Implementation Adoption: The NCISA Model
‘How does national culture influence IS Implementation adoption?’
A case survey approach.
By
Wouter ten Heggeler
University of Groningen Faculty of Economics and Business
MSc Business Administration: Change Management January 2018
Supervisor: dr. J. F. J. Vos 2nd Assessor: prof. dr. A. Boonstra
Word count: 9755
Abstract
The importance of national culture effects on information system (IS) implementations has been noted by several authors. While numerous models exist to explain the resistance to- or the acceptance of IS, the influence of national culture has been neglected. This study focusses on the research question “how does national culture influence IS implementation”. In order to answer this question a case survey strategy was followed, which is a method that allows the researcher to analyze a large sample of cases that contains rich qualitative data. The sample contained 85 cases covering 23 countries and 23 industries. A research framework was developed based upon national culture, user resistance, and user acceptance literature. Hofstede’s national culture dimensions (1980, 1988) were used to identify national culture indications within the data. These indications were connected to resistance and acceptance antecedents (i.e. user adoption antecedents) within the cases who in turn influenced IS adoption. 253 instances were identified within the sample which were summarized in the National Culture Information System Adoption model. Analysis of the data resulted in the conclusion that some national cultures will be more accommodating towards IS implementations. However, the results give implementers the knowledge needed to avoid national culture issues even in countries that are less ideal. This study provides ample future research options and adds to existing national culture, resistance, acceptance and IS adoption literature.
Introduction
Several authors have noted the importance of national culture effects on the adoption of integrated information systems (IS) (Shanks et al., 2000; Straub, 1994; Straub, Keil, & Brenner, 1997). Since the rise of the multinational company, national culture has been gradually introduced in business models as an important factor to consider (Kogut & Singh, 1988; Newman & Nollen, 1996; Schneider & De Meyer, 1991). As such, Straub hypothesized in 1994 that the national cultural dimensions identified by Hofstede (1980) could have a significant influence on the adoption and use of IT. He focused on the adoption of new technologies at that time (i.e. FAX and E-Mail) in two inherently different cultures namely, the U.S.A. and Japan. He found empirical evidence that showed that the adoption of E-Mail was hindered by the national culture of Japan. Since then research on the relationship between national culture and integrated IS adoption has been limited. Several authors have researched this phenomenon (Leidner & Kayworth, 2006; Muk & Chung, 2015; Straub et al., 1997). Yet, an integrated IS adoption model which incorporates national culture effects is still non-existent. The main research question in this study is therefore “how does national culture influence IS adoption?”.
Within IS adoption literature two important factors emerge, user resistance and user acceptance (Van Offenbeek, Boonstra, & Seo, 2013). User resistance towards technology, has been approached from numerous perspectives (J. Ford, Ford, & D’Amelio, 2008; Palmer, 2004; Rivard & Lapointe, 2012). However, none of these perspectives include the national culture of a country. Technology acceptance research is seen as one of the most mature research subjects in information system literature (e.g. Hu, Chau, Liu Sheng, & Tam, 1999). IS acceptance models state various potential causes for a failure of IS adoption yet the national cultures effect have not been extensively researched and incorporated into a model (Delone & McLean, 2003; Venkatesh et al., 2003). Therefore, this study will develop a research framework that includes national culture and user adoption antecedents. Due to the complex nature of culture, rich data will be needed to identify its influence on IS adoption. Rivard & Lapointe (2012), used a case survey approach to study to satisfy the need for rich data in their effort to different forms of user resistance. Consequently, this study facing the same requirements, a case survey approach will be used to gather data.
Furthermore, it will identify cultures that are inherently supportive or resistant of new information systems. This can help practitioners be prepared when implementing such a system. They can adapt their approaches to the implementation based on the national culture to maximize adoption of the information system.
This paper is structured as follows. In the next section, we discuss the prevalent literature on user resistance and user acceptance hereby following the line of reasoning by Van Offenbeek et al. (2013) that these two concepts are critical factors in IS adoption. Furthermore, the literature on national culture is studied as well as the ongoing debate between two predominant theories in this field leading to the research framework for this study. Using a case survey approach the framework is evaluated and in an effort to answer the research question, pattern matching is used to identify causal chains within the research framework (Larsson et al., 1993; Yin, 1994). The paper concludes by showing the patterns found within the research model as well as a discussion on the results. The discussion also goes into the impact this research has on IS adoption literature together with national culture literature.
Literature Review
To build a research framework on IS adoption the literature on user resistance, user acceptance as well as national culture needs to be analyzed to discover the most appropriate operationalization and measures when researching culture effects on integrated IS. Firstly, user adoption and in its extent user resistance and acceptance are analyzed. Next, the basis of national culture theories will be reviewed followed by a discussion on which national culture operationalization is most appropriate for this study.
User Adoption
resisting users (i.e. users that accept therefore use the technology but still openly resist it). They found that user adoption antecedents are based in both resistance and acceptance literature. Therefore, in the next section, both resistance- and acceptance literature is reviewed to find the appropriate antecedents to user adoption.
Resistance to- and Acceptance of Technology
User resistance. User resistance is defined as a set of user behaviors that show
some discontent with the implementation of a new IS (Rivard & Lapointe, 2012). This phenomenon has been analyzed from multiple perspectives. These behaviors can be seen as something positive (J. Ford et al., 2008), or something that has to be overcome (Palmer, 2004), but it is believed to be something that has to be attended in order to increase the quality of the implementation process (Rivard & Lapointe, 2012).
According to Markus (1983), an individual’s resistance towards an implementation process is determined by the interaction between the technology and the context of its use. For instance, if a newly introduced technology negatively influences the balance of power that exists between the individual and his environment, the individual will resist. Moreover, Joshi (1991) states that users employ a three-stage evaluation of a change in terms of its impact on their equity status. He therefore sees the emergence of resistance as a consequence of inequality. Rivard & Lapointe (2012) synthesized the different views on resistance into five basic elements of resistance: manifestations of resistance, subjects of resistance, object of resistance, perceived threats, and initial conditions.
User acceptance. User acceptance is defined as user behaviors that show intend
towards using a new IS. The literature provides various user acceptance models (e.g. Davis et al., 1989; Venkatesh et al., 2003; Venkatesh & Davis, 2000) but the basic concept on which they are built is depicted in Figure 1. One of the core models is the
Individual reactions to
using IT Intentions to use IT Actual use of IT
Technology Acceptance Model (TAM) which was introduced in 1986 by Davis and revolved around two key variables namely, perceived usefulness (i.e. the individual’s idea of how useful the technology in question is going to be) and perceived ease of use (i.e. the individual’s idea of how easy the operation of the technology is going to be) which would influence the attitudes and behavioral intention towards actual system use (Davis et al., 1989). These two variables are still seen as relevant in the models that followed. The most popular and extensive model that is available today was built on TAM namely, the Unified Theory of Acceptance and Use of Technology (UTAUT) (Venkatesh et al., 2003).
