itself is an exchange process based on transactions.

In addition to that, Jung (2001) points out that although transactional leaders provide their employees with incentives, the transactional process itself is based on minimal performance standards. Consequently, the intention of such rewards is to maintain performance levels and stability instead of altering the status quo (Hamstra, Van Yperen, Wisse and Sassenberg, 2014). Another characteristic of transactional leaders is that they pay attention to behavioural strategies such as following clear patterns (Hamstra et al., 2014). This style of action can be understood as being task- and success-oriented (Rafferty and Griffin, 2004), hence the focus on rules and the enactment of these is one of the major concerns of transactional leaders (House, 1971). What is more, Afshari, Bakar and Luan (2009) state that transactional leaders are less effective when it comes to change, since they disregard anything that is close in relationship to any kind of interpersonal or organizational concern.

In contrast to transactional leadership, Bass (1985, 1999) says that transformational leadership is more than just rewarding employees. According to him, transformational leaders should inspire and animate their employees in order to establish an environment that promotes self leadership characteristics. This perception is shared by Den Hartog, Van Muijen and Koopman (1997) who state that transformational leaders are not only motivating their employees to perform their tasks, but also inspire them to do more than expected. Doing so, they are able to nurture their employees in terms of reaching their full potential (Dvir, Eden, Avolio and Shamir, 2002) and as a consequence of this leading them to greater achievements within the team. Yammarino and Bass (1990) add that transformational leaders not only broaden their employees´ enthusiasm, but also comfort awareness and acceptance among them considering their tasks. Similarly, Burns (1978) states that transformational leaders motivate their employees to line up their self-interests for the sake of the group.

In this regard, Bass (1985) mentions that transformational leaders do this by (1) increasing their employees´ level of awareness concerning the importance and value of the anticipated outcomes as well as the way of bringing them about; (2) motivating their employees in a way which makes them prioritize their teams´ interests above their own interests; and finally (3) by enlarging their portfolio of needs and wants. Due to this, transformational leaders are able to positively influence their employees in order to establish a common purpose while at the same time placing emphasis on an organization-wide ethical climate (Veríssimo and Lacerda, 2015).

(14)

be achieved (Veríssimo and Lacerda, 2015); (3) intellectual stimulation arises when leaders support their employees in a way that allows them to take part in decision-making processes, stimulating creativity and innovation in order to come up with solutions for future problems; and (4) individualized consideration which is about the leader´s capability to notice and determine the employees´ individual needs and desires, not only on the immediate but also on the future level. This can be achieved by coaching as well as empowering employees which allows them to grow within their positions (Bass and Avolio, 2008).

Having this pool of intrinsic behaviour tools in mind, the work of Podsakoff et al. (1996) needs to be considered, too. Similar to Bass (1985, 1999), also Podsakoff et al. (1996) emphasize employee empowerment, dividing it into six subdimensions of transformational leadership: (1) articulating a vision can be mentioned as being one of the key characteristics when talking about transformational leadership, motivating employees to go beyond their limits. This is followed by (2) providing an appropriate model, meaning that the transformational leader should act as a role model. Due to this, as stated by Avolio and Gibson (1998) the employees´ performance as well as the degree of self-leadership increases. Next to that, (3) fostering acceptance at group and (4) high performance expectations are crucial, since both deal with the leader´s ability to set up goals and working towards reaching them. In addition to that, (5) individualized support is understood as the leader´s ability to provide employees with assistance on a more personal level in order to maintain a well balanced working atmosphere. Last but not least, the component of (6) intellectual stimulation has to be listed. In this case, Podsakoff et al. (1996) describe a leader`s character trait that should be used to boost self-awareness and creativity, so employees are encouraged to find new ways of thinking instead of using old patterns.

Linking the abovementioned transformational leadership characteristics to the work of Thite (2000) who states that transformational leadership skills are present in successful IT managers or relating to Wang, Li and Hsieh (2013) who say that employees should have access to IT, the importance of transformational IT leadership becomes clear. Especially, since IT is said to increase an organization´s ability “to survive in the highly competitive global marketplace of the 21th century“ (Wu, Straub and Lian, 2015, p.498). Wu et al. (2015) also add that IT governance – in its profile similar to IT leadership - needs to be taken into account because of its significant impact on IT investments. Moreover, Weill and Ross (2004) express that “effective IT governance is the single most important predictor of the value an organization generates from IT“ (pp.3-4), which is affirmed by Kearns and Sabherwal (2007) who state that effective IT governance can generate business value. Next to that, as reported by Silver, Markus and Beath (1995) business managers and hence leaders are expected to use IT more consciously and to take over control in IT implementations (Rockart, Earl and Ross, 1996).

