You are on page 1of 20

2

The Normal Science of Structural


Contingency Theory

L EX D O N A L D S O N

Within organization studies, contingency theory which decision-making authority is concentrated


has provided a coherent paradigm for the at the top of the hierarchy, whereas a large
analysis of the structure of organizations. The organization, one that has many employees, is
paradigm has constituted a framework in which optimally structured by a decentralized structure
research progressed leading to the construction in which decision-making authority is dispersed
of a scientific body of knowledge. The task of down to lower levels of the hierarchy (Child
this chapter is to outline the contingency theory 1 973; Pugh et al. 1 969). There are several con­
of organizational structure and show how tingency factors: strategy, size, task uncertainty
research within this paradigm has proceeded in and technology. These are characteristics of the
a normal science mode. organization. However, these organizational
The recurrent set of relationships between characteristics in turn reflect the influence of
organizational members can be considered to be the environment in which the organization is
the structure of the organization. This includes, located. Thus, in order to be effective, the
but is not restricted to the authority relation­ organization needs to fit its structure to the
ships, the reporting relationships as signified in contingency factors of the organization and thus
the organization chart, the behaviours required to the environment. Hence the organization is
by organizational rules, the patterns in decision­ seen as adapting to its environment.
making such as decentralization, patterns of Each of the different aspects of the organiza­
communication and other behaviour patterns. It tional structure is contingent upon one or more
embraces both the officially prescribed formal of the contingency factors. Thus the task of
organization and the de jacto, unofficial, contingency research is to identify the particular
informal organization (Pennings 1 992). There is contingency factor or factors to which each
no definition of organizational structure that particular aspect of organizational structure
tightly circumscribes its subject matter a priori; needs to fit. This involves the construction of
rather the research projects each look at various, theoretical models of fits between contingency
different aspects of organizational structure and structural factors and their testing against
without claiming their focus to be exhaustive. empirical data. The empirical data usually
Contingency theory states that there is no consist of data comparing different organiza­
single organizational structure that is highly tions as to their contingencies and structures.
effective for all organizations. It sees the The contingency theory of organizational struc­
structure that is optimal as varying according ture will be termed here 'structural contingency
to certain factors such as organizational strategy theory' (Pfeffer 1 982).
or size. Thus the optimal structure is contingent Kuhn ( 1 970) argues that scientific research
upon these factors which are termed the proceeds within the framework of a paradigm,
contingency factors. For example, a small-sized which specifies the core theoretical ideas, the
organization, one that has few employees, is assumptions, language, method and conven­
optimally structured by a centralized structure in tions. The growth of a body of knowledge is
52 STUD YING ORGANIZA TION

marked by paradigm revolutions, when one work organization emerged from the interplay of
paradigm is overthrown and replaced by group dynamics (Roethlisberger and Dickson
another. Such discontinuous changes are radical 1 939). This would enable managers to adopt a
and infrequent. Most of the time science more considerate approach that would elicit
proceeds in a normal science phase guided by employee cooperation. The focus here was on
the ruling paradigm. In such a phase research the bottom-up processes of organizing and the
works on problems within the body of work, benefits of participation in decision-making by
such as resolving anomalies, while leaving the employees from lower levels of the hierarchy
paradigm itself unquestioned. (Likert 1 9 6 1 ) . There were attempts to bring
The study of organizational structure wit­ together these two antithetical approaches of
nessed a paradigm change when the classical classical management and human relations by
management school was overthrown by the new arguing that each approach had its place. Thus
paradigm of contingency theory, as will be seen contingency theories developed in the 1 950s and
below. This inaugurated an era of normal science 1 960s on topics such as small-group decision­
research within the contingency paradigm. making and leadership (see Vroom and Yetton
However, other paradigms arose subsequently 1 973). Around the end of the 1 9 50s scholars
that sharply question the contingency paradigm began to apply this contingency idea to
(Scott 1 992). Thus the study of organizational organizational structures.
structure is presently pluralistic with conflict The key idea in the small group literature was
between paradigms and normal science within that group problem-solving was accomplished
the paradigms (Aldrich 1 992; Donaldson 1 985a, effectively in a centralized structure when the
1 995a; Pfeffer 1 993). Since other chapters in this task was relatively certain but required a less
Handbook deal with other paradigms we will centralized and more richly joined structure
here concentrate on the contingency paradigm. where the task was uncertain in order to generate
The normal science that has been pursued within and communicate the larger amount of knowl­
the contingency paradigm is probably the largest edge and communications needed (Pennings
single normal science research stream in the 1 992: 276). Applied to whole organization
study of organizational structure to date. Thus structures this is equivalent to a hierarchy
in discussing the contingency paradigm there is a which centralizes expertise, communications
considerable volume of normal science research and control for tasks low in uncertainty and a
to report. Hence the concept of normal science in flexible, participatory team network for tasks
organizational studies is quite well illustrated by which are high on uncertainty. A major way to
the work within the contingency theory of have a low uncertainty task is to do the same
organizational structure (see also Donaldson thing repeatedly by avoiding innovation. Thus
1 996). innovation becomes a major underlying con­
tingency factor of the task uncertainty con­
tingency. Increasing scale can lead to low task
uncertainty, in that scale often involves repeti­
ORIGINS OF STRUCTURAL
tion, such as mass production.
CONTINGENCY THEORY Scale also leads to increasing numbers of
employees which in turn leads to specialization.
Up until about the late 1 9 50s academic writing This narrows the scope of each job so that it
about organizational structure was dominated becomes less varied and complex, which in turn
by the classical management school. This held lowers the uncertainty of the task. These low
that there was a single organizational structure uncertainty, repetitive tasks are amenable to
that was highly effective in organizations of all bureaucratic formalization such that they are
kinds. This structure was distinguished by a high pre-specified in job descriptions, standard
degree of decision-making and planning at the operating procedures, rules and training. This
top of the hierarchy so that the behaviour of bureaucratization further reduces the uncer­
lower hierarchical levels and of operations was tainty of those performing the tasks. Thus task
specified in detail in advance by senior manage­ uncertainty is the core contingency concept that
ment, through job definitions, work study and has implications for second-order contingency
the like ( Brech 1957). concepts such as innovation and size.
The classical management school held sway Much of the significance of task uncertainty
for the first half of this century, but was resides in the insight that the more uncertain the
challenged increasingly from the 1930s onwards task the more information that has to be
by the human relations school. This approach processed and this in turn shapes the commu­
focused on the individual employee as possessing nications and control structures (Galbraith
psychological and social needs. An understand­ 1973). The more uncertain the task, the less the
ing of these would allow an appreciation of how work activities can be scheduled in advance and
STR UCTURAL CONTINGENC Y THEOR Y 53