Firstly, Venkatesh et al. (2003) developed four constructs of which three influence behavioral intention (i.e. reactions to using IT), which in turn influence user behavior, and one construct that directly influences user behavior. The significant positive relationship between behavioral intention and use behavior has been the basis of all underlying intention models used in the UTAUT (e.g. Sheppard, Hartwick, & Warshaw, 1988 for an extended overview of the relationship). Moreover, after analyzing various technology acceptance theories, four constructs that play a significant role in user acceptance and usage behavior were identified namely: performance expectancy, effort expectancy, social influence, and facilitation conditions. The four primary constructs are well-defined and rooted in various theories making the suitable for supporting a coding scheme. Moreover, Van Offenbeek et al. (2013) posit that facilitating conditions and social influence constructs of Venkastesh et al. (2003) can actually also lead to resistance instead of acceptance.
National Culture
ideology, rituals, myths, and ceremony (Pettigrew, 1979).
Schein (1985a, 1985b) introduced a model that encompasses both the more observable (e.g. artifacts) and the less observable aspects of culture. Schein’s model consists of three levels namely, basic assumptions, espoused beliefs and values, and artifacts. When researchers started studying culture most followed Schein’s (1985b) advice to focus on the second level, espoused beliefs and values. These values are adopted beliefs isolating what is important to a particular cultural group. They are the answer to why people behave the way they do (Schein, 1985b). They are the ideals, goals and aspirations of individuals and therefore inherently more visible (Schein, 2010). Schein (1985b) argues that they are more easily studied because of their visibility and problems that the other two levels pose. Basic assumptions are mostly invisible and preconscious whereas artifacts might be very visible but on the other difficult to decipher. It is therefore logical that this study will use a value-based approach as well.
The national culture debate. Most popular taxonomies of national culture
as well as the danger of potential overlap between these dimensions when analyzing data, the Hofstede dimensions are chosen as this study’s operalization of culture.
Hofstede’s dimensions. As mentioned before, Hofstede identified five natural
culture dimensions: power distance, uncertainty avoidance, individualism-collectivism, masculinity-femininity, and short-term versus long-term orientation (Hofstede, 1980; Hofstede & Bond, 1988). The dimensions are described as follows.
Power distance. Hofstede (1980) states that powers distance is the degree to which society accepts the inequality of power distribution within institutions and organizations. Therefore, when there is a high power distance, there is a general belief that the masses are dependent on a select few (Hofstede, 1980). For instance, in the superior-subordinate relationship, with high power distance, the subordinate is expected to follow orders without questioning them whereas in a country with low power distance the opposite is true. In such a country the subordinate is encouraged to question the superior and exert greater independence (Hofstede, 1984).
Individualism – collectivism. This dimension describes the preference for a social framework where individuals take care of themselves as opposed to collectivism where individuals expect a group to take care of them in exchange for loyalty (Hofstede, 1980). Therefore, an individualist culture will prefer a competitive relationship among colleagues. Their promotions will be based upon merit and poor performance is met with punishment. On the contrary, collectivist cultures prefer a relationship with their colleagues that is based upon mutual loyalty and poor performance is met with a change in task (Murcia & Whitley, 2007). Logically, when considering employee training, individualist cultures prefer private, personal, training whereas collectivist cultures rely on group training (Murcia & Whitley, 2007).
Masculinity – femininity. Masculinity versus femininity revolves around the preference for achievement, assertiveness, and material success. Scoring high on this dimension is referred to as masculine as opposed to a low score being feminine (Hofstede, 1980). Besides this, feminine cultures are known to focus more on personal goals such as a friendly atmosphere or a comfortable work environment instead of work-related performance goals (Gallivan & Srite, 2005).
Long-term – short-term orientation. The fifth, and last, dimension deals with society’s orientation towards the future. At one end a strong focus on immediate, short-term, results as opposed to being comfortable with sacrificing now with the prospect of long-term benefits (Hofstede & Bond, 1988). From a business perspective, one can observe this is the form of decision-making. Short-term oriented cultures will focus on making decisions that favor short-term business targets whereas long-term oriented cultures will focus on the overall organizational performance on the long-run (Hofstede & Bond, 1988).
The theory reviewed will be used to build a user adoption research framework that includes national culture. The literature shows the complexity of national culture and the user adoption antecedents. This demands a research approach that will allow the identification of nuances of these dimensions and constructs. In the following section the methodology chosen to detect these nuances will be discussed.
Research Methodology
One of the most central activities in organizational research is the development of theory. Traditionally, this has always been done by combining observations from previous literature, common sense and experience (Eisenhardt, 1989). However, to tie this in with data has often been a challenge (Perrow, 1986; Pfeffer, 1982). Since this study will attempt to do both, a method is required that provides in-depth perspectives on the implementation of IS and the complex nature of culture.
what this study targets. Therefore, this study will attempt to replicate and add to this methodology.
Case surveys are a way to aggregate data from multiple qualitative case studies and turn it into semi-quantitative data using coding schemes (Lucas, 1974; Newig & Fritsch, 2009). Larsson et al. (1993) provide a procedure for performing a case survey. The procedure is broken up into four general stages. This study will follow these four stages namely, (1) selecting a group of existing case studies relevant to the research question, (2) design a coding scheme, (3) use multiple raters for coding and (4) analyze the coded data for patterns. The procedure stipulates the use of multiple raters for coding. Yet, this study will be performed by a single rater.
Selecting Relevant Case Studies
In accordance with Larsson et al. (1993) multiple search strategies were used in this case survey to limit the biases related to search strategies. The following strategies were employed, (1) computer search, (2) manual search, and (3) a reference list search. In the next section the various search strategies will be discussed as well as the exclusion- and inclusion criteria.
Search strategies. Firstly, the results were validated by recreating the searches
done by Krijnse Locker et al. (2016). This resulted in the same cases found in their study. The next part of the search strategy was done in four phases. The first phase, a computer search strategy, was reenacting the search done by Krijnse Locker et al. (2016) with a specific focus on the period of 2015-2017 to examine if any new cases studies were done since the publishing of the article. Whereas they went through several keyword combinations this paper skips to their final search strategies.
Firstly, only the Academic Search Premier and Business Source Premier databases were used to exclude non-related field results and the search engine was programmed to only show case studies. Secondly, the following keywords were used to find a satisfactory amount of results: “user” + “resistance” + “information” + “system” + “implementation”. To find only new potential papers to add to the existing data-set the search engine was programmed to only show results between 2015-2017.
“acceptance” + “information” + “system” + “implementation”. However, the amount of potential cases was minimal since most of the promising results were already included in the original data-set. This led to, phase three, a number of manual searches which included experimenting with combining search terms and leaving some out (e.g. “information system” + “implementation”) since as many relevant keywords should to be used to make sure all components of the study are covered (Boonstra, Versluis, & Vos, 2014). Lastly, phase four, consisted of a reference list search was performed on the potential papers found and other literature used in this study.
