THE LOOSENESS OF LOOSE

COUPLING: THE USE AND MISUSE

OF “LOOSE COUPLING” IN HIGHER
EDUCATION RESEARCH
Mari Elken and Martina Vukasovic

ABSTRACT
The term “loose coupling” has been widely employed in higher education
research. Building partly on the “garbage can model” of decision-making, it
proposed an alternative to rational and linear views on organizing and gov-
erning, emphasizing instead ambiguity and complexity. The review of higher
education research literature presented in this chapter demonstrates that the
concept of loose coupling has frequently been used as a background concept,
often taken for-granted either as a point of departure for studies of organiza-
tional processes in higher education or as a diagnosis of the complexity of
higher education organization that inhibits implementation of reforms. This
chapter provides systematization and critical examination of how the term
“loose coupling”/“loosely coupled systems” has been employed in journal
articles focusing on higher education in the last 40 years. It presents a broad
mapping of 209 articles and a more detailed qualitative review of 22 articles,
which employed loose coupling as more than a background concept.
Keywords: Loose coupling; organization; higher education; loosely coupled
systems; differentiation; fragmentation

INTRODUCTION
The concept of organizations as loosely coupled systems is widely used and
diversely understood. While this statement effectively summarizes one of the key
points of this chapter, it is actually almost 30 years older. It is the opening

Theory and Method in Higher Education Research, Volume 5, 53
71

Copyright r 2019 by Emerald Publishing Limited
All rights of reproduction in any form reserved
ISSN: 2056-3752/doi:10.1108/S2056-375220190000005005
53
54 MARI ELKEN AND MARTINA VUKASOVIC

sentence of the 1990 article in which Orton and the “father” of loose coupling,
Weick (though the fatherhood of the concept is rather more complex), review
how the concept has been used since its original publication in 1976 (Orton &
Weick, 1990, p. 203). Weick’s 1976 article, “Educational Organizations as
Loosely Coupled Systems,” along with Clark’s “organisational saga” (1972),
Cohen, March & Olsen’s “garbage can decision-making model” (1972), and
Meyer & Rowan’s “myth and ceremony” (1977) constitutes a key set of 1970s
publications advancing widely used conceptual and theoretical insights based on
analysis of (higher) education organizations.
Not only has loose coupling become a staple ingredient of social science
research, it has also become a widely used and taken-for-granted description of
university organization. Given that the initial article marked its 40th anniversary
in 2016, it is timely to systematically examine how the concept has been used in
higher education research and what kinds of empirical insights have been devel-
oped regarding the loosely coupled nature of higher education.
Consequently, the chapter will:
• introduce the concept of loose coupling based on Weick (1976) and Orton
and Weick (1990);
• map how the idea of organizations as loosely coupled systems has been used
in higher education research, in particular with regard to breadth and depth;
• review articles that have foregrounded the concept; and
• discuss the relevance and possible future avenues for research concerning
loose coupling.

LOOSE COUPLING ACCORDING TO

WEICK (AND ORTON)
While this chapter takes a starting point in Weick’s conceptualization from
1976, it should first be noted that the actual terms “loose coupling” or “loosely
coupled” did not originate from Weick. In his 1976 article, Weick himself points
to at least three prior instances of use of these terms: (1) a 1973 article by
Glassman on persistence and loose coupling in living systems (Glassman, 1973);
(2) an unpublished manuscript by March and Olsen from 1975 titled, “Choice
situations in loosely coupled worlds”; and (3) another unpublished manuscript
by Salancik from 1975 titled, “Notes on loose coupling: linking intentions and
actions.” Moreover, while the 1976 article is attributed to Weick, it is in essence
a report of a National Institute of Education workshop organized in early 1975
in California, which gathered other organizational scholars, including March. In
a short note explaining the origins of the article (http://garfield.library.upenn.
edu/classics1989/A1989T566900001.pdf (accessed on January 29, 2019)), Weick
also implies that the report was published in Administrative Science Quarterly
effectively without peer review, apparently facilitated by the proximity of
Weick’s office to the office of Lodahl, then editor of ASQ. Thus, while Weick is
often cited as a primary reference for “loose coupling”/“loosely coupled” con-
cept, it seems that quite a number of people contributed to its creation. This
The Looseness of Loose Coupling 55

origin story of the 1976 article perhaps explains its content and structure, com-
prising: (1) a literature review; (2) a report of discussions on promises and pit-
falls of loose coupling; (3) a critical reflection on methodology; and (4) an
outline of a research agenda.
Second, the context in which this article was published is also of relevance.
Namely, it was a part of a countermovement within organizational research,
which at the time strongly assumed rationality. Together with the other three
articles highlighted in the introduction, as well as (among others) work by
March and Simon on ambiguity and bounded rationality, the key message was
that there is a “substantial unexplained remainder” of organizational dynamics
which is “intractable to analysis through rational assumptions” (p. 1). It argued
that it is not possible to answer some of the key questions
including what
holds organizations together
unless light is shed on previously taken-for-
granted aspects. In this respect, the concept of loose coupling is expected to act
as a “sensitizing device,” or a new (better) pair of glasses which allows the
researcher to see elements of organization that would otherwise remain in
the dark.
However, when it comes to how such a sensitizing device is to be defined, the
article is somewhat less clear. It refers to “loose coupling imagery,” as well as
“idea of loose coupling” and “coupling mechanisms”
technical and authority
based. It identifies no less than 15 different “connotations of loose coupling,”
some of which are rather abstract and applicable beyond the specific empirical
context of educational organizations
for example, loose-coupling between
means and ends, intentions and actions, organizational structure and organiza-
tional activity, lack of coordination, slow spread of influence/change through
the organization, planned unresponsiveness, and absence of regulations. Other
connotations remain rather close to the school setting that initiated the discus-
sion in the first place
for example, “curricula or courses in educational organi-
zations for which there are few prerequisites” (p. 5). Some of the connotations
as formulated by Weick appear to be conceptually rather close to each other;
compare, for example, “any one of several means will produce the same end”
and “no matter what you do things always come out the same” (p. 5). Some
other connotations appear to resonate with related concepts which originated
around the same time, for example, garbage-can decision-making which argues
that problems and solutions are not necessary causally linked (cf. Cohen et al.,
1972), or decoupling between “talk” and “walk” in organizations (cf. Meyer &
Rowan, 1977). In total, these connotations essentially present a multifaceted
image of coupling, going well beyond structural aspects of organizations.
Compared to the ambiguity concerning what exactly is (or is not) loosely
coupled in (educational) organizations, the article is much clearer with regard to
the promises and pitfalls of loose coupling. The basic premise here is that loose
coupling in general serves a positive function in an organization, but that it also
constitutes a liability. A loosely coupled organization is better at sensing the
complex environment, but runs the risk of “producing faddish responses” (p. 6).
At the same time, loose coupling allows for localized adaptation which does not
affect the whole system, and thus shields the whole organization from too quick
56 MARI ELKEN AND MARTINA VUKASOVIC