(15)

existing patterns (cf. Hamstra et al., 2014) and therefrom affect one´s adaptive system use. As indicated by Keen (1991), “business cannot afford technology-illiterate managers any more than it can afford business illiterate IT professionals“ (p.121). Therefore, considering this statement and the characteristics exhibited above, transformational IT leadership can be defined as supporting employees and demanding the best effort from them in order to work together towards a common goal (Bass, 1985, 1999) while using IT. Accordingly, this research will focus on whether this just abovementioned relation between transformational leadership and IT is still valid when referring to it in the context of adaptive system use, since promoting employees` self leadership characteristics (Bass, 1985, 1999) can increase active thinking which is needed for active use (Jasperson et al., 2005). This leads to the following hypothesis:

H6: Transformational IT leadership is positively associated with adaptive system use.

Especially in an IT-relevant environment, transformational leaders are a decisive factor when it comes to motivating employees since they “offer challenging tasks concerning learning about new technology features...when communicating about potential technology changes“ (Yurov and Potter, 2006, p.436). Supposed that the earlier introduced triggers like novel situations, discrepancies and deliberate initiatives all bring a form of technological change about, because they break existing patterns and nourish active use initiating adaptive system use, transformational leaders are progressively becoming more important in evolving technological societies (Brown, 1994). Based on this, it is assumed that transformational IT leadership may act as a moderator between triggering factors and adaptive system use. Also, such triggers provoking uncertainty can be referred to these days´ job demands which can require not only physical but also psychological effort (Syrek, Apostel and Antoni, 2013), which is especially needed since “we are moving from controlled to accelerated change nearly beyond control“ (Crawford, 2005, p.5). Therefore, transformational IT leaders may not only reduce negative effects of job demands (Syrek et al., 2013), but also provide guidance and support for the employees to reach their full potential (Bass, 1985; Dvir et al., 2002). Altogether, this causes enough reason to believe that transformational IT leadership acts as a moderator of the relationship between each of the abovementioned triggers – novel situations, discrepancies and deliberate initiatives – and adaptive system use. This leads to the following hypotheses:

H7a: Transformational IT leadership moderates the effect of novel situations (in terms of new tasks / changes in system environment / other people´s use) on adaptive system use in a way that the effect will be stronger when transformational IT leadership is sufficient than when it is scarce.

(16)

H7c: Transformational IT leadership moderates the effect of deliberate initiatives on adaptive system use in a way that the effect will be stronger when transformational IT leadership is sufficient than when it is scarce.

Contextual Influences

Just as the study of Sun (2012), also this research takes further influence into consideration. Both facilitating conditions and personal innovativeness in IT are said to influence the effect of the above presented triggers. However, in addition to the original study of Sun (2012) also their influence on transformational IT leadership will be analysed.

Facilitating conditions. Starting with facilitating conditions which influence the external context, Venkatesh, Morris, Davis and Davis (2003) define them as a person´s understanding of how far existing organizational and technical infrastructure can support system use. Furthermore, they can be understood as “individual perceptions of the availability of technological and/or organizational resources that can remove barriers to using a system“ (Venkatesh, Brown, Maruping, Bala, 2008, p.485). In other words, they can be treated as environmental factors which observers believe can have a supporting value to accomplish an act.

Novel situations can provoke perceived threat which according to Bala and Venkatesh (2015) is a user´s perception of new IT and its possible harm on one´s well-being. Hence, employees might perceive new IT as threatening in terms of their performance and status within an organization (Beaudry and Pinsonneault, 2005). Assistance and encouragement during novel situations however can result in exploration and refinement (Baer and Oldham, 2006) of adaptive system use, which leads to the following hypothesis (Sun, 2012):

H8a: Facilitating conditions moderate the effect of novel situations (in terms of new tasks / changes in system environment / other people´s use) on adaptive system use in a way that the effect will be stronger when facilitating conditions are sufficient than when they are scarce.

(17)

sufficient and useful support in the form of personal assistance which on the other hand might be important for the practice of adaptive system use. This leads to the following hypothesis (Sun, 2012):

H8b: Facilitating conditions moderate the effect of discrepancies on adaptive system use in a way that the effect will be stronger when facilitating conditions are sufficient than when they are scarce.