Case collection and selection. When selecting the cases used in this research
they first needed to pass certain inclusion and exclusion criteria. First, as an inclusion criterion, the case must clearly describe an IS implementation process. Examples are implementations of an enterprise systems like ERP and CRM. Second, the description of the process must be in the form of a rich narrative and this narrative must clearly offer a cultural link to user behavior. The last inclusion criterion is that the nationality/region of the organization in which the implementation takes place should be available. Only once a case is subject to all these criteria will it be included. Yet, there are also multiple exclusion criteria. First, the implementation must be in an organizational context. Secondly, to ensure the quality of this research the source must be a peer-reviewed study. Lastly, the source paper must be available to the University of Groningen and written in English. Inclusion and exclusion was mostly done during the search for potential cases. One paper was excluded during analysis since the narrative was found insufficient.
cultural observations when combining the collected quotes into one case. The case collection and selection across different phases can be seen in Table 1.
Analysis
The goal of this is study is theory development and will therefore use an approach that aids this goal namely, pattern matching (Sheu, Yen, & Krumwiede, 2003; Yin, 1994). Within in each of the 85 cases patterns between national culture dimensions and user adoption were identified through user adoption antecedents (Figure 2). Rivard & Lapointe (2012), also using a case survey method, analyzed implementer responses to user resistance by identifying episodes of resistance (i.e. initial resistance led to an implementer responds which led to ex post resistance behavior). A similar approach is used here to identify the patterns within the cases. These patterns were isolated in the form of causal chains, or instances, similar to the episodic approach of Rivard & Lapointe (2012). National culture indications, in form of quotes out of the narrative, are linked to user adoption antecedents, again in the form of quotes out of the narrative, which in turn led to a decreased or increased user adoption. A more detailed explanation of this process, as well as examples, can be found in Appendix C.
Table 1 - Case Collection & Selection
Validation Original Data-Set 70 cases
Exclusion within original data-set 1 case
Total after validation 69 cases
Phase 1: Focus on 2015-2017 0 cases
Phase 2: User Acceptance Search 4 cases
Phase 3: Manual Searches 6 cases
Phase 4: Reference List Search 7 cases
Exclusions: 1 case
The use of multiple cases across multiple countries/regions offers the possibility of identifying cross-cultural patterns hereby supporting the goal of this paper in making the results more generalizable (Larsson et al., 1993) and appropriate for generating theory (Eisenhardt, 1989).
Coding. Once the collection and selection of cases was completed a coding scheme was developed as Larsson (1993) suggests. This research will follow Rivard & Lapointe (2012) as they use a coding scheme based upon literature and then built upon that while coding. In other words, they used an iterative coding scheme.
Initial coding scheme. For this study, a research framework was developed
which is depicted in Figure 2. Whereas the basic technology acceptance model (Figure 1) measures the individual reactions to using IT, this model uses user adoption antecedents. Hereby, the model can be grounded in literature through antecedents found in relevant literature. Within the literature five national culture dimensions (Hofstede, 1980, 2001; Hofstede & Bond, 1988) were found and the codes will be based upon them to measure “National Culture”. The national culture scores of Hofstede (1980) are not used to avoid any controversy surrounding them (i.e. the scores are argued to be outdated, or mispresent cultures) (Ford, Connelly, & Meister, 2003; Gallivan & Srite, 2005; Murcia & Whitley, 2007). For example, the study was performed using 116,000 IBM employees across different countries. Since only IBM employees were interviewed
User Adoption Antecedents National Culture Power Distance Individualism Uncertainty Avoidance Masculinity Long-term Orientation Social Influence (R | A) Facilitating Conditions (R | A)
Performance Expectancy (A) Effort Expectancy (A)
Perceived Threat (R) Manifestation of Resistance (R)
User Adoption
the data will reflect idiosyncrasies of that company (Gallivan & Srite, 2005). Therefore, the national culture will only be observed through situations that indicate a connection towards a national culture dimension. To measure how these national culture indications influence IS user adoption several antecedents were found in the literature. Firstly, in acceptance literature four constructs of the UTAUT model are used (Venkatesh et al., 2003). However, two, namely facilitating conditions and social influence, can also come in the form of resistance according to Van Offenbeek, Boonstra, & Seo (2013). Furthermore, in user resistance literature, two overlying constructs were selected namely perceived threat (Bala & Venkatesh, 2016; Rivard & Lapointe, 2012) and manifestation of resistance (Rivard & Lapointe, 2012). Lastly, for every instance the relation towards user adoption is tracked by analysis of the narrative. Figure 2 shows which antecedents come from which literature by attaching (A) for acceptance literature, (R) for resistance literature, and (R | A) for both. A complete overview of the codes can be seen in Appendix D.
In sum, the coding file states the code attached to the case based on the search strategy, the country of origin, and the implementation description (e.g. a pre-implementation narrative). Next, national culture indications, in form of quotes out of the narrative, are coded based on Hofstede’s dimensions (1980, 2001). This is followed by coding the user adoption antecedent, again in the form of a quote, linked to the national culture indication based on various antecedents found in the literature. The last part of the causal chain answers the question if the national culture indication led to an increase or decrease of user adoption. This is extracted from the narrative and coded. A more detailed explanation of this process, as well as examples, can be found in Appendix C.
Results
In this section the results of the study will be described. First, the case descriptives will be explained followed by a description of the most frequently observed patterns per national culture dimension.
Case Descriptives
between countries. The examined IS were implemented in Australia (3), Central Europe (2), Canada (6), the Caribbean (1), Chile (2), China (5), Denmark (1), France (1), Germany (1), Hawaii (1), Italy (1), Lithuania (1), Malaysia (1), the Netherlands (5), New Zealand (4), Norway (3), Portugal (1), Singapore (1), Spain (1), Sweden (2), Switzerland (1), the UK (15) and the USA (26). The studied organizations operate in 23 different industries and together they implemented around 39 different enterprise systems, all of which were integrative systems. The industries ranged from public institutions like healthcare (26) and education (11), to more traditional sectors like manufacturing (12), energy (5) and financial services (4). The implementation narratives in these cases were mostly post-implementation (56). 11 provided a pre-implementation narrative and 18 cases provided both a pre- and post-pre-implementation narrative. For more details on the cases Appendix A can be consulted. Appendix B provides the sources. From these cases 253 different instances were identified where national culture influenced user adoption through user adoption antecedents. These 253 indications of national culture were also compared to the actual national culture scores by Hofstede (1980). This was done to check if the reservations about the national culture scores are valid. Appendix E shows the comparison and confirms that while observations are close to the scores, the clear difference support the argument that the scores are outdated.
