Content uploaded by Rogier van de Wetering
Author content
All content in this area was uploaded by Rogier van de Wetering on May 14, 2020
Content may be subject to copyright.
Agility through EAM and the Role of Alignment
Americas Conference on Information Systems
1
Improving Agility Through Enterprise
Architecture Management: The Mediating
Role of Aligning Business and IT
Completed Research
Maurice Pattij
Open University, the Netherlands
maurice.pattij@ou.nl
Rogier van de Wetering
Open University, the Netherlands
rogier.vandewetering@ou.nl
Rob J. Kusters
Open University, the Netherlands
rob.kusters@ou.nl
Abstract
The economic environment of contemporary organizations is becoming increasingly dynamic.
Organizational agility fosters sustainable competitive advantage under these turbulent conditions. Prior
research demonstrated that strategic IT alignment could enhance organizational agility. Many
organizations implemented an enterprise architecture management (EAM) function to achieve benefits
such as strategic IT alignment and agility. However, there is little research that explains the pathways
between these focal concepts. Hence, we ground our work in the dynamic capabilities view and develop a
conceptual model to explain how EAM investments lead to agility mediated by strategic IT alignment. We
conducted survey research and collected a sample of 110 respondents. Based on this dataset, we performed
a PLS-SEM and cluster analysis to test our model and associated hypotheses. Our results indicate that EAM
enhances organizational agility. Strategic IT alignment mediates this effect. Lastly, our results showcase the
complementary effect of conducting a PLS-SEM and cluster analysis.
Keywords
Enterprise Architecture Management, Enterprise Architecture, Strategic IT Alignment, Organizational
Agility, PLS-SEM, Cluster Analysis.
Introduction
In the late 1980s, organizations rapidly started to increase the use of information systems (IS) and
information technology (IT) to improve the efficiency and effectiveness of their business processes. As a
result, the IS landscape became larger and more complex. To manage this complexity, organizations started
to model IS systems from various perspectives, including the data, process, and network perspective,
leading to the first enterprise architecture (EA) frameworks to describe EA artifacts (Lapalme et al. 2016;
Zachman 1987). In years past, EA has developed as a practice widely used to manage an organization’s
complex IT and IS landscape. However, the organization’s dynamic environment, resulting from (for
example) the emergence of new disruptive technologies, led to challenges to remain competitive (Lapalme
et al. 2016). To face these challenges and to enhance sustainable competitive advantages, organizational
agility, hereafter referred to as ‘agility,’ is considered crucial (Sambamurthy et al. 2003). Currently, many
organizations implement an enterprise architecture management (EAM) function to manage an
organization’s current EA, develop its desired state, and guide improvement projects using roadmaps and
EA artifacts (Aier et al. 2011). Articulated benefits of EAM include the ability to utilize EA to foster strategic
IT alignment, hereinafter referred to as ‘alignment’ (Aier et al. 2011; Kotusev et al. 2015) and agility
(Abraham et al. 2012; Lux et al. 2010). While decision-makers make substantial investments in their EAM
efforts to align their organization’s business and IT (Winter et al. 2010), practitioners and academics still
struggle to understand the range and reach of these benefits and how they can be achieved in practice
Agility through EAM and the Role of Alignment
Americas Conference on Information Systems
2
(Kotusev et al. 2015). Additionally, there is limited empirical research supporting these claims (Abraham et
al. 2012; Lux et al. 2010; Niemi and Pekkola 2016). Nevertheless, scholars appear to have a common
understanding that EA and EAM benefit realization comprises of benefits resulting directly from EA and
EAM efforts and different levels of indirect effects, such as benefits on an organizational level (Foorthuis et
al. 2016; Niemi and Pekkola 2016; Pattij et al. 2019; Schmidt and Buxmann 2011; Shanks et al. 2018).
Practitioners indicate that alignment is the top goal of their EAM practice (Winter et al. 2010). Both
practitioners and academics suggest alignment as a direct benefit of EAM efforts, and agility as an indirect
benefit (Kotusev et al. 2015; Lux et al. 2010; Winter et al. 2010). However, perceptions of alignment and
agility are paradoxical. Previous research indicates that alignment impedes agility (Liang et al. 2017).
Contrarily, the results of other research proved that alignment is positively associated with agility (Bradley
et al. 2012; Tallon and Pinsonneault 2011). Hence, more research is required to understand the relation
between these focal concepts.
The goal of this research is to contribute to the empirical knowledge base on EAM benefit realization by
addressing the gaps mentioned. To achieve this, we aim to answer the following research questions: (1) To
what extent does EAM influence organizational agility and (2) what is the role of strategic IT alignment
in this value path?
Additionally, we aim to contribute from a methodological perspective. Linear dependence techniques, such
as PLS-SEM, appear to be the most common approach in quantitative research on EA benefit realization
(Foorthuis et al. 2016; Lange et al. 2016; Niemi and Pekkola 2016; Schmidt and Buxmann 2011). These
approaches are characterized by singular causation and linear relationships. However, the literature on
strategic management stresses the importance of configurational theory and suggests that an organization’s
performance results from complex causality of configurations of many variables showing non-linear
behavior. Furthermore, organizational performance could be subjected to equifinality, thus the
phenomenon that a system’s end state can result from a combination of different conditions that follow
different pathways (Fiss 2007; Ketchen and Shook 1996). Hence, to further unravel the complex nature of
EA and EAM benefit realization, we will explore the usefulness of configurational approaches to explore
how different aspects of EAM influence alignment and agility. We will conduct a cluster analysis as a
commonly used technique that is classified as an configurational approach (Ketchen and Shook 1996).