As indicated by its name deliberate initiatives imply that people are mandated to become familiar with certain features and how to use them. Such an external trigger can influence one´s intention to use, which is defined as one´s expectation of how to use enterprise systems for distinct tasks (Saeed, Abdinnour, Lengnick-Hall and Lengnick-Hall, 2010). Resulting from that, people who are instructed to use features might lose their intention to use these features which on the other hand can influence their actual use (Saeed et al., 2010) and thus adaptive system use. Facilitating conditions which earlier were defined as supporting, environmental factors however can help to adapt to the new prescribed task and/or features by representing a broad range of knowledge (Sun, 2012). Moreover, Sun (2012) posits that the presence of facilitating conditions can reduce one´s pressure to adapt immediately since additional assistance and time is provided in case there are any incomprehensions. Therefore, it is hypothesized (Sun, 2012):

H8c: Facilitating conditions moderate the effect of deliberative initiatives on adaptive system use in a way that the effect will be stronger when facilitating conditions are sufficient than when they are scarce.

(18)

H8d: Facilitating conditions moderate the effect of transformational IT leadership on adaptive system use in a way that the effect will be stronger when facilitating conditions are sufficient than when they are scarce.

Personal innovativeness in IT. Personal innovativeness in IT however can be regarded as an internal contextual factor which Sun (2012) explains by saying that adaptive system use in itself is an innovative construct. Using the definition provided by Agarwal and Prassad (1998), personal innovativeness in IT is a person´s willingness to try out new IT. This stands in relation to Amabile et al. (1996) who state that innovation can be treated as the emergence of novelty by individuals or groups. So personal innovativeness in IT refers to any kind of innovation which includes exploring and trying to find new approaches of using existing IT (Ahuja and Thatcher, 2005). In other words, it is the concept of identifying and simplifying already existing patterns by the usage of IT.

Adaptive system use is described to be innovative by nature (Sun, 2012) and therefore it resembles the basic idea of personal innovativeness in IT. Therefore, people who exhibit a higher level of personal innovativeness in IT are more likely to recognize emerging opportunities triggered by novel situations and discrepancies. Such new events can be seen as innovative (Amabile et al., 1996) and hence people with high personal innovativeness in IT will engage more intensely in adaptive system use, indicating a stronger relationship between each of these two triggers and adaptive system use. Furthermore, innovative people tend to be more risky and willing to take control in unstructured situations (Kirton, 1976) which can be beneficial considering the challenging nature of adaptive system use due to the steady motion of features (Sun, 2012). Accordingly, personal innovativeness in IT is hypothesized to moderate the impact of novel situations and discrepancies on adaptive system use positively (Sun, 2012):

H9a: Personal innovativeness in IT moderates the effect of novel situations (in terms of new tasks / changes in system environment / other people´s use) on adaptive system use in a way that the effect will be stronger for individuals with high personal innovativeness in IT.

H9b: Personal innovativeness in IT moderates the effect of discrepancies on adaptive system use in a way that the effect will be stronger for individuals with high personal innovativeness in IT.

(19)

supposedly to be affected by their leaders in a sense that they will follow his/her directions (Zhou, 2003). So deliberate initiatives which are naturally followed by guidance are said to generate strong resistance among innovative people causing less intrinsic motivation (Deci and Ryan, 1987). That is, such controlling ambience can be understood as an assault on autonomy being part of one´s job complexity which is mostly present among creative people (Feist and Gorman, 1998). This leads to the following hypothesis (Sun, 2012):

H9c: Personal innovativeness in IT moderates the effect of deliberative initiatives on adaptive system use in a way that the effect will be weaker for individuals with high personal innovativeness in IT.

As described above, transformational IT leadership is believed to positively influence adaptive system use, since it is about enhancing people to reach their full potential and thus to animate them to look beyond one´s own nose in order to reach their team´s goals (Bass, 1985, 1999; Burns, 1978). On the other hand, personal innovativeness in IT (Agarwal and Prasad, 1999) reflects one´s willingness to experiment with new IT. As a result, an employee carrying this trait (Ahuja and Thatcher, 2005) plus being under the supervision of a transformational IT leader might not feel mandated towards a specific system us. Quite the contrary, the given employee might perceive his/her leader´s presence and support as beneficial as both parties might share similar views, which again fosters leader identification (Felfe and Schyns, 2010). Consequently, the following hypothesis is drawn:

(20)

Methodology

In this section the general preparation of the data as well as the data collection process are described. This is done by introducing the corresponding methodological components in the following subsections: data collection and sample description and measures.