Semi-quantitative data was produced by converting the absolute amount of observations to percentages. These are displayed in two tables. Table 2 depicts the distribution of national culture observations. In Table 2, percentages of the different national culture dimensions are relative to the total amount of observations (e.g. hierarchy is 10,64% out of 253 instances). Percentages of the Low/High columns are relative to their own totals (e.g. 4 hierarchy observations are Low which is 14,81% of the total of 27 observations).
decreased adoption observations. In the next column Table 3 shows 16 observations of high power distance that lead to increased user adoption. This is 9,04% of 76 observations that increased user adoption.
Table 2 - Distribution of National Culture Observations
Power Distance Low High
Hierarchy 27 10,67%* 4 14,81%** 23 85,19% Communication 23 9,09% 3 13,04% 20 86,96% Work supervision 13 5,14% 4 30,77% 9 69,23% De-centralization 10 3,95% 7 70,00% 3 30,00% Job satisfaction 5 1,98% 0 0,00% 5 100,00% Total 78 30,83% 18 23,08% 60 76,92%
Individualism Individualist Collectivist
Relationships 39 15,42% 15 38,46% 24 61,54% Performance promotion 11 4,35% 9 81,82% 2 18,18% Training 7 2,77% 6 85,71% 1 14,29% Company loyalty 1 0,40% 0 0,00% 1 100,00% Total 58 22,92% 30 51,72% 28 48,28%
Uncertainty Avoidance Low High
Technology & innovation 47 18,58% 5 10,64% 42 89,36%
Degree of laws/rules 13 5,14% 1 7,69% 12 92,31%
Punctuality 10 3,95% 5 50,00% 5 50,00%
Total 70 27,67% 11 15,71% 59 84,29%
Masculinity Feminine Masculine
Goals 10 3,95% 0 0,00% 10 100,00%
Assertiveness 9 3,56% 0 0,00% 9 100,00%
Total 19 7,51% 0 0,00% 19 100,00%
Long-term Orientation Short-term Long-term
Targets 24 9,49% 18 75,00% 6 25,00%
Tradition 4 1,58% 2 50,00% 2 50,00%
Total 28 11,07% 20 71,43% 8 28,57%
Table 3 – Distribution of User Adoption Observations among National Culture Dimensions Decreased Adoption 177 69,96%* Increased Adoption 76 30,04% Power Distance High PD 44 24,86%** 16 9,04% Low PD 9 5,08% 9 5,08% 53 29,94% 25 14,12% Individualism Individualist 26 14,69% 4 5,26% Collective 5 2,82% 23 30,26% 31 17,51% 27 35,53% Uncertainty Avoidance High UA 49 27,68% 10 13,16% Low UA 7 3,95% 4 5,26% 56 31,64% 14 18,42% Masculinity Masculine 16 9,04% 3 3,95% Feminine 0 0,00% 0 0,00% 16 9,04% 3 3,95% Long-term Orientation Short-term 20 11,30% 0 0,00% Long-term 1 0,56% 7 9,21% 21 11,86% 7 9,21%
*Relative to the total number of observations (253)
**Relative to the total number of Low Adoption observations (177)
Power Distance
High power distance. Table 3 shows that power distance observations mainly led to a decreased user adoption, specifically high power distance observations. Figure 3 shows the distribution of high power distance observations. The causal chain starts with the national culture dimension and the amount of indications found (i.e. high power distance, 60 indications). Next, the distribution of patterns that were found. For example, two of the high power distance observations lead to an increased (high) performance expectancy which in turn lead to an increased user adoption. By depicting it in this way the most frequent antecedents can be easily identified. All subsequent figures (i.e. figure 3 to 11) function similarly. In this case it shows that the main antecedent that power distance influences is Manifestation of Resistance which consequently leads to decreased user adoption.
The high power distance led to an increase in manifestations of resistance. For example, a high power indication was found in case MS4: “the communication was: this is what is happening!, which in turn led to the manifestation of resistance where users admitted that “people found it hard to commit”. This had a negative effect on user adoption.
However, the second most frequent antecedent affected by high power distance increased user adoption. The high power distance laid the ground for a strong social influence over the users which generally increased their user adoption. For instance, in a case in China the power distance was perceived as being so high that a classic champion for the project wasn’t even needed. This was represented in the following quote: “We suggest that in the Chinese context, the concept of a champion, as distinct from top
Power Distance - High
Performance Expectancy - High Performance Expectancy - Low
Effort Expectancy - Low Social Influence - High Social Influence - Low Facilitating Conditions - High Facilitating Conditions - Low
Perceived Threat - High Manifestation of Resistance - High Manifestation of Resistance - Low
User Adoption - High User Adoption - Low User Adoption - Low User Adoption - High
User Adoption - Low User Adoption - High User Adoption - Low User Adoption - Low User Adoption - Low User Adoption - High User Adoption - Low
management, is not important because the top manager is perceived to be champion.”. The social influence from top management was so strong that the following was said: “What top management insists on will happen.” and “Change is accepted if it is demanded” (RL11). This led to an increase in user adoption.
The remaining antecedents that were frequently observed to be affected by a high power distance are Performance Expectancy (10%) and Facilitating Conditions (11,7%), which both caused the user adoption to decrease.
Low power distance. Table 3 shows that the 18 observations of low power distance are evenly distributed between negatively (9) and positively (9) influencing user adoption. The impact of low power distance is by itself already very low since there are only 18 out of 253 observations (7,1%). This end of the continuum was mostly linked to a high performance expectancy. A good example of this is a project where the participants noted that it was not top-down and hierarchical driven: “This also shows that the participants in the SAP project felt that the implementation was a collaborative venture and was not top-down and hierarchically driven.”. This indication of low power distance led to a point where the participants stated, “all of the goals for our unit were reached” indicating an increase in performance expectancy (CS23). However, this was, as can be seen in Figure 4, only found 4 times and it can therefore be said that in the end a low power distance did not have a big influence either positive or negative.
Power Distance - Low
Performance Expectancy - High Performance Expectancy - Low
Social Influence - High Facilitating Conditions - High Facilitating Conditions - Low
Perceived Threat - High Perceived Threat - Low Manifestation of Resistance - High
User Adoption - High User Adoption - Low User Adoption - High User Adoption - High User Adoption - Low User Adoption - Low User Adoption - High User Adoption - Low User Adoption - Low
18 4 (22,2%) 2 (11,1%) 2 (11,1%) 3 (16,7%) 2 (11,1%) 1 (6%) 1 (6%) 3 (16,7%)
In sum, especially a high power distance can have a significant influence on the implementation of an integrative IS system. Most notably an increase in manifestations of resistance can be expected, which has a negative influence on user adoption. On the other hand, a greater social influence gained by high power distance will make it easier to make users commit to the project.