Theory and Model Development
As an appropriate theoretical lens to underline the effects of an organization’s dynamic environment, we
will use the dynamic capabilities view (DCV) to develop our model and to conceptualize EAM. The DCV
builds upon the resource-based view (RBV), which argues that scarce and valuable firm resources that are
hard to copy or replace are critical ingredients of sustained competitive advantages. Firms’ resources
include “all assets, capabilities, organizational processes, firm attributes, information and knowledge etc.”
(Barney 1991, p. 101). The DCV entails leveraging a firm’s resources to create capabilities that support
organizations to adapt to its dynamic environment. The original work of Teece et al. (1997, p. 516) defines
a dynamic capability as “the firm’s ability to integrate, build, and reconfigure internal and external
competences to address rapidly changing environments.” Synthesis of today’s academic knowledge base
reveals different views on what dynamic capabilities are and how they should be conceptualized. Some
views take a higher abstraction level and regard them as organizational ‘abilities’ (Teece et al. 1997; Winter
2003), while others perceive them as ‘capacities’ (Helfat et al. 2007). Like Eisenhardt and Martin (2000),
we take a more concrete view and consider dynamic capabilities as specific processes that realize new
resource configurations required to adapt to market change or even induce market change.
Enterprise Architecture Management
EAM has been conceptualized as a set of processes that continuously supports the development of EA (Aier
et al. 2011). Other researchers extend this view and articulate the importance of leveraging EA (Van de
Wetering 2019a) and EAM resources (Lux et al. 2010) to face environmental dynamics. Abraham et al.
(2012) underpin the differences between EAM as a dynamic capability to respond to anticipate
environmental changes, and as an improvisational capability to respond to unanticipated environmental
changes. Building upon these conceptualizations, we argue that positioning EAM as a dynamic capability
in existing typologies is not straightforward. First, activities like EA documentation (i.e., process to facilitate
Agility through EAM and the Role of Alignment
Americas Conference on Information Systems
3
the EA description), EA programming (i.e., process to facilitate setting architectural rules and guidelines),
and EA governance (i.e., process to ensure compliance with EA) could be classified as learning capabilities
(Schmidt and Buxmann 2011) that aim to improve the performed of tasks by repetition (Teece et al. 1997).
Second, EA implementation, thus the initiation and execution of change (Schmidt and Buxmann 2011), can
be considered as transformation capabilities. These capabilities include activities to recombine or renew
assets and resources (Teece et al. 1997). Third, EA communication and support and stakeholder
participation involve the integration and coordination of resources (Helfat et al. 2007; Teece et al. 1997).
Such capabilities have been identified as constituents of orchestration capabilities (Helfat et al. 2007).
We ground our EAM conceptualization in the previous work on EAM (Abraham et al. 2012; Shanks et al.
2018; Van de Wetering 2019b), dynamic capabilities, and resources orchestration literature (Queiroz et al.
2018; Teece et al. 2016). Therefore, we regard EAM to be a specific dynamic capability that allows
organizations to systematically orchestrate resources in the process of adapting to the changing economic
environment and for improving firm performance. We define EAM as a set of processes that (1) manages
leveraging EA resources to develop EA and (2) guides the continuous reconfiguration and renewal of
resources to transform an organization’s EA to new resource configurations, to adapt to or induce market
change.
Organizational Agility
Reviewing literature reveals a certain high-level consensus that agility comprises the ability to sense and
respond to opportunities and threats. This is suggested as an important ability to achieve competitive
advantages and includes activities to induce innovative changes in an organization’s products, services,
channels, and processes (Sambamurthy et al. 2003; Tallon et al. 2019; Tallon and Pinsonneault 2011).
However, a more detailed view shows differences in conceptualizations. Agility has been conceptualized as
a dynamic capability (Sambamurthy et al. 2003) and as an ordinary, or operational capability, and thus the
regular processes that are required to run the daily operations (Tallon and Pinsonneault 2011; Van de
Wetering 2019a; Winter 2003). The line between dynamic capabilities and 0perational capabilities is blurry
and depends on different facets, such as the extent of change and the radicality of change. Capabilities, such
as agility, could even have both operational as dynamic purposes (Helfat and Winter 2011). Our view on
agility remains within the threshold of the operational capabilities. We conceptualize agility as an output-
oriented capability to fulfill changing customer and market demands, to improve existing processes, or to
reduce costs (Tallon and Pinsonneault 2011; Van de Wetering 2019a). Thus, as a capability that
encompasses responding to external changes by making adjustments required to make a living in the
present time (Helfat and Winter 2011; Winter 2003).