a response to negative signals from the environment, effectively contributing to

organizational perseverance. And while being able to “seal off” a problematic
part of an organization is good in that respect, it also poses obstacles for repair.
A loosely coupled organization is thought to be more creative with regard to
novel solutions but, given that change spreads slowly and weakly through the
organization, the benefits of such novel solutions may not arrive where and
when they are needed. Loosely coupled organizations are organizations charac-
terized by high professional autonomy, and as such it may be easier and cheaper
to coordinate them in a manner that resonates with these characteristics, but
that also comes with a “non-rational system of fund allocation” (p. 8) suggesting
that funding is, reportedly, not an effective incentive for change in such
organizations.
With regard to the research agenda, the 1976 article advocates for an inventory
of elements in (educational) organizations that could be (loosely or tightly) cou-
pled and for a mapping exercise of patterns of coupling, as well as functions and
dysfunctions that arise from such patterns. It stresses that such endeavors should
be based on context rich methodologies as well as longitudinal and comparative
studies, so that what was previously invisible can be seen. It also links to Weick’s
other work, arguing that more attention should be made on how people make
sense of their loosely coupled organizations (Weick, 1995). However, the research
agenda section actually starts with a suggestion that “more conceptual work has
to be done before other lines of inquiry on this topic are launched” (p. 16). This
effectively serves as an admission that the concept is ambiguous, in particular
given that it comes with the following warning: “if researchers start stalking the
elusive loosely coupled systems with imperfect language and concepts, they will
perpetuate the blandness of organizational theory” (p. 16).
Yet, it seems that the concept rapidly got a life of its own. In late 1980s,
Orton and Weick conducted a review of approximately 15 years of studies utiliz-
ing the concept of loose coupling/loosely coupled systems and found that “like a
linguistic Trojan horse, the loose coupling concept has preceded loose coupling
theory into various strongholds of organizational studies” (Orton & Weick,
1990, p. 203). They admit that the concept is “underspecified” and that its face
validity comes coupled (!) with significant imprecision, but argue that such
imprecision may have abetted its wide use. In an almost ironical manner, the
rather open conceptualization (or, loose definition) of loose coupling is likely
what also led to its success and spread.
Reflecting the initial context in which the 1976 article appeared, they reiterate
that the term “loosely coupled” should allow researchers to simultaneously
acknowledge (and see) rationality and order in an organization (hence “cou-
pled”), as well as indeterminacy and independence of its elements (hence
“loosely”). They also postulate two distinct interpretations of the concept. On
the one hand, there is a unidimensional notion of loose coupling
a scale rang-
ing from tightly to loosely coupled elements in a system. On the other hand,
there is a dialectical notion that is based on the extent to which elements are dis-
tinctive from each other, and responsive to signals external to them (which
includes signals from other parts of the organization). This yields a 2 × 2 matrix
The Looseness of Loose Coupling 57

(Table 1), which not only highlights that loosely coupled systems are those in
which both responsiveness and distinctiveness are present, but also links the idea
of coupling through one other often cited concept, that is, decoupling which ori-
ginates from Meyer and Rowan (1977).
Orton and Weick also classify the different studies they reviewed (reportedly
around 300 of them) into five distinct “voices.” Voice of causation comprises
studies focusing on why one finds loose coupling or not. The three identified
explanations include: (1) causal indeterminacy, which relates to ambiguity cir-
cumstances which give rise to garbage can decision-making approaches; (2) frag-
mented external environment which, among other matter, concerns decoupling
talk from walk; and (3) fragmented internal environment which highlights fluid-
ity of participation and dispersion of interest and involvement. Voice of typol-
ogy focuses on what can be (loosely) coupled: (1) individuals, (2) subunits, (3)
organizations, (4) hierarchical levels, (5) organizations and their environments,
(6) ideas, (7) activities, and (8) intentions and actions. These eight types at simi-
lar levels of abstraction constitute a leaner and sharper typology, compared to
the 15 connotations identified in 1976, some of which were rather context-
dependent. Voice of direct effects focuses on (supposedly) desirable effects of
loose coupling on organization as such, including increased modularity, variety,
and discretion. This is distinct from voice of organizational outcomes or indirect
effects, which highlights how specific loosely coupled organizational aspects
affect organizational performance, resistance to change, buffering of problem-
atic areas, adaptability, member satisfaction, and effectiveness. The voices of
effects (direct and indirect) correspond clearly to the discussion of promises and
pitfalls of loose coupling that Weick presented in 1976. Finally, the voice of
compensations highlights different approaches through which negative aspects
of loose coupling can be redressed, namely leadership, focused attention and
shared values. These five voices constitute building blocks of what Orton and
Weick label “a preliminary model of loose coupling theory”.
Orton and Weick admit that such a schematic representation of the five
voices oversimplifies relationships between and importance of different elements,
and does not necessarily correspond to how different researchers utilizing the
concept would see it. Nevertheless, Orton and Weick maintain that loose cou-
pling, specifically if understood dialectically and as more than a negative defini-
tion, allows for simultaneously grasping connectedness and autonomy (order

Table 1. A Dialectical Interpretation of System Coupling.