Individualism
Individualism is divided into individualist and collective culture indications. There are 58 observations in total (22,92% of 253 instances) of which 30 were individualist and 28 that indicated a collectivist culture (see Table 2). This makes the distribution fairly even across the continuum. Moreover, Table 3 shows that the same is true for individualism influencing user adoption. 31 observations were found to influence user adoption negatively where 27 were found to be a positive influence. Firstly, the individualist observations will be analyzed followed by the collectivist observations.
Individualism. Table 3 gives a clear picture of what an individualist culture’s influence is on user adoption. Of the 30 observations 26 negatively influenced user adoption. Figure 5 displays the distribution of these observations and it shows that the main antecedent is Manifestation of Resistance.
The individualist nature displayed in the following quote: “This shed light on the existence of a continuing struggle to impose one organization identity as being dominant over and against the other competing alternative.”. This led to “us-versus-them” scenarios and even talk about “war between the two adult sites” (RL52). This
Figure 5 - Distribution of Antecedents influenced by Individualist observations
Individualism
Performance Expectancy - Low Effort Expectancy - Low
Social Influence - High Facilitating Conditions - High Facilitating Conditions - Low
Perceived Threat - High Manifestation of Resistance - High
User Adoption - Low User Adoption - Low
User Adoption - High User Adoption - Low User Adoption - Low User Adoption - Low User Adoption - High
example clearly illustrates the manifestations of resistance that can come forth out of an individualistic culture which negatively influences user adoption.
Another antecedent that was found repeatedly as a cause for decreased user adoption was low facilitating conditions (Figure 5, 8 observations). This was due to the ‘individualist’ nature of training provided during the implementation. Case RL12 shows this individualist nature as follows: “team members developed their own skills based in self-study and self-training during the project.” which led to “Project team members complained about their training” which negatively influenced the user adoption of the system.
Collectivism. The results show that collectivism is the polar opposite of an individualist culture. Figure 5 shows that individualist observations generally decrease user adoption, collectivist observations however mostly have a positive influence on user adoption. 23 of the 28 collectivist observations were linked to increasing user adoption (Table 3).
Figure 6 depicts the distribution of collectivist observations and it shows that it has a strong positive effect on facilitating conditions which in turn increases the user adoption (12 observations). While it was training that was most affected on the individualist side, it is the compatibility of the system that mostly benefits of the collectivist culture and therefore increases the facilitating conditions as is illustrated in the following example. The collective culture was indicated by this quote: “Careful attention was given to the development of a participative and flexible culture”, which led to users responding as follows: “Professionals considered the acceptance of EMR to
Figure 6 - Distribution of Antecedents influenced by Collectivist observations
Collectivism
Performance Expectancy - High Effort Expectancy - High
Social Influence - High Social Influence - Low Facilitating Conditions - High Facilitating Conditions - Low
Perceived Threat - Low Manifestation of Resistance - High
28 4 (14,3%) 1 (3,6%) 4 (14,3%) 1 (3,6%) 12 (42,9%) 1 (3,6%) 1 (3,6%) 3 (10,7%)
Manifestation of Resistance - Low
1 (3,6%)
User Adoption - High User Adoption - High
User Adoption - Low User Adoption - High User Adoption - Low User Adoption - High User Adoption - High
be directly linked to their participation in these problem-solving groups” showing a positive effect of facilitating conditions and therefore also user adoption (EX1).
Collectivism is the greatest positive influence on user adoption from all the dimensions (see Table 3). More specifically, based on these results, one can say that collectivist cultures will be more facilitating towards implementing IS systems whereas individualist cultures will make it more challenging.
Uncertainty Avoidance
Uncertainty avoidance is divided into indications of high and low uncertainty avoidance. There are 70 observations in total which makes it the second most observed national culture dimension (27,67% of 253 instances, see Table 2). Table 2 shows most observations were of high uncertainty avoidance (59) versus 11 of low uncertainty avoidance. Moreover, the majority of the observations led to a decrease in user adoption (56 observations) opposed by 14 observations that increased user adoption (Table 3). Specifically, it is high uncertainty avoidance that has the greatest negative influence on user adoption of all the dimensions. Firstly, the high uncertainty avoidance observations will be analyzed followed by the low uncertainty avoidance observations.
High uncertainty avoidance. High uncertainty mostly leads to a decrease in
user adoption (see Table 3). Of the 59 observations, 49 negatively influenced user adoption. Figure 7 depicts the distribution of these observations and shows that it mainly leads to manifestations of resistance and an increased perceived threat. Uncertainty avoidance was identified through quotes such as this: “During the
Uncertainty Avoidance - High
Performance Expectancy - High Performance Expectancy - Low
Effort Expectancy - Low Social Influence - High Facilitating Conditions - High Facilitating Conditions - Low
Perceived Threat - High Manifestation of Resistance - High
User Adoption - High User Adoption - Low User Adoption - Low
User Adoption - High User Adoption - Low User Adoption - High User Adoption - Low User Adoption - Low User Adoption - Low
59 5 (8,5%) 9 (15,3%) 6 (10,2%) 2 (3,4%) 3 (5,1%) 3 (5,1%) 13 (22%) 18 (30,5%)
implementation process tensions often appeared, especially in family firms, as a result of fear of some managers or employees that they could lose part of the responsibilities, power or recognition that had gone with the work they did before the new technology was introduced” (MS10).
The biggest influencing antecedent was manifestation of resistance. Roughly 30% of the observations lead to a manifestation of resistance which in turn led to a decreased user adoption (see Figure 7). An example of this can be seen in case EX3, the uncertainty avoidance indication showed a clear sign for fear of new technology: “I remember that there was a lot of angst with the physicians around jumping into CPOE and starting in with the new technology. I would say fear..”. This indication of uncertainty avoidance lead to a manifestation of resistance: “There was a tremendous pushback by a few doctors early on who swore this was going to make care more dangerous, who swore that docs were going to rebel.” which lead to decreased user adoption.
The second antecedent that was frequently observed was perceived threat. While in 2 instances the increased perceived threat led to higher user adoption out of fear of being let go, the majority of the observations led to a decrease in user adoption (see Figure 7, 11 out of 13). For example: “Before the start of the project, when hearing the double label of the program, nurses were afraid that the time they would save because of the more efficiently organized ward they would create during the project would result in cutbacks on personnel.” is a quote that represents uncertainty avoidance which led to an observation of perceived threat: “The double label of the program used in the organization, seems to create some suspicion among nurses.” (CS20). This example led to a decrease in user adoption.
Uncertainty Avoidance - Low
Performance Expectancy - High Performance Expectancy - Low
Effort Expectancy - Low Social Influence - High Facilitating Conditions - Low
Perceived Threat - Low
User Adoption - High User Adoption - Low
User Adoption - High User Adoption - Low User Adoption - High User Adoption - Low
11 2 (18,2%) 3 (27,3%) 1 (9%) 1 (9%) 3 (27,3%) 1 (9%)
Low uncertainty avoidance. There were relatively little low uncertainty avoidance observations (11 observations, see Table 2). Remarkably, most of these observations, just like high uncertainty avoidance, led to a decreased user adoption (7 out of 11, see Table 3). Figure 8 depicts the distribution of antecedents influenced by low uncertainty avoidance and it shows that performance expectancy and facilitating conditions were both observed to be negatively influences on multiple occasions.