Strategic IT Alignment
Alignment refers to the cohesion between business strategies and goals, and IT strategies and goals
(Luftman and Brier 1999). Researchers argue that this cohesion is a prerequisite to fully benefit from the
large IT investments made by organizations operating in a time in which IT has to potential to transform
industries and markets (Chan 1997; Luftman and Brier 1999; Yeow et al. 2018). Alignment has been
conceptualized as both an intended state or result (Reich and Benbasat 1996), and as a process
(Venkatraman et al. 1993). The relation between alignment as a state and the alignment process can be
described as a continuous cycle. Misalignment triggers the alignment process to facilitate the realization of
alignment, leading to a new state of alignment (Yeow et al. 2018). We can devise two key alignment
dimensions from the literature. Social alignment focuses on shared values, commitment, communication,
and mutual understanding, while intellectual alignment refers to the alignment of strategy, plans,
infrastructure, and processes (Reich and Benbasat 1996; Wu et al. 2015).
Model and Hypotheses Development
Teece et al. (2016) argue that dynamic capabilities, such as EAM processes, are crucial ingredients to foster
agility. Exploratory research of Abraham et al. (2012) provides support for this claim. Their results indicate
that EAM helps organizations to respond to unanticipated change. Survey-based research of Van de
Wetering (2019a) proves that dynamic EA capabilities have a positive effect on market agility. The vigorous
DCV suggests that dynamic capabilities, like the EAM practice, help an organization to remain competitive
(Teece et al. 2016). However, pathways to these competitive advantages could involve specific (e.g., aligned)
configurations of resources and inducing changes in operational capabilities, such as the alignment process
Agility through EAM and the Role of Alignment
Americas Conference on Information Systems
4
(Van de Wetering 2019a; Winter 2003). Supporting this claim, previous research indicates that competitive
advantages, including agility, are an indirect effect of EA and EAM efforts (Bradley et al. 2012; Lux et al.
2010; Niemi and Pekkola 2016; Pattij et al. 2019). Hence, As depicted in Figure 1, we propose that alignment
mediates the effect between EAM and agility.
Figure 1 - Research Model
EAM includes tasks to manage the development and implementation of EA and to ensure that business and
IT stakeholders from all levels participate in EA decision-making (Aier et al. 2011; Schmidt and Buxmann
2011). Using EA artifacts as input, activities like EA stakeholder participation, and EA communication and
support improve the common understanding and partnerships between IT and business (Foorthuis et al.
2016; Schmidt and Buxmann 2011). Architects working on EA implementation activities are reported to be
involved in strategy development as IT stakeholders (Lux et al. 2010). Furthermore, EAM includes EA
planning that guides the long-term planning and transformation of an organization’s IT resources (Lange
et al. 2016). EA planning or migration planning involves the cross-organizational prioritization of
transformation projects (Winter et al. 2010). This includes prioritizing non-business driven IT projects that
are required to transform to the desired state (Schmidt and Buxmann 2011). In a survey-based research,
Luftman and Brier (1999) identified IT involvement in strategy development, a common understanding
between business and IT, business/IT partnerships, and strategically prioritized IT projects as key enablers
of the alignment process. Hence, we propose the following hypothesis.
H1: Enterprise architecture management is positively associated with strategic IT alignment.
The DCV underpins the influence of path dependencies on capabilities and processes. Path dependencies
are the organization’s routines resulting from investments and decisions made in the past (Teece et al.
1997). Path dependencies can constrain future behavior and are therefore seen as determinative of agility
(Tallon and Pinsonneault 2011; Yeow et al. 2018). Some researchers argue that the alignment process
increases organizational inertia by creating rigid routines (Liang et al. 2017). Others argue that the
proximity of business and IT stakeholders in the alignment process could foster agility by identifying and
utilizing strategic options to reduce the constraints of path dependencies (Yeow et al. 2018). An example of
this is purposefully releasing or dropping resources like, for example, legacy processes that constrain agility.
Interestingly, manifestations of path dependencies are not always constraining. The alignment process
includes the routine (i.e., path dependency) of continuous collaboration between business and IT
stakeholders. This routine is bi-directional; thus, it is not only the IT that adapts to the business strategy
but also business seeks opportunities, in order to utilize the potential of IT to increase agility (Tallon and
Pinsonneault 2011). Based on this section’s argumentation, we propose the following hypothesis.
H2: Strategic IT alignment is positively associated with organizational agility.
Methodology
Data Collection and Analyses
To collect data, we distributed a survey among EA stakeholders who were expected to give a judgment on
the EA practice and its coordination effort, the alignment, and the agility of their organization (Foorthuis
et al. 2016). Using e-mail or LinkedIn, we invited 481 potential respondents employed by nationally and
internationally operating organizations located in Belgium, the Netherlands, and Luxemburg. In total, this
led to 110 (23%) useful and complete responses. A quota sampling method, based on industrial categories
derived from previous research, was used to improve external validity (Aier et al. 2011; Foorthuis et al.
2016). The survey was completed by 41 IT architects (37.3%), 11 business architects (10.0%), 17 IT managers
(15.5%), 24 business/management consultants (21.8%), 15 IT/IS consultants (13.6%) and 2 business
managers (1.8%). The dataset reveals that 23 respondents worked in accountancy, finance and banking
(20.9%), 18 in information technology (16.4%), 13 in energy and utilities (11.8%), 12 in healthcare (10.9%),
12 in transport and logistics (10.9%) and 32 in other industries (29.1%). Furthermore, the dataset contains
Agility through EAM and the Role of Alignment
Americas Conference on Information Systems
5
13 organizations with less than 100 employees (11.8%), 22 between 100 and 1000 employees (20.0%) and
75 with more than 1000 employees (68.2%). Harman’s single factor test to control for common method bias
(CMB) indicates that 40% of the variance can be explained by one factor, thus no single factor attributes to
the majority (> 50%) of the variance, suggesting that CMB is not an issue (Podsakoff et al. 2003).
