Practical recommendations based on institutional theory for the
implementation of technology:
A Systematic Literature Review
Master Thesis BA: Change Management
Faculty of Economics and Business
University of Groningen
Supervisor: dr. C. Reezigt
Secondary Evaluator: M. L. Hage
November 28, 2016
L.E. de Vries
S1517481
Word count: 9041
2
Abstract
3
Table of Contents
Introduction ... 4
Theoretical framework ... 5
The institutional environment ... 5
Methodology ... 8
Planning the Systematic Literature Review ... 8
Conducting the Systematic Literature Review ... 9
Reporting the Systematic Literature Review ... 11
Summary ... 11
Results ... 12
Search Results ... 12
Overview of the Selected Articles ... 13
Findings ... 16
Know the institutional environment ... 16
Have an adequate infrastructure ... 16
Create alignment ... 18
The use of isomorphic pressures ... 19
Summary ... 20
Discussion ... 21
Know the institutional environment ... 21
Have an adequate infrastructure ... 23
Create alignment ... 24
The use of isomorphic pressures ... 24
How do the themes relate? ... 25
Further research directions ... 25
Research limitations ... 25
Conclusion ... 27
References ... 28
Appendices ... 36
Appendix A – Search Query ... 36
Appendix B – Data Extraction Form ... 38
Appendix C – Overview of excluded articles ... 39
Appendix D – Quality assessment form and scores ... 91
4
Introduction
Halloween is supposed to be the “sweetest and scariest” day of the year in the USA. Hershey Foods, a
Fortune 500 company famous for their delicious chocolate kisses, usually delivers around $100
million worth of those kisses to the people who celebrate Halloween. The year 1999 however was not
a “sweet” experience at all for Hershey. In 1996 they started implementing a new technology, an
upgrade of their IT systems. 48 months was recommended for the implementation, but Hershey
decided to use 30 months to roll out the new system. The implementation failed, which led to
problems with the supply chain applications. Hershey was unable to deliver the $100 million worth of
Kisses for Halloween, although they had most of the inventory in stock. It caused an 8 percent decline
in stock price and a 19 percent drop in quarterly profit. The implementation of the technology did not
take Hershey down, but 1999 was certainly not the “sweetest” year (Wailgum, 2009).
Nowadays, many organizations implement technology to be more efficient and competitive
(Ahmad, 2014). Furthermore, competition and a demanding environment ask for constant
improvement of technology to stay ahead (Nielsen, Mathiassen & Newell, 2014). Moreover,
technologies can be very complex (Narayanan & Chen, 2012) and many attempts to implement
technology fail because it is difficult in practice (Hitt, Wu, & Zhou, 2002; Kukafka, Johnson, Linfante,
& Allegrante, 2003).
Institutional theory might provide answers to why implementation of technology often fails
(DiMaggio and Powell, 1983; Zucker, 1987; Greenwood and Hinings, 1996). Institutional theory
states that structural and behavioural changes in organizations are driven by the need for
organizational legitimacy instead of competition or a desire for efficiency. This in turn, stimulates
processes of institutionalization which makes organizations more alike, and possibly not more
efficient (DiMaggio and Powell, 1983). This process is called ‘isomorphism’, which is defined as a
“constraining process that forces one unit in a population to resemble other units that face the same set
of environmental conditions” (DiMaggio and Powell, 1983, 149). It is the drive for legitimacy and
becoming more alike which might pose problems when an organization tries to implement technology.
Institutional theory assists in mapping the institutional environment and understand the
possibilities and restrains where such change initiatives take place (Kuipers et al., 2014). Change
practitioners benefit from this knowledge when implementing technology. Furthermore, practical
recommendations on how to implement technology in an institutional environment might decrease the
amount of failing technology initiatives.
Although, a substantial body of academic research exists on the adoption and implementation
of technology, a review of the practical recommendations regarding institutional theory given by this
body of research has yet to be done. Therefore, this study aims to fill in this gap.