The low uncertainty avoidance observations that were coded with “technology and innovation” are 100% accountable for the 4 observations that lead to an increased user adoption. However, low uncertainty avoidance observations in the form of lack of “punctuality” mostly had a negative influence on user adoption through decreased performance expectancy. For example: “The BOMs [bills of material] were loose and standard routings were non-existent, it was very dysfunctional” indicated low uncertainty avoidance which in turn led to decreased performance expectancy which is represented in the following quote: “Consequently, both companies had difficulty adhering to processes that were newly developed by the ERP system” (RL17). This led to decreased user adoption. While there are just as many observations of low facilitating conditions there was no link that came up more than once making it hard to draw any worthy conclusions.
In sum, one can expect a high level of resistance towards new IS systems in countries where uncertainty avoidance is high. Conversely, a country with low uncertainty avoidance will have an easier time gaining support for an IS project but will have to watch out for possibility of decreased data quality which would decrease usability of the system.
Masculinity
Figure 9 shows that masculinity led to an increase manifestations of resistance which in turn decreased user adoption (6 out of 19 observations). Half of these observations (3 out of 6) were what Rivard & Lapointe (2012) call ‘persistence of former behavior’. For example, “the pace and competition in the sales group were high” serving as a masculinity indication which led to the manifestation of resistance quote “The sales staff resisted the additional work involved, the slowing in pace the system created” which in turn led to a decrease in user adoption (RL51).
Therefore, it can be said that cultures of a more masculine nature might be more inclined to resist because of a preference to stick to what they know.
Long- and short-term Orientation
There are 28 observations in total which makes it the second least observed national culture dimension (11,07% of 253 instances, see Table 2). Table 2 shows that most observations were of short-term orientation (20) versus 8 of long-term orientation. Table 3 shows that the majority of the observations led to a decrease in user adoption (21 observations) opposed by 7 observations that increased user adoption. Specifically, it is short-term orientation that has the greatest negative influence on user adoption. Firstly, long-term orientation observations will be analyzed followed by the short-term observations.
Masculine
Performance Expectancy - High Performance Expectancy - Low
Social Influence - High Facilitating Conditions - High Facilitating Conditions - Low Manifestation of Resistance - High Manifestation of Resistance - Low
User Adoption - High User Adoption - Low
User Adoption - High User Adoption - Low User Adoption - Low User Adoption - High User Adoption - Low
19 1 (5,3%) 4 (21,1%) 3 (15,8%) 1 (5,3%) 3 (15,8%) 6 (32,6%) 1 (5,3%)
Long-term orientation. Long-term orientation generally leads to an increase in user adoption (see Table 3). 7 out of 8 observations influenced user adoption positively. Figure 10 displays the distribution of these observations and shows that it mainly has a positive effect on facilitating conditions (5 out of 8 observations). The long-term orientation indications were identified through either the literal use of the word ‘long-term’ (“Training had a long-term rather than a short-term focus”, CS6) or the reference towards an approach that rather uses more than less time (“Rather than being rushed to go live, the new VLE was introduced in an incremental fashion.”, CS6). Long-term oriented cultures seemed to mostly benefit through higher facilitating conditions which leads to higher user adoption. These facilitating conditions were in the form of better training (“Training was really useful”, CS6), better compatibility of the system (“We need to get really busy and at least get the key strategic elements of our long-term plan and get them in place and start to drive Threshold [the IT system] from those business propositions”, CS2) and especially, a better organizational infrastructure to support the project (“This [phased implementation] will allow the necessary business evaluation and also perfect the roll-out processes, techniques, and tools, prior to subsequent roll-out activities.”, MS5).
Short-term orientation. All instances show that that short-term orientation’s influence on user adoption is negative (20 observations, see Table 3). Figure 11 portrays the distribution of these observations and shows that short-term orientation functions as
Long-term Orientation
Performance Expectancy - High Social Influence - High Facilitating Conditions - High Facilitating Conditions - Low
User Adoption - High User Adoption - High
User Adoption - Low User Adoption - High
8 1 (12,5%) 1 (12,5%) 5 (62,5%) 1 (12,5%) Short-term Orientation
Performance Expectancy - Low Effort Expectancy - Low Facilitating Conditions - Low Manifestation of Resistance - High
User Adoption - Low User Adoption - Low
User Adoption - Low User Adoption - Low
20
6 (30%) 1 (5%) 8 (40%) 5 (25%)
the opposite of long-term orientation. Namely, short-term orientation has a negative effect on facilitating conditions. However, short-term orientation also showed to effect performance expectancy and manifestation of resistance negatively on multiple occasions. Similarly to long-term orientation, short-term orientation indications were identified through either the literal use of the word ‘short-term’ (“They/we don’t give a damn - we plan only short-term”, CS2) or the reference towards an approach that rather uses less than more time (“The new system had to be implemented within a limited time frame and there was no room for delay”, MS4).
Short-term oriented cultures seemed to mostly negatively affect facilitating conditions which leads to lower user adoption. The facilitating conditions were mainly hindered in the form of less effective training which is clearly represented in the following quote: “For example, some staff had one day of training three weeks before we went live, and production staff got about a four-hour overview. It was too much too quickly for the staff and a lot of it went right by them” (RL17). Furthermore, again as a direct opposite to long-term orientation, the compatibility of the system was negatively affected: “During development we could not prioritize according to the needs of the business, instead it was a time box. (…) It wasn’t easy” (MS6).
In sum, short- or long-term orientation of a culture mainly influences the facilitating conditions of a project. One can expect that long-term oriented cultures have a more supporting environment for the IS as opposed to a short-term oriented culture.
Discussion
The National Culture Information System Adoption model (NCISA) shown in Figure 12 shows that national culture influences user adoption through various relationships. This summary only depicts the relationships that were observed frequently. All arrows are positive relationships except for the arrows specifically marked as negative. It illustrates the fact that culture mostly, although not exclusively, influence that amount of resistance towards IS adoption. User adoption antecedents influence user adoption as follows. Antecedents based on user acceptance (A) have a positive relationship to user adoption (e.g. performance expectancy is increased, user adoption is increased). Antecedents based on user resistance (R) have a negative relationship to user adoption (e.g. perceived threat is increased; user adoption is decreased). With regard to the antecedents that are based on both resistance and acceptance literature (R | A). If the antecedent surfaces in the form of resistance, the relationship is negative. If it surfaces in the form of acceptance, the relationship is positive.