We used partial least squares structural equation modeling (PLS-SEM) to analyze our path model. PLS-
SEM is a variance-based linear regression method, designed to estimate complex causal path models
containing multiple latent variables (Hair et al. 2016). We estimated our measurement model and structural
model using SmartPLS version 3.2.7. (Ringle et al. 2015). For the bootstrapping procedure, we used 5000
sub-samples to ensure the stability of results. Complementary to the PLS-SEM analysis, we conducted a
cluster analysis to find clusters of cases where statistical variance within the groups is minimized, and the
statistical variance between the groups is maximized. A common use of cluster analysis is to profile these
groups based on characteristics (Hair et al. 2014). We profiled our EAM clusters based on their alignment
and agility scores. We used SPSS version 24 (IBM Corp. 2015) to perform cluster analysis procedures and
analysis of variance (ANOVA) tests. As for sample size requirements, statistical interference issues related
to sample size are not relevant for a cluster analysis (Hair et al. 2014, p. 436). For our PLS-SEM analysis,
we applied a reliable rule of thumb based on statistical power analyses for multiple regression models (Hair
et al. 2016, pp. 25–26), indicating a minimum required sample size of 33 (max. number of arrows pointing
to a construct = 2; lowest R2 = .275). Hence, the data exceeds the minimum sample size requirements.
Measures and Construct Validity
The developed survey consisted of existing empirically validated items to enhance intern al validity. To
improve face and content validity, two scholars peer-reviewed the survey, and two practitioners pretested
the survey. All survey items were evaluated on a 7-point Likert.
Using Schmidt and Buxmann’s (2011) items, EAM was measured as a type II second-order construct (first-
order reflective and second-order formative). Seven first-order constructs comprising four items each,
measured to what extent an organization agreed with statements on (1) EA planning, (2) EA communication
& support, (3) EA documentation, (4) EA governance, (5) EA implementation, (6) EA stakeholder
participation and (7) EA programming (Schmidt and Buxmann 2011). Respondents were asked to what
extent they agreed with statements like, for example, ‘our documentation of the Enterprise IT Architecture
is updated continuously.’
Wu et al.’s (2015) items measured alignment. Eight items measured to what extent a respondent agreed
with statements on an organization’s business goals and product-, quality-, and market-oriented strategies.
For each of the eight items, a counterpart item measured to what extent the IT/IS infrastructure supports
a business goal. Thus, an example business goal-oriented item ‘We attempt to be ahead of our competitors
in introducing new products,’ has an IT/IS infrastructure support counterpart item ‘Our current systems
enable us to introduce new products earlier than our competitors.’ The product of the business-oriented
item and its IT/IS infrastructure support item were included as an item in the first-order construct to
measure alignment (Chan 1997; Tallon and Pinsonneault 2011). This method includes the catalyzing effect
of an advanced IT/IS infrastructural implementation on an organization’s business strategy (Chan 1997).
Agility was constructed as a first-order construct comprising eight items on operational, customer, and
partnering agility (Tallon and Pinsonneault 2011). Operational agility reflects the ability to adequately
redesign business processes as a response to environmental changes. Customer agility refers to the ability
to co-0pt customers in the process of gathering market intelligence to identify opportunities, and to utilize
opportunities in order to gain competitive advantages. Partnering agility concerns the ability to adequately
respond to environmental changes by forming alliances and partnerships with other organizations, such as
suppliers, distributors, and logistic providers (Sambamurthy et al. 2003; Tallon and Pinsonneault 2011).
Respondents were asked to what extent they agreed with statements like, for example, ‘our organization
can quickly and easily respond to changes in the aggregated consumer demand.’
As indicated in Table 1, we assessed first-order latent variables for reliability, convergent validity, and
discriminant validity. Construct reliability was assessed based on the Cronbach α and composite reliability
(CR). For all latent variables, both values exceeded the .7 threshold, indicating satisfactory reliability (Hair
et al. 2016, pp. 111–112). Convergent validity assessment showed that all average variance extracted (AVE)
values exceed the lower limit of .5. We assessed discriminant validity by confirming that the square roots of
Agility through EAM and the Role of Alignment
Americas Conference on Information Systems
6
the constructs’ AVE exceeded inter-construct correlations (Hair et al. 2016, pp. 115–118). To establish that
multicollinearity is not a concern for the formative constructs, we assessed the variance inflation factor
(VIF) values. None of these values exceeded the upper threshold of 5 (Hair et al. 2016, p. 143). Heterotrait-
monotrait ratio of correlation (HTMT) values were used to assess discriminant validity between the EAM,
alignment, and agility constructs. All HTMT values were less than 9, indicating discriminant validity (Hair
et al. 2016, p. 119). As for EAM, all first-order construct to item loadings were significant (p ≤.001) and
exceeded 0.70, suggesting indicator validity (Hair et al. 2016, p. 113). Regression coefficients of the first-
order constructs to the second-order EAM construct are shown in Table 1. All paths were significant (p
≤.001). Thus, measurement model outcomes indicate that the latent variables are fit for their intended use.