5
design, structure, and work of an organization (Ahmed, 2014). These technologies can range from
something simple, such as buying a laptop with accounting software, to something complex, such as
inter-organizational information technology. Due to the different effects and uses of technology, it is
difficult to form a definition. In this study, technology is defined as software or system an organization
implements to increase their efficiency, performance, and competitiveness. This definition is able to
capture the technologies studied in academic literature.
Next, is the theoretical framework of this study to broaden the understanding and knowledge
of the reader regarding the topic of this study.
Theoretical framework
Before implementation, an organization needs to make decisions about the adoption of certain
hardware and/or software technology (Thong, 1999) and on a managerial and professional/technical
staff level in the internal and external environment of an organization (Grover and Goslar 1993;
Preece 1995). Institutional theory provides decisionmakers with information about the institutional
environment and its pressures surrounding an organization.
The institutional environment
The focus of institutional theory is on “the institutional environment surrounding an
organization” (Phang et al., 2008, p. 103). Within this environment are institutional dimensions that
influence individual as well as organizational behaviour (Scott, 2001). These institutional forces are
categorized as the regulatory, normative, or cognitive dimension (Scott, 2001). The regulatory
dimension refers to formal rules and regulations to which an organization needs to comply. The
normative dimension refers to informal values and norms that promote certain behaviour. The
cognitive dimension refers to rules and meanings that are culturally constructed (Bruton, Ahlstrom &
Han-Lin, 2010). From these institutional dimensions three types of pressures emerge i.e., coercive,
normative, and mimetic pressures (DiMaggio and Powell, 1983).
Coercive pressure refers to pressures by others on whom an organization is dependent
(DiMaggio et al., 1983). Such coercive pressures might come from suppliers who controle scarce
resources, customers who are responsible for most of the organization’s turnover, or parent companies
driving changes to organizational structure of its subsidiaries (Messerschmidt and Hinz, 2013).
Normative pressure refers to pressures that are derived from the professionalization of fields
and disciplines (DiMaggio et al., 1983). These pressures might come from interaction between
organizations who operate in the same industry, and these successful organizations might persuade
non-adopters to adopt new technology through promotion of the benefits and costs of the new
technology (Powell, 1991).
6
organization over time, in response to changes in structure and actions of organizations in the same
industry (Messerschmidt et al., 2013). Furthermore, changes in other organizations create uncertainty
and triggers mimetic behaviour from organizations that do not want to fall behind (Messerschmidt et
al., 2013).
These pressures within an institutional environment foster to a process called ‘isomorphism’
(DiMaggio et al., 1983). Isomorphism is defined as a “constraining process that forces one unit in a
population to resemble other units that face the same set of environmental conditions” (DiMaggio et
al., 1983, 149). In other words, organizations within the institutional environment will start to look the
same. Coercive isomorphism takes place when organizations try to avoid sanctions from others on
whom they are dependend. Normative isomorphism occurs when an organization tries to conform to
pressures related to professionalization. Lastly, mimetic isomorphism takes place when organizations
face uncertainty and try to model themselves to organizations that they perceive as successful or
legitimate (Souitaris, Zerbinati, and Liu, 2012). An organization’s actions are legitimate when these
actions are desirable (Suchman, 1995).
Thus, an institutional environment is a complex setting that is influenced by institutional
dimensions and pressures. This complexity makes implementation of, for example, information
technologies difficult (Philippidou, Karageorgiou, Tarantilis, Soderquist, & Prastacos, 2008).
Knowing and understanding these forces make it possible to guide implementation of technology more
precisely. Furthermore, mapping the external and internal processes benefits implementation, because
external processes affect diffusion into an organization and internal processes affect creation from
within (Scott, 2001).
According to Burns and Wholey (1993), the institutional environment influences the structure
and actions of an organization significantly. Additionally, Teo, Wei, and Benbasat (2003) argue that
adoption of technology is not by basing it on intra-organizational or technical criteria, but on the
institutional environment and its pressures. For example, organizations might try to replicate the
success of another organization to be as innovative in their environment (Markus, 1987).