Distinctiveness

þ

Responsiveness þ Loosely coupled system Tightly coupled system

Decoupled system Non-coupled system/not a system

Source: Orton and Weick (1990).

58 MARI ELKEN AND MARTINA VUKASOVIC

and disorder) in organizations, thus being particularly well-suited for addressing

how organizations are socially constructed. They posit that the concept of loose
coupling will motivate researchers to consider structure as “something that orga-
nizations do, rather than merely as something they have,” suggesting that “loose
coupling may be able to do for the study of organizational structure what
bounded rationality did for the study of decision making” (Orton & Weick,
1990, p. 218).
In sum, in both the original article from 1976 and the re-conceptualization
from 1990, the idea of loose coupling is presented as relevant for different
dimensions (individuals, subunits, environment in general, other organizations,
etc.), and as having both beneficial and potentially harmful consequences for
organizations and their performance. Already in 1976 Weick stresses the impor-
tance of deeper engagement with the concept, while in 1990 Orton & Weick
order the different strands of research. Both of the articles have a conceptual
aim, but do not complement conceptual discussions with a specific empirical
examination of university organization.

LOOSE COUPLING IN HIGHER EDUCATION RESEARCH:

A BROAD PICTURE
Approach
In order to examine how the term, “loose coupling” has evolved over time, a lit-
erature review was conducted
with specific focus on journal articles. The
review draws inspiration from systematic review approaches (Tight, 2019), with
some adjustments. The starting point for selecting the studies to be analyzed was
Google Scholar citation records of the Weick 1976 article and the Orton & Weick
1990 article, from 1976 until the end of 2016. Within the set of over 11,000 cita-
tions, we selected articles that were published in: (a) journals specific to higher
education research, and (b) more general social sciences, provided that article
focused on higher education (the list of journals is provided in Appendix 1). This
yielded a total of 238 articles that were subsequently analyzed in terms of their
focus and use of the term, loose coupling. In the screening process, articles that
did not meet the inclusion criteria (e.g., they actually referred to other work by
Weick) were removed from the database. Three articles were removed due to
technical problems (e.g., corrupted files). Thus, the final database includes 209
articles. The screening of the articles focused on the following aspects:
• journal and year of publication;
• which of the loose coupling articles is cited: Orton & Weick (1990); Weick
(1976), or both;
• what is the nature of reference to loose coupling, distinguishing between the
use of loose coupling as (1) a background description (e.g., “universities are
loosely coupled” without any further engagement) or (2) a foreground con-
cept (discussed in more detail or part of the analytical framework); and
The Looseness of Loose Coupling 59

• whether loose coupling is described in more detail (the what) and whether
there is a discussion of antecedents and/or consequences (the how and why) of
loose coupling.
The articles were divided between the two co-authors equally, with an over-
lap of 25 articles that were screened by both co-authors, in order to ensure the
consistency of analysis.

The Screening Results

The results of such “broad picture” screening is presented in Table 2.
While it is expected that many articles analyzing loose coupling in higher
education are published in higher education journals, a bit more than one-fifth
of articles were published in general social science journals. This remains a trend
in recent years too, despite a proliferation of journals specializing on higher edu-
cation. Moreover, despite its age, the concept of loose coupling as conceptual-
ized in Weick (1976), and re-conceptualized in Orton and Weick (1990), seems
to maintain relevance for higher education researchers, with the highest number
of citations coming from the latest 10 years. Compared to the original contribu-
tion, the 1990 re-conceptualization of loose coupling by Orton and Weick has
received far less attention. Only two articles cite both contributions.
Most of the studies use loose coupling only as a background concept. Loose
coupling is a foreground concept in approximately one-fifth of the studies,
though in general social science journals, this proportion is somewhat higher.
Some of the studies provide detailed descriptions of the specific instance of loose

Table 2. Broad Overview of Use of Loose Coupling Concept.

Number of Articles Higher Education Specific General Social Science Total
Journals Journals

Total 162 47 209

Per decade 1977
1986 37 11 48
1987
1996 29 9 38
1997
2006 39 12 51
2007
2016 57 15 72
Citing Weick (1976) 156 45 201
Citing Orton and Weick (1990) 7 3 10
Loose Background 130 36 166
coupling concept
Foreground 32 11 43
concept
Descriptive (what) 23 5 28
Explanatory (how, 17 7 24
why)

Source: Authors.
60 MARI ELKEN AND MARTINA VUKASOVIC

coupling and, though somewhat less prominently, discuss the antecedents and
consequences of loose coupling.
Frequently, loose coupling is also associated with Meyer and Rowan’s decou-
pling argument, and in some instances, decoupling and loose coupling are used
interchangeably. Moreover, the garbage can decision-making model and view of
organizations as organized anarchies by Cohen, March and Olsen is used along
with loose coupling to describe enduring characteristics of universities, in several
instances even the same sentence, for example:
the general structure and design of institutions of higher education is much more adaptive and
restorative than are traditional bureaucracies and hierarchical systems. That is, they are loosely
coupled (Weick, 1976), fluid systems (Cohen & March, 1974) that have a great capacity to sur-
vive environmental disruptions. (Boffo, Dubois, & Moscati, 2008; Cameron & Whetten, 1983)

the situation can be summarised as follows: both in France and Italy the model of governance
which prevailed until recently can be defined as a mix of the political model (Cyert & March,
1963; March & Simon, 1958; Pfeffer, 1982) and the anarchy model and derived versions, one
of which might prevail over the others, according to times and specific situations. (Cohen &
March, 1974; Cohen et al., 1972; Weick, 1976). (Boffo et al., 2008, p. 15).