the more the reliance on ad hoc arrangements. structure in which organizational roles were
Moreover, organizations often have to deal with loosely defined and arrived at by mutual
uncertainty by utilizing diverse bodies of discussion between employees, with knowledge
expertise and this requires departure from being dispersed among the employees who
deference to hierarchy as some of the expertise possessed varieties of expertise germane to the
may be possessed by those at lower hierarchical organizational mission. Burns and Stalker ( 1 96 1 )
levels. Some part of these experts may be argued that where a n organization faces a stable
professionals and this tends to amplify the shift environment then the mechanistic structure is
away from hierarchical control of employees. effective but where the organization faces a
The core assumption of structural contingency high level of technological and market change
theory is that low uncertainty tasks are most then the organic structure is required. The
effectively performed by centralized hierarchy mechanistic structure becomes counter-produc­
since this is simple, quick and allows close tive where a high rate of innovation is needed;
coordination cheaply. As task uncertainty the resulting high uncertainty of the environment
increases, through innovation or the like, then and of the tasks in the organization means that
the hierarchy needs to loosen control somewhat spontaneous cooperation within teams of
and be overlain by participatory, communicative experts, that is, the organic structure, is more
structures. This reduces structural simplicity and effective.
raises costs but is rewarded by the benefits from The Burns and Stalker ( 1 96 1 ) theory was
innovation. As size increases the compact, advanced in a book that gave extensive
simple centralized structure is replaced by a illustrations from qualitative case studies of the
bureaucracy featuring a tall hierarchy and ex­ electronics industry. This is probably the most
tensive specialization. This bureaucracy allows widely received contribution in the structural
decentralization because employees are increas­ contingency theory literature. It provided in one
ingly controlled through formalization (e.g. stroke a synthesis between classical management
rules) and decentralization is increasingly and human relations schools in the mechanistic
required because the increase in scale, internal and organic structures, respectively. It resolved
structural complexity and length of hierarchy the debate between them with the compromise
makes centralization infeasible. Bureaucracy that each was valid in its own place. It also gave
brings disbenefits through rigidity, dysfunctions primacy to task uncertainty, driven by innova­
and some loss of control, but these are more than tion, as the contingency factor.
out-weighed by the increase in predictability, At about the same time as Burns and Stalker,
lower average wages, reduction in managerial Woodward ( 1 958; 1 965) conducted a compara­
overhead and increasing computerization which tive survey study of one hundred manufacturing
bureaucratization also brings. As the organiza­ organizations. She examined their organiza­
tion increases the range and complexity of its tional structures and found them to be unrelated
outputs, that is products or services, or increases to the size of their organizations. Operations
its geographical extensiveness, such as through technology emerged as the key correlate of
becoming a multinational, so it further increases organizational structure (Woodward 1 965).
its structural complexity and decentralization, Where production technology was primitive,
through adoption of a divisional or matrix with single articles or small batches being made,
structure. often mainly by hand and involving craft skills,
This then is the framework that provides the for example, musical instruments, the organ­
underlying theoretical unity of the ideas compos­ ization was fairly informal and organic. Where
ing structural contingency. Such a totalizing production technology had advanced to large­
vision is possible in retrospect, but the theory batch and mass production using more specia­
was developed in more piecemeal fashion, lized machinery, such as in automobile assembly,
through breakthroughs that identified a connec­ work organization was more formalized and
tion between a particular contingency factor or mechanistic and more according to the prescrip­
factors and a structural factor or factors. These tions of classical management. However, with
theoretical insights were typically advanced in further technological advance to more capital
studies that offered empirical support through intensive and automated production so that
field studies of actual organizations. product flowed continuously, such as oil in an oil
The seminal statement that pioneered the refinery, the regimentation of mass production
contingency approach to organizational struc­ gave way to work teams run on organic and
ture was by Burns and Stalker ( 1 96 1). They human relations lines. The progressively greater
distinguished between the mechanistic structure predictability of the technical system and the
in which organizational roles were tightly smoothness of production as technology
defined by superiors who had the monopoly of advanced led first to more mechanistic and
organizational knowledge, and the organic then to more organic structures.
54 STUD YING ORGANIZA TION

The Woodward ( 1 965) model was more problematic. The solution is higher levels of
complex than that of Burns and Stalker ( 1 96 1 ), integration provided by more integrating per­
being of three rather than two stages. However, sonnel in project teams and the like, coupled
it shared a similar conceptualization of structure, with interpersonal processes that defuse conflict
as mechanistic or organic, and had some through taking a problem-solving approach.
similarity in contingency factor in that techno­ Lawrence and Lorsch ( 1 967) advanced their
logically induced uncertainty was a common­ theory in a comparative study of different
ality. Moreover, Woodward, like Burns and organizations in three industries: containers,
Stalker, held that the future belonged to the processed foods and plastics. They demonstrated
organic, human relations, style of management also that organizations whose structures fitted
and that this would be forced upon management their environments had higher performance.
by technological change. The task of research Hage ( 1 965) advanced an axiomatic theory of
and academic writing in this approach was to organizations, similar to Burns and Stalker, in
bring these models and findings to the attention which centralized, formalized organizations
of managers so that they could avoid the produced high efficiency but low innovation
inefficiencies that both Woodward ( 1 965) and rates while decentralized, less formalized
Burns and Stalker ( 1 96 1 ) depicted resulted from organizations produced low efficiency but high
failure to adapt organizational structure to innovation rates. Thus which structure was
technological change rapidly enough. optimal depended upon whether efficiency or
Unlike Burns and Stalker ( 1 96 1 ), Woodward innovation was the organizational objective.
( 1 958; 1 965) used quantitative measures of Hage and Aiken ( 1 967; 1 969) demonstrated the
organizational structure, such as the span of validity of the theory in a study of health and
control of the first line supervisor, the number of welfare organizations.
levels of management in the hierarchy and the Perrow ( 1 967) argued that knowledge tech­
ratio of direct to indirect labour. Woodward nology was a contingency of organizational
( 1 965) gives many quantitative results showing structure. The more codified the knowledge used
associations between operations technology and in the organization and the fewer the exceptions
various aspects of organizational structure. encountered in operations, the more the
There is also one table ( 1 965: 69, Table 4) organization could be centralized in decision­
which shows not only an association between making.
technology and an aspect of organizational Thompson ( 1 967) advanced a book-length
structure (average span of control of the first theory of organizations containing many theor­
line supervisor), but also that organizations etical ideas and propositions. He distinguished
which conform to the association had high closed system organizations versus organizations
performance and organizations which deviated which are open systems transacting with their
had lower performance. Woodward ( 1 965) environments. He argued that organizations
argued that where the organizational structure attempt to insulate their core production
fits the organizational technology this caused technologies into a closed system to render
superior performance to those organizations them efficient through buffering the core from
whose organizational structure is in misfit to the the environment. External perturbations are
technology. dealt with by forecasting, inventories and other
Burns and Stalker and Woodward worked in mechanisms. Thompson ( 1 967) distinguished
the UK. Pioneering contributions came also also three different technologies: long-linked,
from the US. Lawrence and Lorsch ( 1 967) have mediating and intensive. M oreover, he distin­
been credited with initiating the term 'contin­ guished three different levels of interdependence
gency theory' to identify the then fledgling between activities in the workflow - pooled,
approach to which they made a major contribu­ sequential and reciprocal - and identified the
tion. They theorized that the rate of environ­ differing coordination mechanisms to handle
mental change affected the differentiation and each interdependency. He theorized that inter­
integration of the organization. Greater rates of dependencies between activities in the organiza­
environmental change require certain parts of tional workflow had to be handled at different
the organization, such as the R&D department, hierarchical levels, thus generating the design of
to face high levels of uncertainty relative to other the organization. Thompson ( 1 967) further
parts, such as the production department. This argued that the environment directly shaped
leads to large differences in structure and culture the organizational structure, with different
between departments, with R&D being more parts of the organizational structure being
organic internally and production being more specialized to conform to the requirements of
mechanistic. This greater differentiation makes different parts of the environment. Thompson
coordination between these two departments, for theorized also about organizational politics, as
instance to innovate a new product, more had Burns and Stalker and Perrow. The main
STRUCTURAL CONTINGENCY THEOR Y 55