Data Collection and Sample Description

To investigate the topic of this study, a survey instrument focusing on team members was developed. The pilot version was then presented to three subjects to see whether the provided items were clear, but also to check how long answering the survey would take. This resulted in a response time of around ten minutes and slight changes in expression to provide linguistic clarity. Afterwards, the pilot was shown to a senior researcher in this field who helped the researcher to further adjust the wording regarding certain items.

Furthermore, the survey instrument was available both in English and in Dutch. The Dutch translation of the survey items concerning the transformational IT leadership theme was adopted from two former students (Biernath, 2014; Sietsma, 2014) of the University of Groningen. Side by side, all of the remaining items were translated by one of the researcher´s acquaintances and finally evaluated by the senior researcher in the field. The reason why next to the English version also a Dutch version was taken into account can be justified in the fact that this study is part of the graduation process of a Master´s programme at a Dutch university. What can be derived from this is that this study constitutes an international character and hence does not only focus on Dutch organizations, but also on organizations outside the Netherlands.

With the help of a colleague, the actual survey instrument was distributed to organizations that use IT on a daily basis. In respect to this, no clear line between industries was drawn. However, it was made explicit that using any form of IT on a regularly basis was a necessary criterion. Following this notion, a variety of organizations was approached.

At first, a general information-mail introducing the researcher together with the aim of this study was sent to a total of 5298 organizations, asking for collaboration. Contact information was partly acquired using the electronic database Orbis. One week later, another e-mail including access to the survey link was sent out. However, organizations which expressed that they were not able to participate in the study before launching the second email were not further contacted. Furthermore, family members and friends who have own companies or work for smaller companies were asked for participation.

(21)

which seemed unrealistic considering the number of items. Thus, the actual and final response rate constituted 2.3% (120 valid surveys). Out of the 120 participating subjects, 45 were female and 75 were male. Moreover, 25 people answered the survey using the Dutch version while 95 went for the English version. The participants´ average age was 30 years, while their tenure constituted around three years. The ordinary team size amounted to eight people per team.

Table 4

Demographic characteristics of the Sample

Variable Sample composition Team members (n=120) in %

Gender Female 45 37.5%

Male 75 62.5%

Language Dutch 25 20.8%

English 95 79.2%

Age Mean = 30.18; Std. dev. = 6.657; range 21-54 years Tenure Mean = 2.88; Std. dev. = 2.030; range 0-10 years

Team size Mean = 8.17; Std. dev. = 8.590; range 0-55 people per team

Moreover, the majority of the sample was either in possession of a Bachelor´s degree (38%) or a Master´s degree (35%), as shown in Figure 2. Also, most of the respondents (67%) indicated working in a different industry sector than what was provided by the questionnaire. Consequently, 33% work in one of the specified sectors, for instance in financial and/or insurance services (17%) or production (9%) which can be reviewed in Figure 3.

(22)

Figure 3. Industry sector

Measures

In order to achieve a set of reliable outcomes, the given measures are taken and adapted respectively from already affirmed scales from reviewed literature. Furthermore, apart from the demographic-related questions in the beginning and the end of the survey (a total of eight questions) and one open question amid the survey asking for a specific IT situation, all the remaining items of the survey were approached using a 7-point Likert-scale. This was consciously decided by the researcher, since a broader set of items is said to increase validity and reliability of the collected data (Lozano, Garcia-Cueto & Muniz, 2008).

Dependent variable. The main dependent variable in this research is adaptive system use. As described above in the theoretical framework section, adaptive system use can only emerge through one of the prior described triggers or a combination of them (Sun, 2012). In this study, adaptive system use will be measured with the help of Sun´s (2012) particularly for this variable developed constructs. In close relation to this, it has to be noticed that the researcher adapted those constructs and the corresponding items in order to fit an overall IT-related context. An adjusted example item is: “I combined features in our IT tools with features in other applications to finish a task“. Parallel to adaptive system use, also discrepancies as one of the triggers take on the role of being a dependent variable.

(23)

recall one specific incident to which they should refer to. This situating task as well as the items were adapted from Sun´s (2012) research. Compared to his research, in this study the used items were translated into a broader context, referring to IT in general instead of a specific system. Besides, also the concept of transformational IT leadership was added as an additional trigger by the researcher, to see whether it can serve as a further triggering factor to adaptive system use.