In general, a substantial share of the articles use loose coupling as a back-

ground concept and thus only superficially explore the concept itself. The refer-
ences to Weick, 1976 (and far less to Orton & Weick, 1990), can often be found
in introductions to describe specific characteristics of higher education organiza-
tions, without necessarily unpacking what these characteristics actually entail
and very rarely discussing the implications of this explicitly. This is somewhat
ironic given that Weick himself launched the notion of loosely coupled systems
precisely with a view that things which were before just assumed
at that time
rationality in organizations
will be questioned and that the sensitising device
of loose coupling will make previously invisible aspects of organization visible
(Weick, 1976, p. 3).
It is conceivable that preoccupation with rationalized, tidy, efficient, coordinated structures
has blinded many practitioners as well as researchers to some of the attractive and unexpected
properties of less rationalized and less tightly related cluster of events.

While such limited engagement with the concept may be an indication that
the messages from both cited articles
concerning conceptual clarity as well as
the need to empirically identify patterns of loose (or tight) coupling
have not
been taken aboard in the higher education research community, a small number
of articles foreground the concept to some degree.

LOOSE COUPLING IN HIGHER EDUCATION RESEARCH:

A MORE DETAILED PICTURE
Approach
In this section, we focus on the studies that foregrounded the concept (see
Table 2). Given that the notion of loose-coupling concerns intra-organizational
characteristics of higher education institutions, the studies that used it to reflect
The Looseness of Loose Coupling 61

on policy processes or university hospitals were not included, leading to a final

set of 22 articles that were the focus of the review.
We took Orton and Weick’s model of loose coupling theory and its five
voices as the starting point for classifying articles. This way we identified that
the large majority of the articles (p. 17) reflect the voice of typology and focus
on what is being loosely (or tightly) coupled to what. Specifically, these articles
address the relationship between structure and action, the differentiation of tasks
and authority, and horizontal and vertical fragmentation. A much smaller set of
studies (six; note that some of the articles appear under more than one heading)
focuses on different outcomes of loose coupling, thus reflecting voices of direct
effects, compensations and organizational outcomes, but not explicitly referring
to any of these three (hence grouped together). Similarly, five studies which
explore how loose coupling within higher education emerges reflect the voice of
causation, but they actually do not explicitly refer to causes. We also identified
five studies which provide some insights into methodological approaches for
studying loose coupling. Some studies
for example, Bleiklie, Enders, and
Lepori (2015), Kondakci and Van den Broeck (2009), Rhoades (1990)
belong
to more than one of these categories. In the following section, we present each
of these categories.

Types of Loose Coupling: Structure and Action

Articles in this group explore the coupling between: (a) formal structures or
principles, and (b) practices and activities within organization, often referring to
Weick’s loose coupling concept as well as the decoupling argument by Meyer
and Rowan (1977). For example, Townley (1997) refers to loose coupling as a
concept to describe ceremonial conformity, including adoption of formal
requirements while these remain decoupled from practices. Others, for example,
Rhoades (1990) see loose coupling and decoupling as related but different pro-
cesses, constituting one of the four analytical perspectives employed to analyze
change and continuity in higher education (the other three being resource depen-
dence, rational-bureaucracy, and organizations as political arenas).
The relationship between formal structures and actual practices is also the
focus of Barron’s analysis of the Berlin Principles on Ranking Higher Education
Institutions (2017), suggesting a clear decoupling with a clear reference to Orton
and Weick (1990). Similarly, Jarzabkowski and Seidl (2008) analyze the role of
meetings in the social practice of strategy, and the opportunities strategy pro-
cesses provide for emergence of loose coupling. They refer to both de-coupling
and re-coupling between durable organizational structures and temporary meet-
ing structures, albeit without explicitly discussing this in relation to loose cou-
pling. Blaschke, Frost, and Hattke (2014) take a different approach and
juxtapose macro loose coupling of organizations and tight coupling of the micro
patterns of communication, stressing how this view is in line with the “dialectical
nature of loose coupling” (Orton and Weick, 1990).
One of the few studies that explicitly constructs their whole analytical frame-
work around the concept loose coupling is Sapir and Oliver (2017) study of
62 MARI ELKEN AND MARTINA VUKASOVIC

introduction and framing of regulations. They specify that loose coupling exists
in situations in which (1) practices do not neatly match organizational structures
or (2) organizational responses do not clearly reflect external pressures. They
clearly contrast the notion of loose coupling to both decoupling and tight cou-
pling. In their view tight coupling concerns “full compatibility among external
requirements, organizational structures, and work activities,” while decoupling
is a specific strategy to maintain an image of compliance (p. 712). Similar to
Blaschke et al. (2014), they discuss the dialectical notion of loose coupling,
emphasizing simultanous processes of coupling and decoupling.
Overall, several of these studies suggest the relevance of adopting a proces-
sual view on coupling processes, and either explicitly or implicitly hint at the
dialectical view of coupling, where processes of decoupling and recoupling take
place on a continuous basis.