focus of contingency theory, however, remained Other researchers analysed the significance for
upon the way the organizational structure was its structure of an organization going from
shaped so as to meet the needs of the environ­ operating only domestically to being a multi­
ment and the resulting tasks (see Donaldson national (Stopford and Wells 1 972; Egelhoff
1 996). 1 988; Ghoshal and Nohria 1 989). This leads to
In the US, Blau ( 1 970) advanced a theory of adoption of structures such as area divisions and
structural differentiation. This asserted that as product-area matrices. Egelhoff ( 1 988), in
an organization grows in size (employees) so it particular, advances a formal contingency
structures itself more elaborately into increas­ theory based on the underlying information­
ingly numerous sub-units, such as more divi­ processing requirements.
sions, more sections per division, more levels in Other contingency factors, such as environ­
the hierarchy and so on. He also argued that mental hostility (Khandwalla 1 977) and product
organizational growth leads to greater econo­ life-cycle (Donaldson 1 985b), have been identi­
mies of scale with the proportion of employees fied, and their implications for organizational
who are managers or support staff declining. structure theorized. For a model prescribing the
Weber ( 1 968) argued that organizations were optimal organization design required by the
becoming increasingly bureaucratic structures, combination of the strategy and innovation
characterized by impersonal administration, contingencies see Donaldson ( l 985a; 1 7 1 ) .
fostered in part by their increasing size. In the
UK, the Aston Group (named after their
university) argued the need to improve the
T H E STRUCTURAL CONTINGENCY
measurement of organizational structure (Pugh
et al. 1 963). They developed a large number of THEORY MODEL
quantitative measures of different aspects of
organizational structure, with attention to Increases in the innovation rate of a firm may
reliability (Pugh et al. 1 968; Pugh and Hickson reflect competition from other firms through new
1 976). The Aston Group surveyed organizations products, so the ultimate cause is the environ­
of diverse types, spanning manufacturing and ment. For this reason the contingency approach
service organizations and public and private is often termed 'the organization in its environ­
sectors. They empirically distinguished two main ment approach'. However, the environmental
dimensions of organizational structure: structur­ innovation leads the organization to raise its rate
ing of activities (how far the organization adopts of intended innovation which is the immediate
specialization by function, rules and documents) cause of the adoption of an organic structure.
and concentration of authority (centralization of Thus the structure is caused directly by the
decision-making) (Pugh et al. 1 968). They internal factor and only indirectly by the en­
examined a large number of contingency factors vironment. Both the internal and the environ­
and used multiple regression to identify the mental factors are referred to as contingencies
distinct set of predictors of organizational struc­ but a more parsimonious statement of structural
ture. For structuring the main contingency was contingency theory would need refer only to the
organizational size (number of employees): internal factor. Therefore, many contingency
larger organizations are more structured (Pugh factors of structure such as organizational size or
et al. 1 969). For centralization the main contin­ technology are internal to the organization,
gencies were organizational size and whether or though they reflect the environment such as
not the organization under study was a sub­ population size or commercially available tech­
sidiary of a parent organization: decentralization nologies. Thus while it is correct to include the
is higher in larger organizations which are environmental factors as contingencies shaping
independent (Pugh et al. 1 969). structure a sufficient explanation may be
A further structural contingency theory obtained by considering only the internal factors
focuses on the implications of the contingency as contingencies.
of corporate strategy for the organizational The import of the contingency theory may
structure of business corporations. Chandler be summarized briefly in the following way. A
( 1 962) showed historically that strategy leads to small organization, one with few employees, is
structure. Corporations need to maintain a fit organized effectively in a simple structure
between their strategy and their structure ( Mintzberg 1 979) in which there are few levels in
otherwise they suffer lower performance. Speci­ the hierarchy. Decision-making authority is
fically, a functional structure fits an undiversified concentrated in the top manager (who is often
strategy, but is a misfit for a diversified strategy the owner in a small firm) who exercises power
where a multidivisional structure is required for directly over the lower-level employees by directly
effective management of the complexity of instructing them. Thus there is little delegation of
several distinct product markets (Chandler 1 962). authority and there is also little specialization
56 STUD YING ORGANIZA TION

among the employees. As the organization grows is recourse to trial and error learning often
in size, especially in the number of employees, the accompanied by employment of more educated
structure becomes more differentiated. Many and higher trained employees such as profes­
more levels are added in the hierarchy, creating sionals. The organization has to allow them
tiers of middle managers. Some of the decision­ discretion and encourage them to use their
making authority of the top managers is initiative, with the actual division of labour
delegated down to them, commensurate with involving team elements and emerging through
their greater knowledge of local, operational discussion between employees rather than being
matters, such as direction oflower-Ievel personnel imposed by hierarchical superiors. This means
and some decisions on production. This delega­ that the R&D departments are structured more
tion is to a degree forced upon senior managers by organically than the typical production depart­
the increasing burden of decisions that they are ment. While R&D design the new offering, the
facing as organizational size and complexity production operations department makes it and
increases. Again the growth of hierarchy and the sales sells it. The dove-tailing of these require­
geographic spread of personnel makes senior ments means that successful innovation needs
management remote from 'the firing line' and so it coordination across these departments and this
becomes infeasible for them to access all the is achieved by cross-functional project teams or
required information. However, senior managers matrices or product divisions (depending upon
retain decision-making over strategy, policy and the other contingencies such as the degree of
large decisions, induding capital allocation and strategic diversification, see Donaldson J 985b).
budget amounts. As the firm diversifies from a single product or
Throughout the organization there is a greater service to mUltiple products or services, so the
division of labour as operations are broken original functional structure becomes over­
down into their components and allocated to whelmed by the complexity of decision-making.
specific departments and work-groups. Admin­ A multidivisional structure allows this complex­
istration is also increasingly broken into ity to be factored down so that each division
specializations each handled by distinct staff makes the decisions for its own product-market.
roles such as accounting, production planning, This improves the expertise and speed of the
records, personnel and so on. Behaviour is decision-making and relieves the top manage­
increasingly regulated by written job descrip­ ment of overload, allowing them to concentrate
tions, plans, procedures and rules. These on strategic decisions and more selective inter­
constitute an impersonal web regulating ventions in the divisions. The centre retains
organizational members, so that control shifts overall control through treating the divisions as
from direct, personal supervision to impersonal profit centres and creating a corporate staff to
devices. At the extreme in the large organization, monitor divisional performance and plan corpo­
its structure is a machine bureaucracy (Min­ rate strategy. Thus the organization, if large and
tzberg 1 979). The increase in scale and diversified, becomes even more bureaucratic and
specialization means that the work of any one more decentralized.
individual becomes more routine and this This in brief is the contingency theory model
facilitates its bureaucratic formalization, which of the way organizational structure changes as
in turn heightens the routineness and predict­ the contingencies change through growth.
ability of the work. The greater formalization
and predictability of employee behaviour
encourages the senior levels to increase their
THE STRUCTURAL CONTINGENCY
delegation of authority down to lower levels as
they can do so with more confidence that such RESEARCH PARADIGM
discretion will be used as the senior levels intend,
though such control is imperfect as bureaucratic Almost all of this pioneering structural con­
dysfunctions arise (Gouldner 1 954; Merton tingency research was published between
1 949). The greater specialization of personnel 1 960 and 1 970 and was the fruit of a burst of
increase their competence, which again fosters research conducted mainly in the 1 960s. Thus by
delegation, though again with some hazards 1 970 there was a well-established research
(Selznick 1 957). paradigm.
As organizations seek to innovate, in products The theory is sociological functionalism
or services or production processes, so this (Burrell and Morgan 1 979). Just as biological
entails more uncertain tasks. These tasks cannot functionalism explains the way the organs of the
be formalized by the bureaucracy i.e. the tasks human body are structured so as to contribute to
cannot be pre-specified in advance in a rule or human well-being, so sociological functionalism
procedure because this would require knowledge explains social structures by their functions, that
that the bureaucrats do not possess. Thus there is their contributions to the well-being of society
STRUCTURAL CONTINGENC Y THEOR Y 57