Moreover, Figure 12 shows which antecedents were most frequently observed to be influenced by national culture. Firstly, regarding effort expectancy, this antecedent, based in user acceptance literature (Venkatesh et al., 2003), was not frequently observed to be influenced by national culture. This does not however, discard it as being an antecedent to user adoption. The data only suggests that effort expectancy is not influenced by national culture. Furthermore, manifestation of resistance was most
User Adoption Antecedents National Culture Power Distance Individualism Uncertainty Avoidance Masculinity Long-term Orientation Social Influence (R | A) Facilitating Conditions (R | A)
Performance Expectancy (A) Effort Expectancy (A)
Perceived Threat (R) Manifestation of Resistance (R)
frequently observed as being influenced by national culture. Rivard & Lapointe (2012) present manifestation of resistance as the core element of resistance towards IT. This study shows that it is also the core element of IS user adoption when looking at the influence of national culture. The second important element is facilitating conditions, which is based on both resistance and acceptance literature (Van Offenbeek et al., 2013; Venkatesh et al., 2003). It shows that national culture has a great influence on the environment and supporting components surrounding the implementation. While these are important antecedents, the individual influence of the different national culture dimensions are as follows.
Implementations in cultures with a high power distance can expect an increase in manifestations of resistance which impedes user adoption. On the other hand, these high power distance cultures will have the benefit of greater social influence over their users.
With regards to individualism, this dimension paints a clear picture when it comes to user adoption. Individualist cultures are prone to more resistance and are notably less facilitating towards IS implementations then collectivist cultures.
Uncertainty avoidance was frequently observed in relation to user adoption. High uncertainty avoidance cultures will be considerably more resistant to IS implementations than low uncertainty avoidance cultures. However, low uncertainty avoidance cultures are susceptible to decreased punctuality in the usage of a system which could lead to decreased data quality in the system.
Furthermore, long-term orientation mainly had an influence on the facilitating conditions supporting an IS implementation. Short-term orientation has the exact opposite effect namely, negatively affecting the facilitating conditions of an IS implementation. Long-term orientated cultures will therefore be more supportive than short-term orientated cultures.
Lastly, the masculinity dimension was noteworthy because it was the least observed dimension as well as there being no indications of femininity. Therefore, any conclusion might be less supported and will need further investigation. The data suggests that high masculinity led to an increase in manifestations of resistance.
resistance towards the implementation. This does not mean that cultures who do not fit this description are unsuitable for such projects. The analysis of the cases also revealed situations where individualist culture actually benefited from their cultural orientation because the training provided matched their cultural background. Observations of uncertainty avoidance revealed similar situations. While in most cases it will have a negative effect, there were moment where high uncertainty avoidance actually led to a system of higher quality since they were so careful in the development. Masculine cultures can also improve user adoption since the competitive nature causes the users to work harder to make the system a success. The only dimension that was not seen to have a positive side was short-term orientation.
Therefore, it can be concluded that, while there are cultures who will be more supporting of technological change, the success of an implementation lies more in the ability of the practitioner to adapt to the national culture of the region where the implementation takes place.
Theoretical implications
This study has shown the influence of national culture on user adoption through acceptance and resistance antecedents (i.e. user adoption antecedents) and thereby adds to existing IS adoption literature as well as user acceptance and resistance theory. This study answers the call for a comprehensive framework as well as a more extensive country sample regarding to culture effects on IS implementations (D. Ford et al., 2003). With regard to acceptance and resistance theory, this study supports the theory of Van Offenbeek et al. (2013) that acceptance and resistance are both critical factors in the case of IS adoption.
Furthermore, this study add to national culture literature by comparing national culture indications based on Hofstede’s national culture dimensions (1980, 2001) to the controversial national culture scores (Smith, 2006). The observations were mostly close, but still different than the original scores which supports the argument that the Hofstede national culture scores are outdated and are unreliable to use in a study such as this. Further details on this analysis can be found in Appendix E.
Practical implications
Due to continued globalization, companies will become more and more geographically and culturally dispersed (Ricaud, 2006). Tailoring the process of an IS implementation to the national culture preferences will decrease resistance and increase acceptance hereby increasing the chance of successful IS adoption. In multi-national companies, it would therefore mean that regional implementation strategies could be a solution to avoiding national culture issues with the implementation.
For example, it is evident that a short-term perspective will have a negative influence on an IS implementation. Therefore, practitioners can stress the need for a long-term perspective in short-term oriented cultures. With regard to power distance, when a practitioner implements an IS in a high power distance culture he can now exploit the increased social influence to his or her advantage while at the same time being aware of the potential for increased resistance. The individualism dimension can be used to adapt training programs. Individualist cultures prefer personal, one-on-one training whereas collective culture prefer a group approach. In high uncertainty avoidance cultures, relatively high resistance can be expected. Therefore, a practitioner can increase communication efforts to inform users of every step that is being taken to ensure success (e.g. share contingency plans in case of problems) hereby decreasing uncertainty. In this manner, the results of this study will support the implementation of information systems around the world. It also shows that, as said before, successful IS adoption lies in the ability of the practitioner to adapt to the national culture of the region where the implementation takes place.
Limitations & future research
Moreover, not every selected case could be individually discussed due to time and resource restraints which is also a limitation. To account for this all sources are provided in Appendix B. The last limitation is the variety of countries involved in the sample. The distribution between countries is skewed towards the US and the UK. This can be explained by fact that these countries draw bigger attention of researchers or IS implementation are more common in these countries.
The goal of this study is theory development and therefore future research can be directed at testing any of the patterns found in this study as well as testing or potentially revising the entire model. More specifically, the qualitative nature of the study makes the conclusions that were drawn subjective and therefore calls for more quantitative testing of the relationships. The individual relationships should be analyzed, tested and verified empirically and statistically. Furthermore, while this study was comprehensive in the range of cases that were included in the sample, the data-set was skewed towards the UK and USA suggesting that a future research should aim to remedy this.
Also, practitioners who use this study in their implementations might experience unforeseen “side-effects” when for instance forcing a short-term oriented culture to adopt a long-term perspective. Case studies can be built around the success or failure of IS adoption that has accounted for the national culture influences. Lastly, due to the rise in of multi-cultural teams in multinational companies it will be interesting to see how this influences the adoption of IS.
References
Ajzen, I. (1991). The theory of planned behavior. Orgnizational Behavior and Human Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T
Ajzen, I., & Fishbein, M. (1975). Belief, attitude, intention and behavior: An introduction to theory and research. Reading, MA: Addison-Wesley.