Moderating variables. Concentrating on the moderating variables, concepts such as facilitating conditions and personal innovativeness in IT come into play (Sun, 2012). Concerning both concepts in this study the researcher made use of a set of already approved items to ask about possible facilitating conditions and personal innovativeness in IT. Focusing on the facilitating conditions, these were adapted from Venkatesh et al. (2003). The items related to personal innovativeness in IT however were taken from Agarwal and Karahanna (2000). In both cases, the researcher went for slight adjustments to fit a more general IT context. Beyond that, transformational IT leadership was added by the researcher as an additional moderating variable using items based on the work of Podsakoff et al. (1996). For this purpose, the researcher reverted to the work of two former students (Biernath, 2014; Sietsma, 2014) who adjusted Podsakoff et al.´s (1996) original questionnaire regarding transformational leadership. In this sense, they specified the essential dimensions of transformational leadership in an IT context. Examples of this process are “My team leader fosters collaboration between teams by using IT tools“ instead of “My team leader fosters collaboration among work groups“ and “My team leader has a clear understanding of how IT tools should be used to get where we want to be“ in place of “My team leader has a clear understanding of where we are going“.

(24)

Analysis and Results

T

his section elaborates on the analysis as well as the results of this research. At first, the underlying analysis is explained to provide a generic overview of the conducted steps. Afterwards, the results of the research are introduced.

Analysis

In the beginning of the analysis, I used IBM SPSS Statistics 23 to conduct an explanatory factor analysis with varimax rotation. This was intended to identify information about the relationships between the constructs´ different items and their scores (Hinton, 2004). Moreover, items were only to include if they fulfilled the criterion of having a loading of at least 0.5 (Song, Bij and Song, 2011). Based on this condition, an EFA was then realized for the constructs belonging to Transformational IT Leadership and Adaptive System Use, but also for the independent and moderating variables. By virtue of the results of the EFA analyses, items not loading to constructs or loading to more than one construct were dropped. For instance, the EFA for the transformational IT leadership construct shows that all of the initial six dimensions remain while some of the items had to be deleted. The results of this last-mentioned EFA are presented in Table 5.

Table 5

Explanatory factor analysis loadings and Cronbach´s alpha (Transformational IT Leadership)

Construct Item Factor Loadings Cronbach´s alpha

Articulating an IT-vision TLAV 1 .802 α = .803

TLAV 3 .744

Providing an appropriate IT-role model TLAM 2 .847 α = .919

TLAM 3 .785

Fostering the acceptance of group goals through IT TLFG 1 .863 α = .815

TLFG 2 .779

High performance expectations with IT TLPE 2 .944 n/a

Individualized support TLIN 2 .918 α = .860

TLIN 3 .870

Intellectual stimulation with IT TLIS 2 .991 n/a

(25)

Table 6

Confirmatory factor analysis loadings, t-values and Cronbach´s alpha

Construct Item Factor Loadings t-value Cronbach’s alpha Adaptive System Use

Trying new features TR 1 .81 10.21 α = .835

TR 2 .71 8.44 TR 3 .89 11.61 TR 4 .62 7.12 Feature substituting FS 1 .89 11.90 α = .875 FS 2 .81 10.27 FS 3 .82 10.61 Feature combining FC 2 .94 13.15 α = .888 FC 3 .84 10.99 FC 4 .78 9.85 Feature repurposing FR 2 .92 12.98 α = .917 FR 3 .88 11.99 FR 4 .90 12.38 FR 6 .75 9.50

χ2 = 129.97; df. = 71 ; RMSEA = .084; NFI = .94; CFI = .97; GFI = .87

Novel situations

Changes in System Environments SE 1 .63 5.39 α = .846

SE 2 .98 7.24

Other People’s Use OU 2 .96 9.67 α = .864

OU 3 .71 7.01

Discrepancies DP 1 .89 10.76 α = .905

DP 2 .93 11.48

Articulating an IT-vision TLAV 1 .85 10.86 α = .803

TLAV 3 .80 10.01

Providing an appropriate IT-role model TLAM 2 .89 12.29 α = .919

TLAM 3 .95 13.62

Fostering the acceptance of group goals through IT TLFG 1 .86 10.74 α = .815

TLFG 2 .80 9.81

Individualized support TLIN 2 .80 9.10 α = .860

TLIN 3 .95 11.03

Personal Innovativeness in IT PIIT 1 .84 10.55 α = .859

PIIT 3 .82 10.27

PIIT 4 .82 10.15

Facilitating Conditions FCOND 1 .67 5.81 α = .743

FCOND 3 .43 6.61

χ2 = 128.77; df. = 116; RMSEA = .030; NFI = .93; CFI = .99; GFI = .90

RMSEA - Root Mean Square Error of Approximation; NFI - Normed Fit Index; CFI - Comparative Fit Index; GFI - Goodness of Fit Index