(Merton 1 949; 1 975; Parsons 1 9 5 1 ; 1 964). The As the research progressed it became more
organizational sociological branch of function­ sophisticated in at least four senses. Firstly,
alism posits that organizational structures are increasing attention was paid to the operational
shaped so as to provide for effective functioning definitions of concepts. For example, Woodward
by the organization (Pennings 1 992). Structural ( 1 965) measured organizational performance in
functional organizational theory proceeds in the a vague way. Later researchers were more precise
following way. Variations in organizational and recorded their definitions more explicitly,
structures are identified. These are explained by for example Child ( 1 974). Secondly, there was
each different organizational structure function­ increasing attention to reliability of measure­
ing effectively in its situation. The structure fits ment. Woodward ( 1 965) did not report the
the contingency which in turn fits the environ­ reliability of her measurements and used
ment. Fit is the underlying key. Organizations approaches that yield low reliability, such as
move into fit by adjusting their structure to their single item measures. Later researchers sought to
contingencies and this produces the observed boost reliability by using multiple item measures,
association between contingency and structure. for example the Aston Group (Pugh et al. 1 968).
The emphasis on the adaptation by the organiza­ It is now commonplace among research reported
tion to its environment makes structural in the better journals to report the reliabilities of
contingency theory part of adaptive functional­ variables. Thirdly, the theoretical models used to
Ism. explain any particular aspect of organizational
The functionalist theoretical base has meant structure went from using one contingency
that the contingency paradigm can be pursued factor, for example technology in Woodward
both by sociologists interested only in the ( 1 965), to using several, such as in Pugh et al.
explanation of organizational structure, for ( 1 969), that is from mono-causality to multi­
whom the functionality of a structure is purely causality. Fourthly, the analysis of data uses
a cause, and management theorists for whom more sophisticated statistics. Woodward ( 1 965)
the effectiveness outcomes of structures inform used only simple statistics whereas, by the late
their prescriptive advice to managers. In the 1 960s, multivariate statistics and statistics that
history of contingency theory both values have took account of sample size were being used (e.g.
motivated researchers (Hickson, personal com­ Pugh et al. 1 969).
munication). Pioneering structural contingency theory
The method used in contingency research work often used surveys of organizations at
tended to follow that used by Woodward ( 1 965). one point in time, that is a cross-sectional
A comparative study is made across a number of method. From these data inferences were made
different organizations (or across different sub­ that causation flowed in particular ways, that is
units within the same organization if they are the from contingency to structure. This adaptive
object of theoretical interest). Each contingency functionalist interpretation is a convention in
and structural factor is measured, either as a structural contingency research. Nevertheless,
quantitative scale or as a series of ordered cate­ the correlational method left room for other
gories. Each organization is allotted a score on causal interpretations. For example, Aldrich
each contingency and structural factor. The ( 1 972) reanalysed the Aston data and argued the
cross-distribution of scores of the organizations correlations were compatible with a model in
on a pair of contingency and structural factors is which structure caused size - the opposite of the
then examined to see whether there is an causal interpretation advanced by the Aston
association; this is done by cross-tabulation or Group (Pugh et al. 1 969). These alternative
correlation. The theory that associations interpretations constitute challenges to the
between contingency and structure reflect an paradigm. There has been some progress in
underlying fit is then tested. Organizations con­ resolving some of these questions of causality in
forming to the association are contrasted with favour of contingency determinism, as will be
those that deviate. If the conforming organiza­ seen below.
tions outperform the deviant organizations then The theory and empirical evidence deployed in
this signifies that the association is a fit between the structural contingency theory paradigm are
contingency and structure. Thus in much positivist. The organization is seen as being
research the empirical association is taken as forced to adjust its structure to material factors
approximating the fit (Child 1 975; Drazin and such as size and technology. Ideas and values do
Van de Ven 1 985; Woodward 1 965); however in not figure prominently as causes. Moreover,
other research the fit model is derived from little scope is seen for choice or human volition.
theory (Alexander and Randolph 1 985; Donald­ There is little information in most contingency
son 1 987). It is desirable to unite the empirically analyses about who exactly makes the structural
and theoretically derived fit models over the decisions or what their motives are or how the
course of the research programme. structures are implemented (Pugh et al. 1 969;
58 STUD YING ORGA NIZA TlON

Blau and Schoenherr 1 9 7 1 ) . Thus the analysis is Lammers and Hickson 1 979; Mansfield and
depersonalized and at the level of the organiza­ Poole 1 98 1 ; McMillan et al. 1 973). This
tion as a collective entity pursuing its objectives. interest continues through to the 1 990s and
There is thus the absence of an analysis at the has spawned a great deal of research in the
level of the human actors (Pennings 1 992). Such intervening period (as examples, Conaty et al.
an analysis would identify actors in the processes 1 983; Hickson and McMillan 1 98 1 ; Routamaa
of redesigning organizations, their beliefs, ideals, 1 985). The initial orientation of most research­
values, interests, power and tactics. Much of the ers is that they expect that they may find the
criticism from outside of the paradigm revolves contingency-structure relations of the pioneer­
around the perceived neglect of an action-level ing studies but that such general assertions are
analysis in structural contingency theory to be treated cautiously until verified empiri­
research (Silverman 1 970). Indeed the validity cally in each particular, new setting. The
of talking about 'the organization' rather than studies of replication and generalization con­
the individuals that compose the organization stitutes much of the normal science research in
has been challenged sociologically and philoso­ the structural contingency literature.
phically (Silverman 1 970). However, Donaldson The Aston Group gave emphasis to replica­
( l 985a) has offered a defence of organizational­ tion (Child 1 972a; Hinings and Lee 1 9 7 1 ; Inkson
level constructs, arguing that they are cogent and et al. 1 970). The multiple dimensions of
indispensable in organization theory. Key organizational structure found in the pioneering
phenomena such as organizational centralization study were not found in some replication studies,
and organizational performance cannot even be some of which found a single main dimension
discussed unless a collectivity-level analysis of (Child 1 972a; Grinyer and Yasai-Ardekani 1 980;
the organization as a system is made (see also 1 98 1 ; Hinings and Lee 1 97 1 ) . This is a major
Donaldson 1 990). difference in the Aston Group literature and
The adaptive functionalism, contingency-fit there have been attempts to resolve it, through
model and comparative method constitute the examination of method issues, such as the
core of the paradigm of structural contingency measurement of the variable and whether the
theory. They provide a framework in which status of the organization (as between indepen­
subsequent researchers work. dent or dependent) affects the results (Donald­
son et al. 1 975; Greenwood and Hinings 1 976;
Mansfield 1 973; see also Reimann 1 973; Star­
buck 1 98 1 ). The different findings are seen as
THE NORMAL SCIENCE PHASE:
supporting different theoretical views and as
REPLICATION AND GENERALIZATION refuting or confirming Weber ( 1 968).
In contrast, the main contingency-structure
By about 1 970 there was an established findings of the original study have been
contingency theory paradigm and those coming supported: size is the major contingency of the
afterward could orientate their efforts within this bureaucratic structuring of the activities aspect
tradition and contribute to its evolving literature of organizational structure. Replication studies
(for a collection see Donaldson 1 995b). bear this out (Pugh and Hinings 1 976). Further
The pioneering contingency studies had studies show that this finding generalizes across
produced evidence of connections between organizations of many types and nations in
contingencies and organizational structure, diverse locations. For example, Donaldson
but these might be flukes or idiosyncrasies or ( 1 986: 74) reviews thirty-five studies of the rela­
reflect biases of their authors. Therefore there tion between the contingency of organizational
was a need for replication, that is for studies size and the structural variable of degree of
by other, independent researchers to see specialization by function; all the studies found a
whether or not they found the same phenom­ positive correlation. The studies include
ena. Replication studies are seldom on the organizations from fifteen countries: Algeria,
same organizations, so the studies provide also Canada, Egypt, Finland, France, Germany,
a test of generalization, that is whether the India, Iran, Japan, Jordan, Poland, Singapore,
original findings hold in studies of new Sweden, the UK and the USA (respectively,
organizations, in settings that differ in some Zeffane 1 989; Hickson et al. 1 974; Badran and
way, such as type of organization or country, Hinings 1 9 8 1 ; Routamaa 1 985; Zeffane 1 989;
from the pioneering studies (Fletcher 1 970). Child and Kieser 1 979; Shenoy 1 98 1 ; Conaty et
For instance, during the 1 970s there arose an al. 1 983; Azumi and McMillan 1 98 1 ; Ayoubi
interest in whether different national cultures 1 98 1 ; Kuc et al. 1 98 I ; Tai 1 987; Horvath et al.
require different forms of organizational struc­ 1 98 1 ; Bryman et al. 1 983; Blau et al. 1 976). Thus
ture that render the general structural con­ the size-functional specialization relationships
tingency theories false (Hickson et al. 1 974; generalizes globally and is not confined to Anglo-
STR UCTURAL CONTINGENC Y THEOR Y 59

Saxon nations such as the UK or the USA where The topic area of strategy and structure has
these sorts of relationship were originally been studied in greater detail and so is a suitable
identified (for a review see Donaldson 1 996). one to examine for causality.