Unfortunately, ‘IT researchers have yet to ask how institutions influence the design, use, and
consequences of technologies, either within or across organizations’ (Orlikowski and Barley, 2001, p.
153). It is likely, that the institutional environment influences organizations in their choice and design
of technology. But often, technology is treated as a unitary object, ignoring its unique properties
(Volkoff, Strong & Elmes, 2007). This might be another reason why implementation of technology
often fails. Again, this shows the importance of recommendations on implementation of technology.
Therefore, this study will answer the following research question:
7
This study contributes to the current knowledge of institutional theory and technology
implementation. Regarding institutional theory, this study expands the knowledge by giving insight in
recommendations found in the academic literature. This study also broadens the knowledge of the
field of technology implementation regarding the aspect of institutional theory.
The main aim of this study is to provide practitioners with, on institutional theory based,
practical recommendations for implementation of technology in organizations. The method for
gathering these practical recommendations from the academic literature is a systematic literature
review.
8
Methodology
To find key scientific contributions to the field of change management in relation with
technology and institutions a Systematic Literature Review (SLR) is used. According to Leseure,
Bauer, Birdi, Neely, and Denyer (2004), a SLR should be replicable, parsimonious, and competent. In
other words, a literature search using the same databases and search strings should result in the same
list of found articles. Secondly, the SLR should be written in a clear and concise way. And lastly,
enough evidence needs to be brought forward by the search in order to create valuable insights. To
adhere to these three criteria, three phases described by Kitchenham (2007) are used: 1) planning the
review; 2) conducting the review; and 3) reporting the review. Each phase is build-up of several stages
that guide the process of the SLR. The first stage is concerned with setting up the SLR, the second
phase with gathering and processing relevant information, and the third phase with describing what the
gathered and processed information tells us. This process seems linear, but Lavallee, Robillard, and
Mirsalari (2014) advice to approach a SLR in an iterative way. Especially, if the SLR is performed by
a person who is not an expert on the subject. By describing each stage and its phases comprehensively,
repeatability of the SLR is increased.
Planning the Systematic Literature Review
Identifying major terms and keywords. To find articles for a systematic literature review,
relevant search terms and keywords need to be identified. To identify these search terms and keywords
a strategy developed by Khan, Basri, Dominic, and Amin (2013) is used. Their strategy consists of the
following four steps: 1) derive major terms from the research question; 2) search for alternative
spellings and synonyms; 3) verify the keywords in any relevant papers; and 4) use the Boolean “OR”
operator to integrate alternative spellings and synonyms and use the “AND” operator to integrate the
major terms.
9
were adapted to the search engine used by the databases.
Databases. The databases used in this study are “Academic Search Premier”, Business Source
Premier”, and “Elsevier ScienceDirect”. These databases were selected because the majority of articles
related to the topic are expected to be published in journals that are represented by these databases.
Both “Academic Search Premier” and “Business Source Premier” use the same search engine.
To search for conjugations in this search engine an ‘*’ is placed at the end of the search term. For
example, ‘theor*’ will find results containing ‘theories’, ‘theory’, etc. Furthermore, by using “ “
around a search term the search engine will search for the complete expression. For example,
“organizational change” will give articles containing ‘organizational change’ and not organizational
and change separately.
“Elsevier ScienceDirect” on the other hand, uses a different search engine where using the
search term ‘theory’ will automatically result in plurals. In appendix A you will find two sets of
search strings where the first is used in the databases “Academic Search Premier” and “Business
Source Premier”, and the second set is used in the database “Elsevier ScienceDirect”.
Conducting the Systematic Literature Review
Search Strategies. In the database “Academic Search Premier”, the advanced search option
was used. After each field search, the articles were filtered on publish date (2000+), peer reviewed
journals, academic journals, and English language. For Academic Search Premier the advanced search
option was selected and for each search string the field was set to all text. Furthermore, the articles
were filtered for publish date 2000 and further, peer reviewed academic journals, and written in
English.