The Cronbach´s alphas for each of the constructs were above .7 (ranging from .743 to .919), which according to Hinton (2004) indicates a reliable scale. Additionally, several goodness of fit statistics as proposed by Raykov and Maroulides (2006) are considered, since goodness of fit should not be established on a single index. In this sense, the goodness of fit statistics reveal that the models are appropriate.

(26)

Index (GFI) = .87. For transformational IT leadership including independent and moderating variables the scores are as follows: χ2 = 128.77, df. = 116, RMSEA = .030, NFI = .93, CFI = .99, and GFI = .90.

Secondly, descriptive statistics and correlations were determined for the constructs in the conceptual model. The results are presented in Appendix B.

Third, a path analysis using Lisrel 8.80 was conducted to analyse the relationships between each of the constructs. More specifically, possible moderation effects between the independent and dependent variables were tested. Thus, it has to be specified that the four dimensions of adaptive system use were separately taken into account. In terms of transformational IT leadership serving as a trigger that is an independent variable, its six underlying dimensions and their influences on the four dimensions of adaptive system use were tested distinctly, too. Contrariwise, when relating to transformational IT leadership as a moderating variable its six dimensions were not tested independently, they were averaged to simplify the structural model to be analysed and mean centred (Kenny and Judd, 1984; Aiken, West and Reno, 1991).

All things considered, as proposed by the program some modification indices were included to the model in order to obtain a better, final result (see Table 7). The goodness of fit statistics for the final model of the path analysis are: χ2 = 59.29, df. = 46, RMSEA = .049, NFI = .99, CFI = .99, and GFI = .98.

Results

The results of the path model (Table 7) exhibit to some extent support for the anticipated research hypotheses. Starting with H1, all three subcategories of novel situations do show ties to adaptive system use. However, the relationships are not merely of a positive nature. New task negatively relates to feature substituting (b = -.07; p < .05), while it supports feature combining (b = .34; p < .001) and feature repurposing (b = .30; p < .01). Though, it does not offer any relation to trying new features. Next, changes in system environments are positively related to feature combining (b = .31; p < .05) and to feature repurposing (b = .35; p < .01), while they do not show any relation to trying new features nor feature substituting. Other people´s use only indicates a positive relation to feature combining (b = .14; p < .05). H2 is not supported, not showing any positive relation between discrepancies and adaptive system use at all. The only present relationship is negative and exists between discrepancies and feature substituting (b = -.01; p < .05). H3 is partially supported, offering a positive relationship between deliberate initiatives and feature repurposing (b = .11; p < .05). Next, also H4 is only partially supported. In terms of novel situations its subcategory other people´s use does indicate a positive relationship towards discrepancies (b = 1.63; p < .001). Focusing on H5, there is no significant relation at all. Meaning that H5 is not supported.

(27)

Table 7

Non-standardized path coefficients for the conceptual model 1/2

Independent variables: Adaptive system use Discrepancies Team members

(n=120) Trying new

features

Feature substituting Feature combining Feature repurposing Novel

Situations

New Task - .08 - .07* .34*** .30** - .13 Changes in System Environments .16 - .08 .31* .35** - .03

Other people´s use .00 .04 .14* - .06 1.63***

Discrepancies .03 - .01* - .02 - .01 Deliberate Initiatives - .01 - .04 .06 .11* - .03 Trans-formational IT Leadership Articulating an IT-vision .13 .05* - .06 - .16*

Providing an appropriate IT-role model .13 - .06 - .07 .10 Fostering the acceptance of group goals through IT - .21* .03 .13* .20**

High performance expectations with IT .55 - .04 .12 .47

Individualized support .07 - .04 - .43 .20

Intellectual stimulation with IT - .03 .00 .55 - .34 Novel

Situations* TITL

New Task*TITL .10 - .02 .00 .00

Changes in System Environments*TITL .11 - .07* .10 - .11

Other people´s use*TITL - .15 - .02 - .16 .07

Discrepancies*TITL .17 - .07 .12 - .14 Deliberate Initiatives*TITL - .07 .07* - .14 .04 Novel Situations* Fcond New Task*Fcond .17 .00 - .07 .16