Dynamics of Strategy and Structure


CAUSAL DYNAMICS
The explanation of the correlations between
The discussion thus far has concentrated on the strategy and structure is the functionalist theory
results mainly from cross-sectional studies that that there is a fit between certain strategies and
correlate contingency and structure at the same certain structures (Chandler 1 962). Research
point in time. The contingency literature inter­ into performance has initially focused on
prets these associations according to its own whether divisional structures outperform func­
theoretical paradigm of adaptive functionalism tional structures (for example, Armour and
and contingency determinism. The question Teece 1 978). However, this is not the same as
arises as to whether this interpretation is correct. contingency theory, which holds that it is not the
Each of the major theories in structural con­ structure per se but rather whether or not it fits
tingency theory focuses on only certain cou­ the strategy, that is salient for performance. This
plings of contingency and structural factors (for requires the operationalization of a model that
example, size and bureaucracy or strategy and specifies certain combinations of strategy and
structure); indeed, critics object that there is no structure as fits and other combinations as
singular contingency theory, only a collection of misfits. Donaldson ( 1 987) advanced such a
contingency theories that constitute at best a model drawing on the work of Chandler ( 1962)
contingency approach. However, it is possible to and others.
abstract from these disparate offerings one Corporations in fit are shown to outperform
common, underlying theory. This may be those in misfit, providing empirical validation
termed the structural adaptation to regain fit (Donaldson 1 987). M oreover, fit is at a period
(SARFIT) theory (Donaldson 1 987). This holds prior to performance, adding confidence that fit
that there is fit between each contingency and is a cause and performance an effect. Hamilton
one (or more) aspect of organizational structure and Shergill ( 1 992; 1 993) have also empirically
such that fit positively affects performance and validated a very similar fit model by showing
misfit negatively affects performance. An that it relates positively to performance.
organization initially in fit changes its contin­ Organizations in fit for a number of years have
gency and thereby moves into misfit and suffers superior growth in performance during those
declining performance: this causes adoption of a years to those in misfit over the same period.
new structure so that fit is regained and per­ This means that being in fit leads to increasing
formance restored. Hence the cycle of adapta­ performance and so fit should be seen as a cause
tion is: fit, contingency change, misfit, structural and performance as a consequence. Hill et al.
adaptation, new fit. This causal model underlies ( 1 992) have also shown that the fit of strategy
many structural contingency theories (Burns and and structure is positively related to perfor­
Stalker 1 96 1 ; Lawrence and Lorsch 1 967; mance. Thus the proposition that the fit between
Williamson 1 970; 1 9 7 1 ; Woodward 1 965). strategy and structure affects performance
Commentators have argued against the receives support and some of this is from
SARFIT type of idea and have contested each research in which the temporal dimension lends
component part. They reason that the correla­ support to the causal inference that fit affects
tions between contingencies and structure signify performance. The functionalist theory that
causal processes different to those in the corporations align their structure with their
SARFIT model (Aldrich 1 972). The errors or strategy because of the underlying fit is
uncertainties in theoretical interpretation are supported empirically.
seen as made possible by limitations in the cross­ Some studies of organizational change have
sectional method. The call is made by commen­ sought for a correlation between contingency
tators for structural contingency theory studies change and structural change, during the same
to move beyond cross-sectional or synchronic time period or the immediately following time
research designs into those that study organiza­ period. Their results have been mixed and have
tional change through time, that is longitudinal tended to throw into doubt structural contin­
or diachronic studies (Mansfield and Poole 1 9 8 1 ; gency theory (Dewar and Hage 1 978; Dyas and
Galunic and Eisenhardt 1 994). Thus part of Thanheiser 1 976; Inkson et al. 1 970; Meyer
normal science has been the move to make 1 979). While contingency theory states that
studies through time in order to reveal the actual contingency causes structure, this is the long-run
causal paths. effect which flows through intermediary stages
60 STUD YING ORGA NIZA TiON

such as misfit. Thus contingency change initially to consider reverse causation in which the
only leads to misfit which eventually leads to presumed contingency factor actually results
structural change and new fit. This more from the structure. The possibility arises, there­
elongated and closely specified causal model fore, that the positive correlations between
better represents structural contingency theory. strategy and structure arise through structure
This should be the subject of empirical tests in causing strategy. However, Donaldson ( 1 982)
studies of organizational change. examined this and found no effect of divisiona­
Donaldson ( 1 987) combined data from studies lization on subsequent diversification. The
of strategy and structure in five countries correlation between strategy and structure does
(France, Germany, Japan, the UK and the not arise through structure causing strategy.
USA). First the data were analysed in the This adds confidence that the causal dynamics
traditional way: an association was sought are those identified in the SARFIT model.
between change in the contingency of strategy
and change in structure in the immediately
following period. There was no positive strat­
STRA TEGIC CHOICE
egy-structure association ( 1 987: 1 3), thus repro­
ducing the null finding of previous studies of
organizational change. Structural contingency theory is deterministic in
Then the data were analysed by examining that contingency causes structure (albeit with
each of the separate stages of the SARFIT model time lags) . The organization bows to the
and this was confirmed. Of the 87 corporations imperative of adopting a new structure that fits
that moved from fit into misfit, 83 per cent did so its new level of the contingency factor in order to
by changing their level of the contingency of avoid loss of performance from misfit. This
strategy, typically by diversifying ( 1 987: 1 4). determinism has been much criticized. Some
Thus the cycle of structural adaptation is authors reject such situational determinism,
initiated by contingency change, as SARFIT asserting instead that organizational managers
holds. Turning to the second step in the SARFIT have a free choice (Whittington 1 989) and some
model, the data were analysed to see whether speak of 'free will' (Bourgeois 1 984). Child
misfit led to structural change. Of those ( 1 972b) argues, more moderately, that the
corporations in misfit 39 per cent subsequently contingencies have some influence but that
changed their structure, whereas of those in fit there is a substantial degree of choice, which he
only 9 per cent did so ( 1 987: 1 4) . This confirms terms 'strategic choice' (see also Reed 1 985;
that misfit causes structural change. Of the Pennings 1 992).
corporations that changed their structure, 72 per Child ( 1 972b) argues that choice for managers
cent moved from misfit into fit and only 5 per and other organizational controllers arises from
cent moved from fit into misfit ( 1 987: 1 4). Thus several sources. He points out the decision­
the structural change was overwhelmingly making process that intervenes between con­
adaptive, that is, adoption of the divisional tingency and structure, so beginning to sketch an
structure to fit with the more diversified action-level analysis. Managers (and other
corporate strategy. Hence the misfit causes organizational controllers) vary in their response
structural adaptation as SARFIT holds. Thus to the contingency according to their percep­
each separate stage of the SARFIT model was tions. their implicit theories, preferences, values,
validated. interests and power (Child 1 972b). The pioneer­
When organizational change is examined by a ing structural contingency theorists make some
model that more accurately captures the full mention of these factors but nevertheless assert
processes involved in structural adaptation then the contingency imperative (Woodward 1 965).
structural contingency theory is confirmed. For Child ( l 972b) these action-level factors
Where the simplistic model that contingency gain strength from the room for manoeuvre
change leads to structural change is used to afforded by weaknesses in the systems impera­
analyse data it leads to the erroneous conclusion tives. An organization in misfit may suffer
that structural contingency theory is not performance loss, but this may be of small
supported. This is normal science at work: degree relative to other causes of performance. A
resolving findings contrary to theory by showing corporation in a dominant market position, such
that the empirical testing procedure was erro­ as monopoly or oligopoly, or a corporation in a
neous, in this case by not examining a properly protected industry, has sufficient excess profit, or
articulated model of the theory. organizational slack, that it can absorb a
Contingency theory holds that strategy leads decrement in performance, due to structural
to structure. However. Hall and Saias ( 1 980) misfit. without the profit level becoming unsa­
argue that structure leads to strategy. Bourgeois tisfactory, that is. dropping below the satisficing
( 1 984) criticizes contingency research for failing level. Thus managers of such organizations may
STRUCTURAL CONTINGENCY THEOR Y 61