β (EAM)
1
2
3
4
5
6
7
8
9
1. EA Planning
.148
.85
2. EA Communication & Support
.176
.70
.86
3. EA Documentation
.154
.67
.76
.83
4. EA Governance
.180
.61
.80
.70
.87
5. EA Implementation
.153
.58
.69
.66
.70
.83
6. EA Stakeholder Participation
.181
.58
.78
.66
.79
.65
.91
7. EA Programming
.161
.69
.78
.75
.78
.78
.70
.82
8. Strategic IT Alignment
-
.35
.49
.42
.52
.50
.40
.49
.83
9. Organizational Agility
-
.32
.38
.33
.37
.33
.27
.31
.79
.81
AVE
.72
.73
.69
.76
.69
.82
.68
.69
.65
CR
.91
.92
.90
.93
.90
.95
.89
.94
.94
Cronbach α
.87
.88
.85
.90
.85
.93
.84
.93
.92
VIF
2.3
4.6
2.9
4.0
2.8
3.3
4.3
-
-
Table 1 - Reliability, Validity and Path Coefficients of First-Order Constructs
For cluster analysis, we computed the means of the EAM dimension, alignment, and agility. Each variable
was converted to a standardized score, resulting in variables with a standard deviation of one and a mean
of zero. We used this procedure to eliminate the effect of scale differences and to enable easier comparison
between the variables (Hair et al. 2014, pp. 434–435).
Results
Hypotheses Testing
To evaluate the structural model, we assessed the coefficients of determination (R2) values, Cohen’s effect
sizes (f2), path coefficients (β) and Stone-Geisser’s predictive relevance (Q2) (Hair et al. 2016).
Figure 2 - Estimated Effects of the Structural Model
The PLS-SEM estimation of the structural model is depicted in Figure 2. The model indicates a positive and
significant effect of EAM on alignment (β =.524; p ≤.001). The estimation of alignment explains 27.5% of
the variance (R2 =.275). Cohen’s effect size (f2 =.38) is greater than .35, indicating a strong effect. Hence, we
accept H1. Furthermore, the results indicate a positive and significant effect between alignment and agility
(β =.803; p ≤.001). Based on Cohen’s effect size, the results indicate a strong effect (f2 =1.22). The estimation
of agility explains 61.5% of the variance (R2 =.615). These results lead to acceptance of H2. Additionally, we
established mediation. First, we assessed the total indirect effect of EAM on agility, which is both positive
and significant (β =.384; p ≤.001). Second, we analyzed the direct effect of EAM on agility. This effect is not
significant (β =-037; p =.642). The results of these steps indicate indirect-only mediation, implying full
mediation (Zhao et al. 2010). We evaluated predictive relevance by assessing Stone-Geisser’s Q2.
Agility through EAM and the Role of Alignment
Americas Conference on Information Systems
7
Blindfolding is a sample reuse technique that can be used to calculate Q2 values for latent variables. We
executed the blindfolding procedure and calculated the Q2 values for EAM (Q2 =.496), alignment (Q2 =.363),
and agility (Q2 =.173). All values were greater than zero, thus indicate predictive relevance for endogenous
latent variables in our PLS path model (Hair et al. 2016, p. 202).
Post-Hoc Cluster Analysis
First, we performed a hierarchical cluster analysis using Ward’s method and included the squared
Euclidean distance as a similarity measure to determine the number of clusters (Hair et al. 2014, p. 456).
Assessing the agglomeration schedule showed the largest percentual increase at the step where two clusters
are integrated into one. However, two cluster solutions are seldom useful for meeting research objectives
(Hair et al. 2014, p. 461). Hence, we identified that the step from three to two clusters demonstrated the
second-highest percentual increase (24.3%). Thus, we chose a three-cluster solution as appropriate. Next,
we used the non-hierarchical k-means procedure to establish the final cluster solution (Hair et al. 2014, pp.
462–465). ANOVA showed significant differences (p ≤.001) in variable means within the clusters. Post-hoc
least-squares difference (LSD) analysis showed significant differences between all pairings within the
clusters (p ≤.001), suggesting discriminating observations in the cluster solutions (Hair et al. 2014, p. 465).
The final cluster solution, the standardized variable means, and ANOVA results are described in Table 2.
Cluster
1 (n=24)
2 (n=41)
3 (n=45)
Cluster
Variables
EA Documentation
-1.191
-0.237
0.851
EA Planning
-1.156
-0.031
0.644
EA Programming
-1.372
-0.120
0.841
EA Implementation
-1.337
0.076
0.644
EA Communication & Support
-1.385
-0.114
0.843
EA Governance
-1.221
-0.289
0.915
EA Stakeholder Participation
-1.263
-0.189
0.846
Mean of cluster variables
-1.275
-0.129
-0.798
External
Variables
Strategic IT Alignment
F (ANOVA) = 24.89**
-0.694
-0.303
0.646
Organizational Agility
F (ANOVA) = 9.792**
-0.612
-0.090
0.409
Significance levels ANOVA: ** p ≤.001
Table 2 - Results of Cluster Analysis
We will briefly describe the three clusters from the final solution. (1) The means of the EAM dimensions in
this cluster are smaller than the other two. Compared to other clusters, this cluster scores low on alignment
and agility. However, the mean of the EAM dimensions is small (-1.275) compared to alignment (-.694) and
agility (-.612). We further refer to this cluster as low EAM performers. (2) In this cluster, all means of the
EAM dimensions, alignment, and agility are higher than cluster 1 but lower than cluster 3. All means, except
EA implementation, are below zero, which is the sample average. Alignment shows a mean (-.303) that is
just a bit lower than the lowest-scoring dimension, EA Governance (-.289). The mean of agility (-.090) is
close to the overall sample mean of zero. We further refer to this cluster as medium EAM performers. (3)
This cluster demonstrated the highest means for the EAM dimensions, alignment, and agility. The mean of
alignment (.646) is just a bit smaller than the lowest-scoring EAM dimensions EA planning and EA
implementation (.644). The mean of agility (.409) is lower than the mean of alignment. We refer to this
group as high EAM performers.