Changes in System Environments*Fcond .16 .04 - .04 .05

Other people´s use*Fcond - .01 - .01 - .03 .02

Discrepancies*Fcond .01 .02 .05 - .02

Deliberate Initiatives*Fcond .16* - .08* - .14* .01

(28)

Non-standardized path coefficients for the conceptual model 2/2 Trans-formational IT Leadership *Fcond Articulating an IT-vision*Fcond .08 .04* .10* .15**

Providing an appropriate IT-role model*Fcond .20 .02 .04 .11 Fostering the acceptance of group goals through

IT*Fcond

.08 .09* .06 - .07

High performance expectations with IT*Fcond .03 .00 - .09 - .03

Individualized support*Fcond - .04 - .04 .05 - .03

Intellectual stimulation with IT*Fcond - .15 - .03 - .06 - .13 Novel

Situations* PIIT

New Task*PIIT - .02 .03 - .01 - .26

Changes in System Environments*PIIT - .13 .02 - .05 - .18

Other people´s use*PIIT - .09 .06 .02 - .13

Discrepancies*PIIT - .55* .06 .11 .05 Deliberate Initiatives*PIIT .08 .00 - .01 .04 Trans-formational IT Leadership *PIIT Articulating an IT-vision*PIIT - .12 - .02 .17* .01 Providing an appropriate IT-role model*PIIT - .04 .01 - .11 .01 Fostering the acceptance of group goals through

IT*PIIT

- .13 .04 .10 .15*

High performance expectations with IT*PIIT .08 - .06 .32* .28*

Individualized support*PIIT .05 - .03 - .35* - .20*

Intellectual stimulation with IT*PIIT .04 .08 .12 - .07

(29)

First, there is a positive relationship between articulating an IT-vision and feature substituting (b = .05; p < .05) and a negative one between articulating an IT-vision and feature repurposing (b = -.16; p < .05). Yet, there are no relationships between articulating an IT-vision and trying new features or feature combining. Second, there are relations between fostering the acceptance of group goals through IT and trying new features (b = -.21; p < .05), between fostering the acceptance of group goals through IT and feature combining (b = .13; p < .05) and between fostering the acceptance of group goals through IT and feature repurposing (b = .20; p < .01). As before, also in this case one of the dimensions of adaptive system use is not affected, namely feature substituting.

Additionally, H7a, H7b and H7c hypothesize positive relationships between novel situations and adaptive system use, moderated by transformational IT leadership. The results show that H7a and H7b are not supported. In terms of H7a there is even a negative moderating effect of transformational IT leadership in case of the relationship between novel task´s subcategory changes in system environments and the feature substituting dimension of adaptive system use (b = -.07; p < .05). Focusing on H7c, there is partial support. Its result indicates that transformational IT leadership only moderates the relationship between deliberate initiatives and adaptive system use in terms of feature substituting (b = .07; p < .05).

(30)

dimensions of adaptive system use being moderated by facilitating conditions.

Further, I hypothesized that the relationships between the triggers and adaptive system use are moderated by personal innovativeness in IT. This is put down in writing by H9a, H9b, H9c and H9d. In this regard, H9a, H9b and H9c are not supported, whereas H9d is partly supported. Concerning H9b, opposed to what was assumed there was a negative moderating effect of personal innovativeness in IT in case of the relationship between discrepancies and trying new features (b = -.55; p < .05). That followed, there are moderating effects of personal innovativeness in case of the relationships between four out of the six transformational IT leadership dimensions and adaptive system use. Foremost, there is a positive moderating effect of personal innovativeness in IT in case of the relationship between articulating an IT-vision and feature combining (b = .17; p < .05). Consequently, there are no moderating effects in case of the relationship between personal innovativeness in IT and the remaining three dimensions of adaptive system use. Secondly, a positive moderating effect of personal innovativeness in IT in case of the relationship between fostering the acceptance of group goals through IT and feature repurposing (b = .15; p < .05) is revealed. Subsequently, there are no relationships between fostering the acceptance of group goals through IT and the other three dimensions of adaptive system use moderated by personal innovativeness in IT. Thirdly, there is evidence for positive moderating effects of personal innovativeness in IT in case of the relationships between high performance expectations with IT and feature combining (b = .32; p < .05) and between high performance expectations with IT and feature repurposing (b = .28; p < .05). Thus, this logically means that there are no significant moderating effects of personal innovativeness in IT in case of the relationships between high performance expectations with IT and the remaining two dimensions of adaptive system use. Focusing on the last of the four dimensions of transformational IT leadership, the following can be said. There are negative moderating effects of personal innovativeness in IT in case of the relationships between individualized support and feature combining (b = -.35; p < .05) as well as between individualized support and feature repurposing (b = -.20; p < .05). Additionally, there are no relationships between individualized support and the remaining two dimensions of adaptive system use, moderated by personal innovativeness in IT.