retain a misfitting structure that they prefer for a 1 980). It is therefore false to see large corpora­
long time. Again, Child ( I 972b) argues that tions as seldom having to make structural
when a misfit is no longer tolerable and fit must adaptations. For example, Fligstein ( 1 985: 386,
be restored this can be done by retaining the Table 2) shows that, among the largest 1 00 US
structure and altering the contingency to fit the corporations, 71 adopted the multidivisional
structure. Thus there is no imperative to adapt structure, over the years 1 9 1 9 to 1 979. Even
structure to contingency for there is an large, wealthy corporations can face perfor­
alternative route to regain fit. In these ways the mance downturns that lower their performance
imperative to adopt a structure for a given below the satisficing level. This may arise in part
contingency is softened considerably and a larger through an economic recession, increased inter­
role for choice is seen. The strategic choice national competition, deregulation of industry
theory has been widely received and constitutes a and so on.
considerable challenge to structural contingency Critics assert that, whereas contingency theory
theory. It thereby becomes a candidate for depicts the organization as having to respond to
refutation in the normal science program of the environment, the organization may alter the
structural contingency theory. environment to make it more munificent for the
The argument of Child ( l 972b) that the organization (Perrow 1 986; Pfeffer and Salancik
systems imperatives are weaker than pioneering 1 978). This makes it easier for the organization
structural contingency theory supposed has been to be profitable and thus to avoid having to
examined and is not as valid as generally make structural alterations. Perrow ( 1 986) draws
presumed. Commentators point out that in the on the analysis of Hirsch ( 1 975) which shows
Aston research into bureaucratic structure, the that organizations enjoyed greater profitability
contingency factors accounted for only about in the pharmaceutical than the phonograph
half the variance in structure, so that much industry, because the greater government reg­
variance may be due to strategic choice. ulation of pharmaceuticals is a barrier to entry
However, the variance in structure explained that reduces competition. Presumably such a
by contingencies is understated due to measure­ benign environment would be attractive to many
ment error. Donaldson ( 1 986: 89) showed that organizational managements, yet they have not
the true correlation between size and functional all succeeded in bringing such a favourable
specialization after correcting for measurement alteration of the environment into place. This
error is 0.82. This means that 67 per cent of indicates the resilience of the environment and of
variance in structural specialization is accounted powerful institutions such as the government.
for by size, which is well over half. Of the The degree of regulation of the US pharmaceu­
remaining 33 per cent of the variance in tical industry is atypical, reflecting public
structure, some will be due to other contingency concern about drugs being more harmful than
variables and some will be due to time-lags in pop records. In fact, governmental policy in
adaptation of structure to size and the other several countries (Australia, New Zealand, the
contingencies. Thus the proportion of structural UK and the US) is increasingly to deregulate
variance available to be explained by choice is industries in order to increase competition with
under 30 per cent at best. And it may well be less the intent of curbing organizational slack and
than 30 per cent because of any other causes of forcing organizational adaptations. Thus the
structure that might exist. idea that environmental re-engineering is a ready
Research into strategy and structure shows alternative to organizational adaptation is
that organizations in misfit may delay adoption overstated and becoming less feasible with time.
of a new. fitting structure for lengthy periods, up A misfitting structure is seen as tolerable, given
to decades (Channon 1 973; Donaldson 1 987; a modicum of organizational slack, because the
Dyas and Thanheiser 1 976). Structural adapta­ negative effects of misfit on performance are seen
tion empirically tends to occur when the as minor, especially for a wealthy organization
organization in misfit has low performance enjoying market dominance such as an oligopoly
(Donaldson 1 987). This is consistent with the (Child 1 972b). However, a study by Hamilton
strategic choice argument (Child 1 9 72b). How­ and Shergill ( 1 992; 1 993) compared the perfor­
ever, the study that reveals this phenomenon mance effect of structural misfit with that of
(Donaldson 1 987; Rumelt 1 974) is of large industry concentration, an index of market
Fortune 500 corporations, that is, the pillars of domination or oligopoly. Industry concentration
American capitalism. Many of the studies of accounted for 28 per cent of profitability, and
structural adaptation to changing contingencies structural fit (to strategy) accounted for 16 per
are of large corporations (Channon 1 973; cent ( 1 993: 79). Thus the effect of organizational
Donaldson 1 987; Dyas and Thanheiser 1 976; structural misfit is similar in magnitude to that of
Fligstein 1 985; Mahoney 1 992; Palmer et al. market domination. Structural misfit is not
1 987; 1 993; Pavan 1 976; Rumelt 1 974; Suzuki trivial in its performance effect relative to
62 STUD YING ORGANIZA TION

market domination. For most firms, the degree probably greater than in larger organizations,
of organizational slack enjoyed through market where the CEO has less influence, sharing it with
domination would be almost exhausted by staff specialists, and decision-making is more
structural misfit so that performance would bureaucratized (as the authors accept ( 1 988:
decline below the satisficing level, leading to 564» . Moreover, the effect of the size con­
structural adaptation. tingency variable is restricted in a study just of
Strategic choice theory argues that an small organizations. Thus the Miller et al. ( 1 988)
organization in misfit can regain fit by altering study likely overstates the impact of CEO
its contingency to fit its structure, thereby personality and understates the effect of the
avoiding the necessity of changing a structure size contingency that would typically apply in
that the managers prefer. In fact, empirical organizations in general. Indeed Miller and
research shows that 95 per cent of corporations Droge ( 1 986: 552) found no relationship between
that move from misfit to fit do so by changes CEO personality and organizational structure in
involving structural adaptations (Donaldson large firms. Similarly, Miller and Toulouse
1 987). Corporations overwhelmingly attain fit ( 1 986: 1 397) found more numerous effects of
by adapting structure to the contingency of CEO personality on organizational structure in
strategy. Only 5 per cent of corporations move small than in large firms. Thus the effect of CEO
from misfit into fit by altering just their strategy personality on organizational structure that is
contingency to attain fit with their existing present in small firms fails to generalize
structure. Corporations do not in reality use the completely to larger firms. Thus the effects of
contingency adjustment route to fit. The CEO personality is mainly restricted to small
difference is so marked as to raise doubts that firms rather than the large corporations where
contingency adaptation is an alternative route. institutionalization of the organizational struc­
Where strategy change produces a new fit, such ture means that impersonal contingency factors
as by the corporation reducing its level of hold sway.
diversification and thereby moving from misfit Fligstein ( 1 985) shows that the functional
into fit with its existing functional structure, this background of the CEO affects structure.
may be caused by very poor performance forcing However. the functional background of the
the sell-off of non-core businesses rather than be CEO is itself affected by the structure and by the
motivated by the search for fit with a preferred corporate strategy, that is by a contingency of
structure. Instead of alternative routes to fit and structure (Fligstein 1 987). Thus it is not clear
choice, the research supports the view that that CEO background is a cause of structure that
corporations select a strategy and then tailor the itself is independent of structure and of the
structure to fit (Chandler 1 962; Christensen et al. structural contingencies. Many of the individual­
1 978). level factors that Child ( l 972b) and others see as
Thus the normal science programme of shaping structural decisions may themselves be
solving deficiencies identified in extant work in affected by organizational structure, strategy,
the paradigm of structural contingency theory size, or other contingency. For instance, power
has been able to answer to a substantial degree to affect selection of structure is presumably
the criticisms advanced from the strategic choice itself affected by the existing organizational
camp. The systems imperatives are strong and structure; similarly, the interest of a manager
constrain to a high degree the choice open to would be affected by their position in the
managers and others deciding upon organiza­ structure (see also Donaldson 1 996).
tional structures. Organizations, even large and The main attempt by Child ( 1 973) to forge an
wealthy ones, bow to the dictate of having to fit actor-level theory of structure holds that bureau­
structure to contingency in order to avoid cratic formalization is affected by the degree of
intolerable performance loss. If there is much specialization and qualifications among the
choice it is mostly restricted to timing of struc­ administrative staffs who are the architects of
tural changes (see also Donaldson 1 996). bureaucratization - specialization leads to
There have been some moves towards formalization. Thus the theory is itself struc­
demonstrating the role of individuals in the tural, explaining structure by structure. This
shaping of organizational structure, through adds to our knowledge yet is not a replacement
showing that characteristics of individuals add of structural by an action-level analysis.
to the explanation of structure by the con­ The strategic choice theory has provided the
tingencies. For instance, Miller and his col­ stimulus for a closer examination of several
leagues show that structure is affected by the issues in structural contingency theory. The
personality of the CEO (Miller et al. 1 988; Miller results support structural theory in its original
and Droge 1 986; Miller and Toulouse 1 986). form with the determinism intact.
However, the Miller et al. ( 1 988) study is of small Strategic choice theory often has a negative
organizations wherein the effect of the CEO is aspect in that it seeks to assert a role of
STRUCTURA L CONTINGENC Y THEOR Y 63