In the database “Business Source Premier”, the advanced search option was used. After each
field search, the articles were filtered on publish date (2000+), peer reviewed journals, academic
journals, and English language. For Business Source Premier the advanced search option was selected
and for each search string the field was set to all text. Furthermore, the articles were filtered for
publish date 2000 and further, peer reviewed academic journals, and written in English.
In the database “Elsevier ScienceDirect”, the same search strings were used, but adapted to fit
the Elsevier ScienceDirect search engine (see appendix A). The ‘Expert search’ function was used to
search in all text of the articles. The search strings were inserted in the text block and restricted by
searching for articles (document type), articles that are published since 1999, and articles in press.
Each search result was exported to the computer in RIS format, and then imported in
Mendeley Desktop. After import, a mechanical and manual search was performed among all articles to
find and remove duplicates.
10
The databases “Academic Search Premier” and “Business Source Premier” have the option to
apply filters to the articles. The following filters were applied: 1) peer reviewed article; 2) articles
published after the year 1999; and 3) articles written in English. The “Elsevier Sciencedirect” database
also gave the possibility to apply filters, but compared to the other two databases choices were limited.
The filter for year was set to articles that are published after the year 1999. Filtering for ‘peer
reviewed’ articles was not possible in “Elsevier Sciencedirect”,
The exclusion criteria ‘describing an empirical study’, was performed manually on all
remaining articles, together with the first inclusion criteria to speed up the process. Hereafter, the
second inclusion criteria was applied followed by the third inclusion criteria. The exclusion criteria
“before the year 2000” was applied because no articles were found when applying a search filter on
the years before 2000.
Table 1. Inclusion and Exclusion criteria
Inclusion Criteria
1 the recommendations are based on institutional theory
2 the article is based on a situation of change and related to technology 3 the article gives practical recommendations to the field of change management
Exclusion Criteria
1 the article is not describing an empirical study 2 the article is not peer reviewed
3 the article is published before the year 2000 4 the article is not written in English
11
criteria are not applicable and scored with ‘N/A’. This influences the total possible score, and thus the
total possible score needs to be adjusted. To adjust the total possible score, the formula ‘total possible
score = 28 – (number of ‘N/A’ x 2)’ is used.
Reporting the Systematic Literature Review
Data analysis and synthesis. After the quality assessment, the selected articles were
documented using a data extraction form (see appendix B). For the analysis and synthesis of the
recommendations, three steps proposed by Thomas and Harden (2008) was used. In the first step, the
findings were coded based on meaning and context. Secondly, the coded findings were structured
based on meaning and context into descriptive themes using axial coding. In step three, the analytical
themes were developed. According to Lavallee et al. (2014), synthesis benefits from an iterative
approach. They state that structuring the large amount of data is difficult, which makes revisiting
papers necessary. Therefore, for the development of the descriptive and analytical themes an iterative
approach was used to improve the quality of the synthesis.
Summary
12
Results
Search Results
The result of the selection process is presented in figure 1. The databases Academic Search
Premier, Business Source Premier, and Elsevier ScienceDirect were consulted on August 3 2016. The
search resulted in 215 articles in Academic Search Premier, 354 articles in Business Source Premier,
and 118 articles in Elsevier ScienceDirect. In total 687 articles were found using the search strings in
the selected databases. A mechanical check for duplicate articles resulted in 42 duplicates and a
manual check found none, meaning that 645 articles remained. Applying the inclusion and exclusion
criteria resulted in removing 607 articles, meaning that 10 articles met the criteria. Appendix C shows
the excluded articles including their reason for exclusion.
Figure 1. Flowchart of article selection process
Business Source Premier: 354 Potentially relevant articles Academic Search Premier: 215 Potentially relevant articles Elsevier ScienceDirect: 118 Potentially relevant articles Total of 687 potentially relevant articles
Total of 645 articles for
screening
Total of 10 articles meet