One additional, not hypothesized, relationship has been proposed by Lisrel 8.80. This suggestion occurs in the form of a relationship between facilitating conditions acting as an independent variable and one of the dimensions of adaptive system use. Resulting from this, the relationship between facilitating conditions and feature combining was found to be positive (b = .32; p < .05).

(31)

Discussion and Conclusion

In this section, both general discussion and conclusion of this research are presented. In this context, theoretical but also practical implications are taken on to establish a broader understanding of the given topic and ergo how to apply that obtained knowledge. Further, limitations and directions for future research are discussed.

As said before, the main goal of this research was to expand Sun´s (2012) concept of adaptive system use by analysing user revisions of IT-based features and their interplay with transformational IT leadership. The reason for that is justified in the fact that IT adoption as well as IT use have evolved into an important part of IS literature (Venkatesh et al., 2003), with post-adoption stages being the most time-consuming (Jasperson et al., 2005). Furthermore, according to Thite (2000) transformational leaders are suitable to guide technological changes. Secondly, this study aimed to explore whether transformational IT leadership exerts influence when acting as a moderating variable between the triggers and the concept of adaptive system use. Hence, the raised research question was: “Does transformational IT leadership influence adaptive system use and if so in which way?”. The different interactions were studied through Sun´s (2012) concept of adaptive system use and an adjusted version of Podsakoff et al.´s (1996) original questionnaire regarding transformational leadership.

(32)

Louis and Sutton (1991). The reasoning behind this statement is that the triggers used by Sun (2012) do not seem to cause active thinking in a way that employees´ active use behaviour (Jasperson et al., 2005), in this example represented by the different types of revisions, is affected adequately. This indicates that employees partially treat the effects those triggers bring about as expected IT events, as they deal with the given IT-related circumstances in a habitual manner not being required to think actively (de Guinea and Webster, 2013). Therefore, there might be other triggers causing adaptive system use, too.

The results for transformational IT leadership acting as a trigger indicate that only two out the six transformational IT leadership dimensions affect the dimensions of adaptive system use somehow. Additionally, not all of these relations were positive. This means that transformational IT leadership does not act as the significant trigger it was assumed beforehand, as it does not provoke active thinking sufficiently and ergo active use. So it seems that transformational IT leaders are not as effective as expected when it comes to change (Afshari et al., 2009). This is in contrast to Hamstra et al. (2014) who state that transformational leaders should be able to motivate their employees and at the same time increase their active thinking in a way which enables them to break existing patterns (Sun, 2012). In this line, one could argue whether the presence of a transactional leader would be more useful, as setting up performance criteria (House et al., 1988) while being success-oriented (Rafferty and Griffin, 2004) might push the employees towards feature revisions as they know what they receive in return for their duties. Overall, it can be said that people perform various adaptive system use behaviours contingent upon different triggering conditions. In this spirit, there are cases where certain triggers generate more than one of the revision behaviours introduced by Sun (2012). For instance, while deliberate initiatives only relate to one of adaptive system use´s dimensions, novel situations represented by changes in system environments impact adaptive system use triggering two of its dimensions (cf. Table 7).

On the other hand, the findings partially confirm that transformational IT leadership moderates the impact of the triggers on adaptive system use. However, only the relationship between deliberate initiatives and adaptive system use is moderated by transformational IT leadership; and even here this moderated relation impacts only one of the dimensions of adaptive system use. Nevertheless, though transformational IT leaders do not seem to fully affect their employees´ feature revisions while these are being triggered by externalities, it also means that in some cases the help transformational IT leaders give away to their employees by offering challenging tasks (Yurov and Potter, 2006) somehow influences adaptive system use. Even if it only holds true for the feature substituting dimension, this implies that transformational IT leaders may play a decisive role on how employees adopt certain behaviours.