Types of Loose Coupling: Differentiation of Tasks and Authority

Studies that focus on the differentiation between different sub-systems within
higher education in particular address authority structures of academic and
administrative work.
In one of the earliest analysis of loose coupling, Telem (1981) distinguishes
between formal organizational structure and functional structure, suggesting
that the two do not necessarily overlap. At the same time, Childers (1981) ana-
lyzed decision-making in universities, specifically the political aspects of bureau-
cratic and collegial decision-making. While she argued that the loose coupling
concept (as well as the garbage can model) reflects the existing unstructured rela-
tionships between authority and tasks, structured relationships also existed, that
is, “bureaucratic strength does not necessarily decrease collegial strength”
(p. 41). A bit later, Lutz (1982) explored the basic tenet of four theories that had
emerged in the end of 1970s
loosely coupled systems, garbage can organiza-
tion, organized anarchies and contingency theory
all of which suggest that
flexible organizations are more adaptable and resilient. While his empirical
material was based on a set of limited personal observations, he noted that uni-
versities are rather bureaucratic in many aspects of their organization, thus also
characterized by inflexibility. Hence, he argues that universities would better be
explained by concurrent tight coupling and decoupling, in this manner also
emphasizing the dialectical characteristic of loose coupling.
Differentiation of tasks and authority within higher education has also been
explored more recently. Bleiklie et al. (2015) juxtapose “two alternative models
of control. That is hierarchical-bureaucratic and the loosely coupled one”
(p. 876), stressing that control in universities (as knowledge-intensive profes-
sional organizations) is, by necessity, a combination of both. They analyze
intra-organizational control in relation to two dimensions: (1) centralization of
power and (2) formalization of social relationships. In this context, loosely cou-
pled organizations are those organizations with low levels of centralization of
power and formalization of social relationships. Similarly, Thoenig and
Paradeise (2016) examine why some universities have more capacity to operate
The Looseness of Loose Coupling 63

as strategic actors, suggesting that coupling concerns interdependence, specifi-

cally interdependence between heterogenous academic units, and between the
academic and administrative sphere.
Overall, these articles seem to consistently emphasize the co-existence of tight
and loose coupling between different authority structures within universities, in
particular between academic and administrative sides. In terms of loose cou-
pling, this suggests that a fruitful area for future research would be to identify
the conditions and practices of coupling between these two subsystems.

Types of Loose Coupling: Horizontal and Vertical

Organizational Fragmentation
Rather than viewing loose coupling between structure, action or tasks, here
loose coupling is an organizational feature, being placed in the structural rather
than human side of organizations (Kezar & Eckel, 2004). Birnbaum (1989), for
example, highlights loose coupling between subunits of an organization, stres-
sing that while these subunits may have weak linkages between themselves, they
are characterized by strong linkages (tight coupling) internally (interestingly,
while Birnbaum does refer to Weick elsewhere in the article, he refers to work
by Cyert and March (1963) and Simon (1964) when discussing internal fragmen-
tation). Loose coupling within universities
here understood as organizational
fragmentation
is a starting point in several of the articles (Kondakci & Van
den Broeck, 2009; Simsek & Louis, 1994). Specifically, Bleiklie et al. (2015) see
loose coupling as low levels of decentralization of power and formalized rela-
tionships, while Thoenig and Paradeise (2016) consider loose coupling to com-
prise segmented nature of universities with heterogenuous subunits.
Change in terms of increasing or decreasing fragmentation also features
prominently. Ogawa’s (2002) study identified how reform processes in Japan
shifted universities from having virtually no linkages to what could be character-
ized as loosely coupled organizations. In contrast, Pilbeam (2006) identifies a
process of change from tight to loose coupling in UK in relation to increased
pressure for generating additional revenue streams. Specifically, he highlights
lack of “common language that permits shared understanding between disci-
plines,” “differences in objectives and goals” between departments and that
“one element can adapt and develop independently from the rest of the organi-
zation” (p. 308) as examples of loose coupling. While in the case of the UK, this
concerns weakening of horizontal linkages between academic subunits, Ogawa
(2002) stresses strengthening of vertical connections between graduate schools
and top administration.
While most of these studies focus on organizational units, loose coupling

in terms of horizontal linkages
has also been used studied on the individual
level. In study of how performance management had altered individual behav-
iour, Teelken (2015) examined policy, organizational as well as individual level,
contrasting loose coupling and tight coupling between individual academics.
In sum, articles that concern organizational fragmentation represent a diverse
set of arguments. While horizontal fragmentation is already a rather often used
64 MARI ELKEN AND MARTINA VUKASOVIC

example of loose coupling, vertical coupling, the relationship between horizontal

and vertical coupling, and relationships between organizational and individual
levels could be promising avenues for further research.

Outcomes of Loose Coupling

In one of the earliest articles reviewed here, Ecker (1979) argues that an impor-
tant aspect of universities being loosely coupled systems is that leaders need to
be comfortable with ambiguity and accept that organizational efficiency must
be a “subordinate goal” to organizational creativity. Ellström (1983) identified
organized anarchies, garbage can decision-making and loose coupling as three
distinct imageries of what he labelled as the anarchistic model of organization.
In his conceptualization, in the anarchistic model one can expect unclear orga-
nizational goals and ambiguous organizational processes, characterized by
foolishness, randomness and play. Both of these studies thus emphasize loose
coupling as a source for disorder and ambiguity within higher education
institutions.
Others have explored the consequences of loose coupling for a change pro-
cess. Simsek and Louis (1994) take as their starting point loosely coupled orga-
nizations, and explore how they respond to challenges of strategic planning.
They identify that both revolutionary and evolutionary changes take place
simultaneously. Kondakci and Van den Broeck (2009) also assume loose cou-
pling as a starting point, stressing that this leads to less interdependency,
increased specialization, reduced predictability and increased authority in sub-
units; thus enhancing improvization, self-reflection, and individual translation
of practices (p. 461).
Loose coupling has been also identified as one of the conditions for organiza-
tional adaptiveness. Specifically, Tahar, Niemeyer, and Boutellier (2011) pro-
pose organizational ambidexterity as a concept to reconcile organizational loose
coupling and centralized authority. Similarly, Kezar and Eckel (2004) argue that
loose coupling presents a number of benefits for organizations regarding their
responses to environment, leading to “greater organizational learning, flexibility,
and ability to respond to external conditions, leading to improved decision mak-
ing” (p. 381).
Overall, studies that discuss the outcomes of loose coupling refer to loose
coupling leading to increasing ambiguity, whereas this ambiguity also represents
a space for adaptability and flexibility. Yet, in very few of these studies are these
causal arguments explicitly analyzed.