managerial choice by showing that managers Drazin and Van de Ven ( 1 985) have modelled
select structures that are less than optimal for the fit as a line of iso-performance and have meas­
situation (Child 1 9 72b), thereby exercising a ured the degree of misfit between a contingency
capriciousness for which they should be held variable and several different structural variables
morally culpable (see especially Whittington of each organization. This brings out the
1 989). Thus choice is manifested by selecting a desirability of considering fit not just between a
structure different from that which the con­ contingency and a structural variable, but
tingencies determine to be most effective. between a contingency variable and all of the
However, a second, more positive, sense of structural variables for which it is a contingency.
choice is that managers select the structure which Such a multistructural concept of fit more fully
moves the organization into fit with the reflects the underlying fit notion and so is to be
contingencies thereby increasing organizational welcomed. In turn, it opens the door to a more
effectiveness through bowing to the system fully multivariate model in which all the
imperatives. Thus they exercise choice and are contingency factors and all the structural
the human actors making the system respond but variables for which they are contingencies are
the outcome is beneficial for the organization and considered simultaneously for each organization
in conformity with contingency theory. (Randolph and Dess 1 984). This multidimen­
Support for this positive view of managerial sional model of fit would more richly capture the
choice is provided in research by Palmer et al. idea of fit. It would be more complex, but not
( 1 993). They show that the adoption of the too complex, as each structural variable has in
multidivisional structure in large US corpora­ practice only a limited number of contingencies.
tions was greater among corporations whose Many structural variables have as their con­
CEO had a graduate degree from an elite busi­ tingencies only a limited set of contingency vari­
ness school. Palmer et al. ( 1 993) argue that such ables, mostly restricted to one or a few out of the
CEOs would have acquired the idea of the variables of size, strategy, task uncertainty and
multidivisional structure through such educa­ public accountability. Clarifying the exact few
tion. The adoption of the multidivisional contingencies that apply to each different aspect
structure by large US corporations was over­ of structure and including them in multivariate
whelmingly rational adaptation to changes in the models that exhaustively capture fit and then
strategy contingency. They adopted the multi­ measuring this multivariate fit and its impact on
divisional structure to bring themselves into fit performance is the next step in fit research. It
between strategy and structure (Donaldson constitutes an important agenda item for future
1 987). Thus the effect of business education on contingency research.
divisionalization is encouraging evidence that the
education of managers in the results of structural
contingency research hastens their adoption of
THE CHALLENGE OF OTHER PARADIGMS
more effective organizational structures, as the
pioneering researchers hoped (Woodward 1 965).
As part of the growing pluralism in the study of
organizations, since about the mid 1 970s new
paradigms have arisen in sociology and econom­
FIT AND PERFORMANCE
ics which offer explanations of organizational
structure additional to those available i n
As has already been pointed out, contingency structural contingency theory (Pennings 1 992;
theory centrally holds that there is a fit between Davis and Powell 1 992). These include resource
the organizational structure and the organiza­ dependence (Pfeffer and Salancik 1 978), institu­
tional contingency that affects organizational tional (Powell and DiMaggio 1 99 1 ), population­
performance. There has been renewed interest in ecology (Hannan and Freeman 1 989) and agency
the conceptualization and operational measure­ (Jensen and Meckling 1 976) theories and
ment of fit during the 1 980s and subsequently. transaction cost economics (Williamson 1 985).
This is quite marked among researchers in the Some of these theories are outlined in other
US. Such developments include the critical work chapters of this Handbook. Elsewhere a detailed
of Schoonhoven ( 1 98 1 ). Others have sought to discussion and critique is offered of each of these
investigate the empirical relationship between organization theories and arguments in favour
their operational definition of fit and organiza­ of contingency theory are presented (Donaldson
tional performance, assessed in various ways 1 995a). Our view is that while these newer
(Alexander and Randolph 1 985; Argote 1 982; organization theories have something to con­
Drazin and Van de Ven 1 985; Gresov 1 989; tribute that supplements contingency theory it
Gresov et al. 1 989; Van de Ven and Drazin remains the core explanatory theory of organiza­
1 985). tional structure (see Donaldson 1 995a).
64 STUD YING ORGANIZA TION

REFLECTIONS ON THE STRUCTURAL theoretical paradigm. Such eclecticism is to a


CONTINGENCY THEORY PARADIGM degree resisted by the hard-core adherents of
each of the organization theory paradigms (see
Aldrich 1 992). However, the more typical
The normal science of structural contingency contemporary researchers seek to accommodate
theory has been pursued by a number of these differing ideas within their research models
scholars, as we have seen. However, it is not (Fligstein 1 985; Palmer et al. 1 993). While there
popular in all quarters and has probably are difficulties in realizing integration between
declined in popularity since 1 970. There have the diverse contemporary paradigms (see
arisen many new and different approaches, for Donaldson 1 995a), the attempt to re-integrate
example, institutional theory in the US (Meyer the field is greatly to be commended. This
and Scott 1 983) and action theory in the UK eclectic use may be becoming the largest use of
(Silverman 1 970). The US has witnessed a structural contingency theory.
particular profusion of new organizational Since structural contingency theory began as a
theories (see Donaldson 1 995a). Much organiza­ synthesis between the opposed ideas of the
tional structural research has been conducted classical management and human relations
under their ambit. It has been suggested that schools, it is not inappropriate that it in turn
career incentives lie more in innovating a new should become synthesized with other organiza­
paradigm than in persevering with an older tion theories in a wider model. The issue then
paradigm (Aldrich 1 992; Mone and McKinley becomes whether structural contingency theory
1 993). Moreover, awareness of alternative views is to be a minor or major part of that new
has combined with specific negative findings synthesis. Proponents of structural contingency
within the structural contingency research so theory will see it as providing the major com­
that some researchers interpret their findings as ponent of the new synthesis (Donaldson 1 995a).
challenges to the paradigm and advance funda­ Proponents of the other organization theories
mental changes. For example, Cullen et al. will see structural contingency theory as provid­
( 1 986) restudied the Blau ( 1 970) theory and ing only a minor part and their own preferred
variables across time; they interpret their theory as providing the major component of the
negative findings as meaning that Blau's theory new synthesis. This may well be one of the main
needs to be seen only as a theory of static scale debates in the immediate future of organization
rather than the size change dynamics that Blau studies.
claimed. In such cases, the researchers are not
treating negative findings as puzzles to be solved,
as is the way in normal science mode. BIBLIOGRAPHY
Thus the normal science of structural con­
tingency theory has been pursued only by some
students of organization. Nevertheless their Aldrich, Howard E. ( 1 972) 'Technology and organiza­
results have indicated that considerable progress tional structure: a re-examination of the findings of
has now been made in solving puzzles and the Aston Group', Administrative Science Quarterly,
advancing a strengthened structural contingency 1 7 : 26-43.
theory. Moreover, while structural contingency Aldrich, Howard E. ( 1 992) 'Incommensurable para­
theory is but one of several theories in the digms? Vital signs from three perspectives', in
research literature, the teaching literature is quite Michael Reed and Michael Hughes (eds). Rethinking
opposite. Books on organizational structure and Organization: Nel<" Directions in Organization Theory
design continue to rely greatly on structural and Analysis. London: Sage.
contingency theory and findings (Bedeian and Alexander, Judith W. and Randolph, W. Alan ( 1 985)
Zammuto 1 99 1 ; Child 1 984; Daft 1 986). 'The fit between technology and structure as a
Given the increasing theoretical pluralism of predictor of performance in nursing sub-units',
the field of organizational structure studies, Academy of Management Journal, 28: 844-59.
many contemporary empirical researchers take Argote, Linda ( 1 982) 'Input uncertainty and organiza­
the contingency-structure relationship as basic tional coordination in hospital emergency units',
and then add on variables and interpretations Administrative Science Quarterly, 27: 420-34.
from the newer structural paradigms, such as Argyris, Chris ( 1 972) The Applicability of Organiza­
institutional theory, in eclectic fashion (for tional Sociology. London: Cambridge University.
examples, Fligstein 1 985; Palmer et al. 1 993). Armour, Henry Ogden and Teece, David J. ( 1 978)
In this way the contingency theory endures in the 'Organizational structure and economic perfor­
mainstream of research among researchers who mance: a test of the multidivisional hypothesis',
maintain allegiances to more than one organiza­ Bel/ Journal of Economics, 9( 1 ) : 1 06-22.
tion theory. This eclecticism between theories Ayoubi, Z.M. ( 1 98 1 ) 'Technology, size and organiza­
marks the breakdown of each as a distinct tion structure in a developing country: Jordan', in
STRUCTURAL CONTINGENCY THEOR Y 65