Discussion
This research makes several contributions to the IT and IS management knowledge base. First, our PLS-
SEM results enrich the findings of earlier research that identified IT capabilities as a mediator between
EAM and agility (Pattij et al. 2019; Shanks et al. 2018; Van de Wetering 2019a). This outcome implies that
multiple mediators are involved in achieving agility through EAM. However, in line with previous research
we argue that EAM benefit realization could be complex and involves the interconnection of multiple
concepts (Niemi and Pekkola 2016; Schmidt and Buxmann 2011). Thus, our results might reveal just the tip
of the iceberg. Hence, we promote extensive explanatory research on the relation between EAM and agility
to gain a more comprehensive notion of EAM benefit realization. Second, we enrich previous research of
Agility through EAM and the Role of Alignment
Americas Conference on Information Systems
8
Bradley et al. (2012). The results of this research indicate a positive effect of EA maturity on agility mediated
by alignment. While maturity represents a state of an organization’s EA at a certain time, our results
emphasize the importance to have a continuous focus on EAM to develop EA. However, research on how
exactly EAM influences EA and its maturity is scarce (Niemi and Pekkola 2016). Hence, we suggest that
further research should be conducted to fill this gap. Third, we contribute to IS studies that have
investigated the paradoxical relationship between alignment and agility. As an answer to Tallon et al.’s
(2019) question, if organizations can be both agile and aligned at the same time, our cluster analysis results
suggest that this is the case for high EAM performers. In contrast to the findings of Liang et al. (2017), our
PLS-SEM results show that intellectual alignment facilitates agility. This outcome is consistent with the
results of Bradley et al. (2012) and Tallon and Pinsonneault (2011). Like Tallon et al. (2019), we argue that
other factors, such as organizational demographics, could explain contradictory results. In their discussion,
the potential effect of the differences between industry types and the extent of inertia and risk-taking that
characterizes them is articulated. Additionally, we argue that organizational characteristics like inertia and
risk appetite are the result of a more complex configuration of factors, such as demographics. Whereas our
results and those of Tallon and Pinsonneault (2011) are not limited to one specific industry, the research of
Liang et al. (2017) is focusing on the shipyard industry and that of Bradley et al. (2012) on healthcare.
Interestingly, not only the shipyard industry is suggested to be prone to inertia (Tallon et al. 2019) but also
healthcare as a result of institutionalized organizational behavior (Wang et al. 2015). Hence, further
research is required to understand inertia and its role in the relationship between alignment and agility.
Perhaps configurational approaches could support this research. Fourth, supported by our results, we
demonstrated the added value of the DCV to conceptualize EAM as dynamic capability for orchestrating the
continuous reconfiguration of an organization’s resources to adapt to its dynamic environment. Taking this
novel approach, we contribute to the academic EA and the DCV knowledgebase (Shanks et al. 2018; Van de
Wetering 2019a). Fifth, we contribute to the knowledgebase on the usage of complementary methods in
IT/IS research. While our PLS-SEM results suggest linear causality, this method does not reveal to what
extent non-linear behavior and equifinality play a role in EAM benefit realization. Hence, we performed a
post-hoc cluster analysis to investigate if such behavior is present. Our results revealed three clusters of our
EAM dimensions that were mainly distinguished by their performance scores. These clusters showcase
organizations with relatively high, medium, and low scores for alignment and agility. Given the differences
in alignment and agility performance scores, our cluster solution is not appropriate to identify equifinality.
In previous research, Aier et al. (2011) conducted a cluster analysis to identify configurations of EAM design.
In contrast to our clusters, their three cluster solutions show more variation in cluster variable means. Thus,
for example, one of their clusters contains a low mean for a cluster variable called ‘design impact’ variable
and a high mean for another variable called ‘IT operations support.’ Their ‘design impact’ variable shows
some similarities with our EA programming variable, and ‘IT operations support’ with our EA
communication & support variable. In their profiling efforts, like our results, the majority of EAM effect
means, such as alignment, significantly differ across the clusters. However, the extent of dissolution of
information silos is not significantly different across their clusters, suggesting equifinality. The results of
our study compared to that of Aier et al. indicate that there is no consolidated view on the importance of
non-linearity and equifinality in EAM benefit realization. Hence, including configuration approaches in
research designs is a valuable avenue for future exploratory research.