The Emergence of Loose Coupling

Among the articles selected for review, very few explicitly concern why and how
loose coupling of universities emerges. Cameron (1984, p. 136) used “Janusian”
thinking as a means to describe the notion, where loose coupling is a means for
“organizations to cope better with unpredictable environmental events.” In this
manner, loose coupling is implicitly viewed as a means to manage
The Looseness of Loose Coupling 65

environmental demands. Townley (1997) expanded on this, arguing that the

environments in which institutions are embedded in are complex and consist of
multiple logics rather than one stream of pressures pushing toward conformity
and isomorphism. At the same time, as argued by Rhoades (1990), not all envir-
onments are equally relevant, and this variance would also lead to varying pat-
terns of loose coupling within higher education institutions.
While the previous studies emphasize institutional environments as a source
for loose coupling, others have suggested that loose coupling is a consequence of
multiple goals and objectives. For instance, Birnbaum (1989) in his model of
cybernetic higher education institutions, dedicates a whole section to discussing
loose coupling, and how universities address tensions between multiple and con-
flicting goals by “assigning responsibility for these goals to different subunits”
(p. 247). In this manner, loose coupling becomes a means to address multiple
tasks in organizations without necessarily determining who specifies these tasks.
Ogawa (2002) also analyzed the emergence of coupling, and traces it back to the
nature of the reform that introduced and proliferated graduate schools and
research centres.
It is obvious that the studies in this review do not present a full range of pos-
sible sources for loose coupling, and that this remains an area where additional
connections within literature can be made, e.g., toward discussions of disciplin-
ary differences.

Some Notes on Methodology

Most of the studies reviewed are qualitative case studies, often interview-based
and with a smaller number of comparative studies. Jarzabkowski and Seidl
(2008) also utilize observations of meetings and in this manner identify coupling
processes. Barron (2017) also relies on observations of ranking related confer-
ences, involvement in a university rating, media articles, and other documents
related to rankings. Bleiklie et al. (2015) analysis is an autoethnography in sev-
eral institutions.
Quantitative approaches are an exception. Childers (1981) employed a quan-
titative quasi-experimental design, using a survey instrument. Blaschke et al.
(2014) analyze micro foundations of leadership, governance and management in
universities by focusing on communication of strategic issues between university
governing bodies. Methodologically, to identify loose coupling between govern-
ing bodies and strategic issues, they identify which governing structures are men-
tioned in relation to which strategic issues. To identify tight coupling between
bodies and issues, they perform hierarchical cluster analysis to identify five sig-
nificant clusters.
This suggests that analysis of loose coupling is a subject to a variety of
methodological approaches. This is also not surprising, given the varied and
multiple (or sometimes lacking) operationalizations of loose coupling in existing
literature.
66 MARI ELKEN AND MARTINA VUKASOVIC

OUTLOOK FOR LOOSE COUPLING RESEARCH IN

HIGHER EDUCATION
It is clear that, despite the lack of explicit conceptual and empirical advance-
ment, loose coupling has become a classic in higher education research. From a
critical viewpoint, one could argue that being a classic could also mean that the
term obtains a more ritual purpose. Specific references can become a signaling
device for legitimating a claim that one is a part of a specific research commu-
nity (for ritual and other uses of “the classics,” see Stinchcombe, 1982). Such rit-
ual use is not necessarily problematic in itself. However, in this case, it seems to
be the predominant use and as such has implications for advancement of organi-
zational studies in higher education. A minority of 209 articles identified in the
broad review engage with the concept in a more encompassing way, by analyz-
ing coupling patterns, by describing what loose coupling actually entails or dis-
cussing its consequences for intra- or inter-organizational dynamics. The overall
impression in this analysis is that explicit conceptual engagement with the notion
of loose coupling has been reduced over time. The original article by Weick is
far more cited than the reconceptualization by Orton and Weick that aimed to
tighten up some of the ambiguities from the original article. The qualitative
review suggested that, while there are a small number of studies that explicitly
build their conceptual framework around loose coupling (Sapir & Oliver, 2017),
the concept is clearly defined and explicitly operationalized only in rare cases.
Loose coupling of higher education organizations seems to have become a
taken-for-granted empirical fact, given the extent to which it is used as a back-
ground concept to describe specific characteristics of higher education. It seems
to provide more of an “imagery” than a precise concept for empirical analysis.
In that sense, the use is perhaps ironically matching the ambiguity of the original
article, which also presented loose coupling as imagery with a range of connota-
tions. In higher education research, loose coupling seems to denote the discon-
nected, unstructured, informal and sometimes chaotic relationships in higher
education institutions. In this context, there is also a danger of loose coupling
becoming a normative notion that is expected to highlight the benefits for the
flexibility and adaptability of organizations. When associated with a temporal
argument, it can also become a means to describe a specific kind of the past,
which is viewed as more desirable than the more managerial and tightly coupled
university that is being facilitated by recent change processes. Voices of warning
against normative use of the concept were also expressed in early literature
(Lutz, 1982).
Recent studies that emphasize universities as more complete organizations
(Brunsson & Sahlin-Andersson, 2000; Seeber et al., 2015) or organizational
actors (Krücken & Meier, 2006) challenge this taken-for-granted view of endur-
ing loose coupling in universities. Given that both some of the earlier and later
studies that engage with the concept (see, for example, Bleiklie et al., 2015;
Childers, 1981; Ecker, 1979) pointed out that multiple couplings co-exist and
that there is a complex co-existence of academic and administrative steering, it
is timely to revisit what loose coupling in higher education really entails, both in
The Looseness of Loose Coupling 67