Bala, H., & Venkatesh, V. (2016). Adaptation to Information Technology : A Holistic Nomological Network from Implementation to Job Adaptation to Information Technology : A Holistic Nomological Network from Implementation to Job Outcomes. Management Science, 62(1).
https://doi.org/http://dx.doi.org/10.1287/mnsc.2014.2111
Boonstra, A., Versluis, A., & Vos, J. F. J. (2014). Implementing electronic health records in hospitals: a systematic literature review. BMC Health Services Research, 14(1), 370. https://doi.org/10.1186/1472-6963-14-370
Davis, F. D., Bagozzi, R., & Warshaw, P. (1989). User acceptance of Computer Technology: A Comparison of two Theoretical Models. Management Science, 35(8), 982–1003. https://doi.org/10.1287/mnsc.35.8.982
De Long, D. W., & Fahey, L. (2000). Diagnosing cultural barriers to knowledge management. Academy of Management Perspectives, 14(4), 113–127. https://doi.org/10.5465/AME.2000.3979820
Delone, W. H., & McLean, E. R. (2003). The DeLone and McLean Model of Information Systems Success: A Ten-Year Update. Journal of Management Information Systems, 19(4), 9–30.
https://doi.org/10.1080/07421222.2003.11045748
Eisenhardt, K. M. (1989). Building Theories from Case. The Academy of Management Review, 14(4), 532–550.
Ferneley, E. H., & Sobreperez, P. (2006). Resist, comply or workaround? An examination of different facets of user engagement with information systems. European Journal of Information Systems, 15(4), 345–356.
Ford, D., Connelly, C. E., & Meister, D. B. (2003). Information systems research and Hofstede’s culture’s consequences: An uneasy and incomplete partnership. IEEE Transactions on Engineering Management, 50(1), 8–25.
https://doi.org/10.1109/TEM.2002.808265
Ford, J., Ford, L., & D’Amelio, A. (2008). Resistance to Change: The Rest of the Story. The Academy of Management Review, 33(2), 362–377.
Gregory, K. L. (1983). Native-View Paradigms: Multiple Cultures and Culture Conflicts in Organizations. Administrative Science Quarterly, 28(3), 359. https://doi.org/10.2307/2392247
Hall, E. T. (1983). Dance of Life. New York: Anchor.
Hofstede, G. (1980). Culture’s consequences: National differences in thinking and organizing. Beverly Hills: Sage.
Hofstede, G. (1984). Cultural dimensions in management and planning. Asia Pacific Journal of Management, 1(2), 81–99. https://doi.org/10.1007/BF01733682 Hofstede, G. (1998). Identifying Organizational Subcultures: An Empirical Approach.
Journal of Management Studies, 35(1), 1–12. https://doi.org/10.1111/1467-6486.00081
Hofstede, G. (2001). Culture’s consequences: Comparing values, behaviors, institutions and organizations across nations. Thousand Oaks: Sage.
Hofstede, G. (2006). What Did GLOBE Really Measure? Researchers’ Minds versus Respondents’ Minds. Journal of International Business Studies, 37(6), 882–896. Hofstede, G. (2010). The GLOBE debate: Back to relevance. Journal of International
Business Studies, 41(1), 339–1.
Hofstede, G., & Bond, M. H. (1988). The Confucius connection: From cultural roots to economic growth. Organizational Dynamics, 16(4), 5–21.
https://doi.org/10.1016/0090-2616(88)90009-5
Hu, P. J., Chau, P. Y. K., Liu Sheng, O. R., & Tam, K. Y. (1999). Examining the Technology Acceptance Model Using Physician Acceptance of Telemedicine Technology. Journal of Management Information Systems, 16(2), 91–112. https://doi.org/10.1080/07421222.1999.11518247
Javidan, M., House, R. J., Hanges, P. J., Dorfman, P. W., & Gupta, V. (2004). Culture, leadership, and organizations: The GLOBE study of 62 societies. Sage. Retrieved from
https://books-google-nl.proxy-ub.rug.nl/books?hl=nl&lr=&id=4MByAwAAQBAJ&oi=fnd&pg=PP1&dq=GLOB
E&ots=7heEHqcb7F&sig=8Ox79dAl1CLd2sHJ030bLcI-qcU#v=onepage&q=GLOBE&f=false
Joshi, K. (1991). A Model of Users’ Perspective on Change : The Case of Information Systems Technology Implementation. MIS Quarterly, 15(2), 229–242.
https://doi.org/10.2307/249384
Kim, H. W., & Kankanhalli, A. (2009). Investigating user resistance to information systems implementation: A status quo bias perspective. MIS Quarterly, 33(3), 567– 582.
Kogut, B., & Singh, H. (1988). The Effect of National Culture on the Choice of Entry Mode. Journal of International Business Studies, 19(3), 411–432.
Krijnse Locker, N., Vos, J. F. J., & Boonstra, A. (2016). Understanding national culture effects on user behavior in integrative IS implementations. Procedia Computer Science, 100, 289–297. https://doi.org/10.1016/j.procs.2016.09.157
Kroeber, A. L., & Kluckhohn, C. (1952). Culture: A critical review of concepts and definitions. Harvard University: Papers. Peabody Museum of Archaeology & Ethnology.
Larsson, R., Driver, M., Eneroth, K., Finkelstein, S., George, D., Kreiner, P., …
Lindgren, U. (1993). Case Survey Methodology: Quantitative Analysis of Patterns across Case Studies. Academy of Management Journal, 36(6), 1515–1546.
Lee, H. G., & Clark, T. H. (1996). Market Process Reengineering through Electronic Market Systems: Opportunities and Challenges. Journal of Management
Information Systems, 13(3), 113–136.
https://doi.org/10.1080/07421222.1996.11518136
Leidner, D. E., & Kayworth, T. (2006). Review: A review of culture in information systems research: Toward a theory of information technology culture conflict. MIS Quarterly, 30(2), 357–399. https://doi.org/10.2307/25148735
Lucas, W. A. (1974). The Case Survey Method: Aggregating Case Experience. Santa Monica: Rand.
Markus, M. L. (1983). Power, Politics, and MIS implementation. Communications of the ACM, 26(6), 430–444.
Martin, J., & Meyerson, D. (1988). Organizational cultures and the denial, channeling and acknowledgement of ambiguity. In Managing ambiguity and change (pp. 93– 126). New York: Wiley.
Moore, G. C., & Benbasat, I. (1991). Development of an Instrument to Measure the Perceptions of Adopting an Information Technology Innovation. Information Systems Research, 212519(3), 192–222.
Muk, A., & Chung, C. (2015). Applying the technology acceptance model in a two-country study of SMS advertising. Journal of Business Research, 68(1), 1–6. https://doi.org/10.1016/j.jbusres.2014.06.001
Murcia, M. A. P., & Whitley, E. A. (2007). The effects of national culture on ERP implementation: a study of Colombia and Switzerland. Enterprise Information Systems, 1(3), 301–325. https://doi.org/10.1080/17517570701504294
Newig, J., & Fritsch, O. (2009). The Case Survey Method and Applications in Political Science. APSA 2009.
Newman, K. L., & Nollen, S. D. (1996). Culture and Congruence: The Fit between Management Practices and National Culture. Journal of International Business Studies, 27(4), 753–779.