Our results are relevant for decision-makers as well. Developing and implementing EA is not a one-time
activity, but requires continuous adaption to guide the constant reconfiguration of an organization’s
resources (Abraham et al. 2012; Teece et al. 1997). Our study embraces a broader dynamic capability view
when it comes to EA deployment. Organizations should actively invest in the EAM practice to leverage its
potential to effectively and efficiently deploy and reconfigure EA assets and resources. Our findings justify
the investment of EAM efforts to achieve alignment and, ultimately, agility that firms need to achieve
competitive advantages in a dynamic environment (Winter et al. 2010). Thus, we provide some guidance as
well as key insights on how to achieve business value from EAM.
Some limitations constrain our outcomes. First, we used a quota sampling approach. Future research could
apply a stratified sampling strategy to improve external validity (Saunders et al. 2015). Second, repeating
our research with data collected from a broader geographical area data could improve the generalizability
of our results. Third, we used a single informant strategy. Collecting data for all constructs from a single
person could lead to CMB as a result of self-reporting bias. To minimize this risk, a matched-pair survey
where different respondents complete survey items (e.g., IT and business executives) could be useful.
Agility through EAM and the Role of Alignment
Americas Conference on Information Systems
9
REFERENCES
Abraham, R., Aier, S., and Winter, R. 2012. “Two Speeds of EAM—A Dynamic Capabilities Perspective,”
Enterprise Architecture Research and Practice-Driven and Practice-Driven Research on Enterprise
Transformation. (131:August 2014), pp. 111–128.
Aier, S., Gleichauf, B., and Winter, R. 2011. “Understanding Enterprise Architecture Management Design
– An Empirical Analysis,” in 10. Internationale Tagung Wirtschaftsinformatik (WI), pp. 645–654.
Barney, J. 1991. “Firm Resources and Sustained Competitive Advantage,” Journal of Management (17:1),
pp. 99–120.
Bradley, R. V., Pratt, R. M. E., Byrd, T. A., Outlay, C. N., and Wynn, D. E. 2012. “Enterprise Architecture,
IT Effectiveness and the Mediating Role of IT Alignment in US Hospitals,” Information Systems
Journal (22:2), pp. 97–127.
Chan, Y. E. 1997. “Business Strategic Orientation, Information Systems Strategic Orientation, and Strategic
Alignment.,” Information Systems Research, pp. 125–150.
Eisenhardt, K. M., and Martin, J. A. 2000. “Dynamic Capabilities: What Are They?,” Strategic Management
Journal (21), pp. 1105–1121.
Fiss, P. C. 2007. “A Set-Theoretic Approach to Organizational Configurations,” Academy of Management
Review (32:4), pp. 1180–1198.
Foorthuis, R., van Steenbergen, M., Brinkkemper, S., and Bruls, W. A. G. 2016. “A Theory Building Study
of Enterprise Architecture Practices and Benefits,” Information Systems Frontiers (18:3), pp. 541–
564.
Hair, J. F., Black, W. C., Babin, B. J., and Anderson, R. E. 2014. “Multivariate Data Analysis Seventh
Edition,” Pearson New International.
Hair, J. F., Hult, G. T. M., Ringle, C., and Sarstedt, M. 2016. “A Primer on Partial Least S quares Structural
Equation Modeling (PLS-SEM) 2nd Ed,” SAGE Publications.
Helfat, C. E., Finkelstein, S., Mitchell, W., Peteraf, M., Singh, H., Teece, D. J., and Winter, S. G. 2007.
“Dynamic Capabilities and Organizational Process,” Dynamic Capabilities: Understanding Strategic
Change in Organizations.
Helfat, C. E., and Winter, S. G. 2011. “Untangling Dynamic and Operational Capabilities: Strategy for the
(N)Ever-Changing World,” Strategic Management Journal (32:11), pp. 1243–1250.
IBM Corp. 2015. “IBM Corp. (2015). IBM SPSS Statistics for Windows, Version 24. Armonk, NY: IBM
Corp,” 2015.
Ketchen, D., and Shook, C. 1996. “The Application of Cluster Analysis in Strategic Management Research :
An Analysis and Critique,” Strategic Management Journal (17:6), pp. 441–458.
Kotusev, S., Singh, M., and Storey, I. 2015. “Consolidating Enterprise Architecture Management Research,”
Proceedings of the Annual Hawaii International Conference on System Sciences (2015-March:June
2016), pp. 4069–4078.
Lange, M., Mendling, J., and Recker, J. 2016. “An Empirical Analysis of the Factors and Measures of
Enterprise Architecture Management Success,” European Journal of Information Systems (25:5), pp.
411–431.
Lapalme, J., Gerber, A., Van Der Merwe, A., Zachman, J., Vries, M. De, and Hinkelmann, K. 2016.
“Exploring the Future of Enterprise Architecture: A Zachman Perspective,” Computers in Industry
(79), pp. 103–113.
Liang, H., Wang, N., Xue, Y., and Ge, S. 2017. “Unraveling the Alignment Paradox: How Does Business-IT
Alignment Shape Organizational Agility?,” Information Systems Research (28:4), pp. 863–879.
Luftman, J., and Brier, T. 1999. “Achieving and Sustaining Business-IT Alignment,” California
Management Review (1), pp. 109–122.
Lux, J., Riempp, G., and Urbach, N. 2010. “Understanding the Performance Impact of Enterprise
Architecture Management,” 16th Americas Conference on Information Systems 2010, AMCIS 2010
(4:August), pp. 3002–3011.