conceptual and empirical terms. Having this in mind, loose coupling is an issue
which remains empirically unresolved.
One way to address these issues is to shift focus from loose coupling to cou-
pling, where loose coupling is one possible form of coupling in higher education
institutions. This requires clarification vis-à-vis other adjacent concepts that
entail coupling
including tight coupling, decoupling and recoupling. One
possible solution is to revisit Orton and Weick’s dialectical conceptualization
that uses the dimensions of distinctiveness and responsiveness to delineate dif-
ferent patterns of coupling (see Table 1). Moreover, to unpack the dynamic
nature of coupling patterns, coupling should also be viewed as a specific orga-
nizational process rather than a property of an organization. This would
require analysis of the conditions under which specific patterns of coupling are
maintained (or change). In other words, the questions that should be asked
are: what kind of internal and external drivers stabilize specific coupling pat-
terns? What kind of drivers lead to loose coupling transforming into full orga-
nizational fragmentation (or even dissolution)? What kind of drivers tighten
organizational coupling? This implies that it would be relevant to examine hor-
izontal coupling as well as vertical coupling between levels of governance. This
processual focus requires attention to both the causes as well as the effects of
loose coupling.
In addition, it would be fruitful to consider whether and how multiple cou-
pling processes interact, given that recent analysis of organizations emphasizes
hybridity, with multiple organizational forms and practices side by side
(Greenwood, Raynard, Kodeih, Micelotta, & Lounsbury, 2011). If loose cou-
pling can be identified according to one dimension, it should not be taken as a
given that this is applicable for all dimensions. Instead, organizations can be
expected to sustain multiple coupling patterns. However, we need more empiri-
cal knowledge of the conditions under which such hybridity of couplings takes
place.
This kind of analysis requires considerable precision regarding the definitions
and operationalizations of coupling and the specific elements that are being
analyzed. Higher education research would benefit from going back to the
basics
revisiting what are the different elements that are coupled, empirically
identifying the patterns of coupling and theoretically accounting for specific
antecedents and consequences of such patterns. Unless this is done, we will
remain superficial in the analysis of how specific coupling patterns change. Such
opaqueness is also an important limitation for international comparisons of
organizational change processes in higher education. After all, merely calling
something “loosely coupled” is a rather generic description, and not very infor-
mative for analytical purposes.

ACKNOWLEDGEMENT
The authors would like to thank the participants of the ECPR 2017 conference
in Oslo for their comments on the initial version of this chapter. Mari Elken
thanks the participants of the NIFU writing group, in particular Taran Thune,
68 MARI ELKEN AND MARTINA VUKASOVIC

for their helpful feedback. Part of the work by Martina Vukasovic on this chap-
ter was conducted in the framework of the Odysseus project “Higher Education
Governance”, Research Council Flanders (FWO), grant number G.OC42.13N2.

REFERENCES
Barron, G. R. S. (2017). The Berlin principles on ranking higher education institutions: Limitations,
legitimacy, and value conflict. Higher Education, 73(2), 317
333. doi:10.1007/s10734-016-
0022-z.
Birnbaum, R. (1989). The cybernetic institution: Toward an integration of governance theories.
Higher Education, 18(2), 239
253.
Blaschke, S., Frost, J., & Hattke, F. (2014). Towards a micro foundation of leadership, governance,
and management in universities. Higher Education, 68(5), 711
732. doi:10.1007/s10734-014-
9740-2.
Bleiklie, I., Enders, J., & Lepori, B. (2015). Organizations as penetrated hierarchies: Environmental
pressures and control in professional organizations. Organization Studies, 36(7), 873
896.
doi:10.1177/0170840615571960.
Boffo, S., Dubois, P., & Moscati, R. (2008). Changes in university governance in France and in Italy.
Tertiary Education and Management, 14(1), 13
26. doi:10.1080/13583880701814132.
Brunsson, N., & Sahlin-Andersson, K. (2000). Constructing organizations: The example of public
sector reform. Organization Studies, 21(4), 721
746. doi:10.1177/0170840600214003.
Cameron, K. S. (1984). Organizational adaptation and higher education. The Journal of Higher
Education, 55(2), 122
144.
Cameron, K. S., & Whetten, D. A. (1983). Models of the organizational life cycle: Applications to
higher education. The Review of Higher Education, 6(4), 269
299.
Childers, M. E. (1981). What is political about bureaucratic-collegial decision-making? The Review of
Higher Education, 5(1), 25
47.
Clark, B. R. (1972). The organizational saga in higher education. Administrative Science Quarterly,
17(2), 178
184. doi:10.2307/2393952.
Cohen, M. D., March, J. G., & Olsen, J. P. (1972). A garbage can model of organizational choice.
Administrative Science Quarterly, 17(1), 1
25. doi:10.2307/2392088.
Cohen, M. D., & March, J. G. (1974). Leadership and ambiguity: The American college president.
New York, NY: McGraw-Hill.
Cyert, R. M., & March, J. G. (1963). A behavioral theory of the firm. Englewood Cliffs, NJ: Prentice-
Hall.
Ecker, G. (1979). Administration in higher education: Making the most of ambiguity. The Review of
Higher Education, 3(1), 23
31.
Ellström, P.-E. (1983). Four faces of educational organizations. Higher Education, 12(2), 231
241.
Glassman, R. B. (1973). Persistence and loose coupling in living systems. Behavioral Science, 18,
83
98.
Greenwood, R., Raynard, M., Kodeih, F., Micelotta, E. R., & Lounsbury, M. (2011). Institutional
complexity and organizational responses. The Academy of Management Annals, 5(1),
317
371.
Jarzabkowski, P., & Seidl, D. (2008). The role of meetings in the social practice of strategy.
Organization Studies, 29(11), 1391
1426. doi:10.1177/0170840608096388.
Kezar, A., & Eckel, P. (2004). Meeting today’s governance challenges. A synthesis of the literature
and examination of a future agenda for scholarship. Journal of Higher Education, 75(4),
371
399.
Kondakci, Y., & Van den Broeck, H. (2009). Institutional imperatives versus emergent dynamics: A
case study on continuous change in higher education. Higher Education, 58(4), 439
464.
doi:10.1007/s10734-009-9204-2.
Krücken, G., & Meier, F. (2006). Turning the university into an organizational actor. In G. S. Drori,
J. W. Meyer, & H. Hwang (Eds.), Globalization and organization: World society and organiza-
tional change (pp. 241
257). New York, NY: Oxford University Press.
The Looseness of Loose Coupling 69