Niemi, E. I., and Pekkola, S. 2016. “Enterprise Architecture Benefit Realization: Review of the Models and
a Case Study of a Public Organization,” The DATA BASE for Advances in Information Systems (47:3),
pp. 55–80.
Pattij, M., Van de Wetering, R., and Kusters, R. 2019. “From Enterprise Architecture Management to
Agility through EAM and the Role of Alignment
Americas Conference on Information Systems
10
Organizational Agility: The Mediating Role of IT Capabilities,” in 32nd Bled EConference -
Humanizing Technology for a Sustainable Society, pp. 561–578.
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., and Podsakoff, N. P. 2003. “Common Method Biases in
Behavioral Research: A Critical Review of the Literature and Recommended Remedies,” Journal of
Applied Psychology, pp. 879–903.
Queiroz, M., Tallon, P. P., Sharma, R., and Coltman, T. 2018. “The Role of IT Application Orchestration
Capability in Improving Agility and Performance,” Journal of Strategic Information Systems (27:1),
Elsevier, pp. 4–21.
Reich, B. H., and Benbasat, I. 1996. “Measuring the Linkage between Business and Information Technology
Objectives,” MIS Quarterly: Management Information Systems (20:1), pp. 55–77.
Ringle, C. M., Wende, S., and Becker, J. M. 2015. “SmartPLS 3. Bönningstedt: SmartPLS.”
Sambamurthy, V., Bharadwaj, A., and Grover, V. 2003. “Shaping Agility through Digital Options:
Reconceptualizing the Role of Information Technology in Contemporary Firms,” MIS Quarterly
(27:2), pp. 237–263.
Saunders, M., Lewis, P., and Thornhill, A. 2015. Methoden En Technieken van Onderzoek, Amsterdam:
Pearson Benelux.
Schmidt, C., and Buxmann, P. 2011. “Outcomes and Success Factors of Enterprise IT Architecture
Management: Empirical Insight from the International Financial Services Industry,” European
Journal of Information Systems (20), pp. 168–185.
Shanks, G., Gloet, M., Asadi Someh, I., Frampton, K., and Tamm, T. 2018. “Achieving Benefits with
Enterprise Architecture,” Journal of Strategic Information Systems (27), pp. 139–156.
Tallon, P. P., and Pinsonneault, A. 2011. “Competing Perspectives on the Link Between Strategic
Information Technology Alignment and Organizational Agility: Insights from a Mediation Model,”
MIS Quarterly (35:2), pp. 463–486.
Tallon, P. P., Queiroz, M., Coltman, T., and Sharma, R. 2019. “Information Technology and the Search for
Organizational Agility: A Systematic Review with Future Research Possibilities,” Journal of Strategic
Information Systems (28:2), Elsevier, pp. 218–237.
Teece, D. J., Pisano, G., and Shuen, A. 1997. “Dynamic Capabilities and Strategic Management: Organizing
for Innovation and Growth,” Strategic Management Journal (18:7), pp. 509–533.
Teece, D., Peteraf, M., and Leih, S. 2016. Dynamic Capabilities and Organizational Agility: Risk,
Uncertainty and Strategy in the Innovation Economy, (58:4), pp. 13–36.
Venkatraman, N., Henderson, J. C., and Oldach, S. 1993. “Continuous Strategic Alignment: Exploiting
Information Technology Capabilities for Competitive Success,” European Management Journal
(11:2), pp. 139–149.
Wang, V., Lee, S. Y. D., and Maciejewski, M. L. 2015. “Inertia in Health Care Organizations: A Case Study
of Peritoneal Dialysis Services,” Health Care Management Review (40:3), pp. 203–213.
Van de Wetering, R. 2019a. “Enterprise Architecture Resources , Dynamic Capabilities , and Their Pathways
to Operational Value,” in International Conference on Information Systems (ICIS) 2019 Proceedings.
Van de Wetering, R. 2019b. “Dynamic Enterprise Architecture Capabilities: Conceptualization and
Validation.,” Business Information Systems, W. Abramowicz and R. Corchuelo (Eds.) (to appear).
Winter, K., Buckl, S., Matthes, F., and Schweda, C. 2010. “Investigating the State-of-the-Art in Enterprise
Architecture Management Method in Literature and Practice,” Proceedings of the Mediterranean
Conference on Information Systems, pp. 1–12.
Winter, S. G. 2003. “Understanding Dynamic Capabilities,” Strategic Management Journal (24:10 SPEC
ISS.), pp. 991–995.
Wu, S. P., Straub, D. W., and Liang, T.-P. 2015. “How Information Technology Governance Mechanisms
and Strategic Alignment Influence Organizational Performance: Insights From a Matched Survey of
Business and IT Managers,” MIS Quarterly (January 2016), p. 29.
Yeow, A., Soh, C., and Hansen, R. 2018. “Aligning with New Digital Strategy: A Dynamic Capabilities
Approach,” Journal of Strategic Information Systems (27:1), Elsevier, pp. 43–58.
Zachman, J. A. 1987. “A Framework for Information Systems Architecture,” IBM Systmes Journal (26:3),
pp. 454–470.
Zhao, X., Lynch, J. G., and Chen, Q. 2010. “Reconsidering Baron and Kenny: Myths and Truths about
Mediation Analysis,” Journal of Consumer Research (37:2), pp. 197–206.