Lutz, F. W. (1982). Tightening up loose coupling in organizations of higher education. Administrative

Science Quarterly, 27(4), 653
669.
March, J. G., & Simon, H. A. (1958). Organizations. New York, NY: Wiley.
Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and cer-
emony. American Journal of Sociology, 83(2), 340
363. doi:10.2307/2778293.
Ogawa, Y. (2002). Challenging the traditional organization of Japanese universities. Higher
Education, 43(1), 85
108. doi:10.1023/a:1012928416523.
Orton, J. D., & Weick, K. E. (1990). Loosely coupled systems: A reconceptualization. Academy of
Management Review, 15(2), 203
223. doi:10.5465/amr.1990.4308154.
Pfeffer, J. (1982). Organizations and organization theory. Boston, MA: Pitman.
Pilbeam, C. (2006). Generating additional revenue streams in UK universities: An analysis
of variation between disciplines and institutions. Journal of Higher Education Policy
and Management, 28(3), 297
311. doi:10.1080/13600800600984363.
Rhoades, G. (1990). Change in an unanchored enterprise: Colleges of education. The Review of
Higher Education, 13(2), 187
214.
Sapir, A., & Oliver, A. (2017). Loose coupling, conflict, and resistance: The case of IPR policy con-
flict in an Israeli university. Higher Education, 73(5), 709
724.
Seeber, M., Lepori, B., Montauti, M., Enders, J., de Boer, H., Weyer, E., … Reale, E. (2015).
European universities as complete organizations? Understanding identity, hierarchy
and rationality in public organizations. Public Management Review, 17(10), 1444
1474.
doi:10.1080/14719037.2014.943268.
Simon, H. A. (1964). On the concept of organizational goal. Administrative Science Quarterly, 9(1),
1
22. doi:10.2307/2391519.
Simsek, H., & Louis, K. S. (1994). Organizational change as paradigm shift: Analysis of the change
process in a large, public university. The Journal of Higher Education, 65(6), 670
695.
doi:10.2307/2943824.
Stinchcombe, A. L. (1982). Should sociologists forget their mothers and fathers. The American
Sociologist, 17(1), 2
11.
Tahar, S., Niemeyer, C., & Boutellier, R. (2011). Transferral of business management concepts to uni-
versities as ambidextrous organisations. Tertiary Education and Management, 17(4), 289
308.
Teelken, C. (2015). Hybridity, coping mechanisms, and academic performance management:
Comparing three countries. Public Administration, 93(2), 307
323.
Telem, M. (1981). The institution of higher education—A functional perspective. Higher Education,
10(5), 581
596.
Thoenig, J.-C., & Paradeise, C. (2016). Strategic capacity and organisational capabilities: A challenge
for universities. Minerva, 54(3), 293
324. doi:10.1007/s11024-016-9297-6.
Tight, M. (2019). Systematic reviews and meta-analyses of higher education research. European
Journal of Higher Education, 9(2), 133
152. doi:10.1080/21568235.2018.1541752.
Townley, B. (1997). The institutional logic of performance appraisal. Organization Studies, 18(2),
261
285.
Weick, K. E. (1976). Educational organizations as loosely coupled systems. Administrative Science
Quarterly, 21(1), 1
19.
Weick, K. E. (1995). Sensemaking in organizations. Thousand Oaks, CA: Sage.
70 MARI ELKEN AND MARTINA VUKASOVIC

APPENDIX 1: JOURNALS EXAMINED FOR THE REVIEW

Journal specializing on higher education

• Assessment and Evaluation in Higher Education

• European Journal of Higher Education
• Higher Education
• Higher Education Policy
• Higher Education Quarterly
• Higher Education Research and Development
• Journal of Higher Education
• Journal of Higher Education Policy and Management
• Minerva
• Quality in Higher Education
• Research in Higher Education
• Review of Higher Education
• Studies in Higher Education
• Tertiary Education and Management

General social sciences journals, only articles focusing on higher education

• American Journal of Sociology

• American Sociological Review
• Academy of Management Journal
• Academy of Management Review
• Annual Review of Sociology
• Administrative Sciences Quarterly
• Educational Administration Quarterly
• European Journal of Education
• Governance
• Journal of Education Policy
• Journal of European Integration
• Journal of European Public Policy
• Journal of Management
• Journal of Management Inquiry
• Journal of Management Studies
• Journal of Organizational Behavior
• Journal of Public Administration Research and Theory
• Journal of Policy Management and Analysis
• Management Science
• Organization
• Organization Science
• Organization Studies
• Policy & Politics
• Policy Sciences
• Policy Studies Journal
• Public Administration
The Looseness of Loose Coupling 71

• Public Administration Review

• Public Management Review
• Public Organization Review
• Regulation and Governance
• Research Policy
• Science and Public Policy
• Strategic Management Journal
• Strategic Organization