D.J. Hickson and CJ. McMillan (eds), Organization firms', A ustralian Journal 0/ Management. Supple­
and Nation: the Aston Programme IV Farnborough, ment to 4( 1 ): 1 -36.
Hants: Gower. pp. 95- 1 14. Child, John ( 1 972a) 'Organization structure and
Azumi, K. and McMillan, CJ. ( 1 98 1 ) 'Management strategies of control: a replication of the Aston
strategy and organization structure: a Japanese Study', Administrative Science Quarterly, 1 7 : 1 63-77.
comparative study', in DJ. Hickson and CJ. Child, John ( 1 972b) 'Organizational structure, envir­
McMillan (eds), Organization and Nation: the Aston onment and performance: the role of strategic
Programme IV Farnborough, Hants: Gower. pp. choice', SOCiology, 6: 1 -22.
1 55-72. Child, John ( 1 973) 'Predicting and understanding
Badran, M. and Hinings, CR. ( 1 98 1 ) 'Strategies of organization structure'. Administrative Science
administrative control and contextual constraints in Quarterly, 1 8: 1 68-85.
a less developed country: the case of Egyptian public Child, John ( 1 974) 'Managerial and organizational
enterprise'. in DJ. Hickson and CJ. McMillan (eds), factors associated with company performance. Part
Organization and Nation: the Aston Programme IV I ', Journal 0/ Management Studies, I I : 1 75-89.
Farnborough, Hants: Gower. pp. 1 1 5-3 1 . Child, John ( 1 975) 'Managerial and organizational
Bedeian, A.G. and Zammuto, R.F. ( 1 99 1 ) Organization factors associated with company performance. Part
Theory and Design. Chicago: Dryden Press. 2: A contingency analysis', Journal 0/ Management
Beyer, J.M. and Trice, H.M. ( 1 979) 'A reexamination of Studies, 12: 1 2-27.
the relations between size and various components of Child, John ( 1 984) Organization: a Guide to Prohlems
oganizational complexity', Administrative Science and Practice, 2nd edn. London: Harper and Row.
Quarterly, 24: 48-64. Child, John and Kieser, Alfred ( 1 979) 'Organizational
Blau, Peter M. ( 1970) 'A formal theory of differentia­ and managerial roles in British and West German
tion in organizations', American Sociological Review, companies: an examination of the culture-free
35(2): 20 1 - 1 8. thesis', in CJ. Lammers and DJ. Hickson (eds),
Blau, Peter M. ( 1 972) 'Interdependence and hierarchy Organizations Alike and Unlike. London: Routledge
in organizations', Social Science Research, I : 1 -24.
and Kegan Paul. pp. 25 1 -7 1 .
Blau, Peter M. and Schoenherr, P.A. ( 1 97 1 ) The
Child, John and Mansfield, Roger ( 1 972) 'Technology,
Structure 0/ Organizations. New York: Basic Books.
size and organization structure', Sociology, 6(3):
Blau. Peter M . , Falbe, Cecilia McHugh, McKinley,
369-93.
William and Tracy, Phelps K. ( 1 976) 'Technology
Christensen, C Roland, Andrews, Kenneth R. and
and organization in manufacturing', Administrative
Bower, Joseph L. ( 1 978) Business Policy: Text and
Science Quarterly, 2 1 ( 1 ): 2 1 -40.
Cases, 4th edn. Homewood, IL: Richard D. Irwin.
Bourgeois III, L.J. ( 1984) 'Strategic management and
Collins, Paul D. and Hull. Frank ( 1 986) 'Technology
determinism', Academy 0/ Management Review. 9:
and span of control: Woodward revisited', Journal 0/
586-96.
Management Studies, 32(2): 143- 1 64.
Brech. E.F.L. ( 1 957) Organisatioll: the Framework 0/
Conaty, J., Mahmoudi. H. and Miller, G.A. ( 1 983)
Management. London: Longmans, Green.
'Social structure and bureaucracy: a comparison of
Bryman, A., Beardsworth, A.D., Keil, E.T. and Ford, J.
organizations in the United States and prerevolu­
( 1 983) 'Research note: organizational size and
specialization', Organization Studies, 4(3): 27 1 -7. tionary Iran', Organization Studies, 4(2): 1 05-28.
Burns, Tom and Stalker, G.M. ( 1 96 1 ) The Management Crawford, Michael ( 1 983) The character, determinants
0/ Innovation. London: Tavistock. and performance effects of inter-unit interactions
Burrell, Gibson and Morgan, Gareth ( 1 979) Socio­ within organisations: a disaggregated systems
logical Paradigms and Organizational Analysis: approach'. PhD thesis, University of New South
Elements 0/ the Sociology 0/ Corporate Life. Wales, Australia.
London: Heinemann. Crawford, Michael ( l 985a) 'Performance loss as an
Capon, N. and Christodolou. C ( 1 987) 'A comparative outcome of aggregate inter-unit strain within
analysis of the strategy and structure of United organisations'. Paper presented at the Academy of
States and Australian corporations', Journal 0/ Management National Meeting, San Diego, Cali­
International Business SllIdies, 1 8 (Spring): 5 1 -74. fornia.
Chandler. Alfred D. Jr ( 1 962) Strategy and Structure: Crawford, Michael ( l 985b) 'Style, effort and outcome:
Chapters in the History 0/ the American Industrial relationships between boundary spanning effort,
Enterprise. Cambridge, MA: MIT Press. conflict style, appreciation and inter-unit disso­
Channon, Derek F. ( 1 973) The Strategy and Structure nance'. Paper presented at the Academy of
0/ British Enterprise. London: Macmillan. Management N ational Meeting, San Diego, Cali­
Chan non, Derek F. ( 1 978) The Service Industries: fornia.
Strategy Structure and Financial Performance. Cullen, J.B., Anderson, K.S. and Baker, D.D. ( 1 986)
London: Macmillan. 'Blau's theory of structural differentiation revisited:
Chenhall. Robert H. ( 1 979) 'Some elements of a theory of structural change or scale', Academy 0/
organizational control in Australian divisionalized Management Journal, 29(2): 203-29.

stribution forbidden.
66 STUD YING ORGANIZA TION

Daft, Richard L. ( 1 986) Organi:ation Theory and Faas, F.A.M.J. ( 1 985) 'How to solve communication
Design, 2nd edn. St Paul: West. problems on the R and D interface', Journal of
Davis, Gerald F. and Powell, Walter W. ( 1 992) Management Studies, 22( 1 ): 83- 1 02.
'Organization-environment relations', in Marvin Fletcher, Colin ( 1 970) 'On replication: notes on the
Dunnette and Laetta M. Hough (eds), Handbook of notion of a replicability quotient and a general­
Industrial and Organizational Psychology, 2nd edn. izability quotient', Sociology, 4(January): 5 1 -69.
Palo Alto, CA: Consulting Psychologists Press. pp. Fligstein, Neil ( 1985) 'The spread of the multidivisional
3 1 6-75. form among large firms, 1 9 19- 1 979', American
Dewar, Robert and Hage, Jerald ( 1 97 8 ) 'Size, Sociological Review, 50: 377-9 1 .
technology, complexity, and structural differentia­ Fligstein, Neil ( 1 987) The intraorganizational power
tion: toward a theoretical synthesis'. Administrative struggle: rise of finance personnel to top leadership in
Science Quarterly, 23( 1 ) : 1 1 1 -36. large corporations, 1 9 1 9 - 1 979', American Socio­
Donaldson, Lex ( 1 976) 'Woodward, technology, logical Review, 52: 44-58.
organizational structure and performance - a Ford, Jeffrey D. and Slocum, John W. Jr ( 1 977) 'Size,
critique of the universal generalization', Journal of technology, environment and the structure of
Management Studies, 1 3(3): 255-73. organizations', Academy of Management Review,
Donaldson, Lex ( 1 982) ' Divisionalization and diversi­ 2(4): 561 -75.
fication: a longitudinal study', Academy of Fry, Louis W. ( 1 982) 'Technology-structure research:
Management Journal, 25(4): 909-· 14. three critical issues', Academy of Management
Donaldson, Lex ( 1 985a) In Defence of Organi:ation Journal, 25(3): 532-52.
Theory: a Reply to the Critics. Cambridge: Cam­ Fry, Louis W. and Slocum, John W. Jr ( \ 984)
bridge University Press. 'Technology, structure, and workgroup effective­
Donaldson, Lex ( l 985b) 'Organization design and the ness: a test of a contingency model', Academy of
life-cycles of products', Journal of Management Management Journal, 27(2): 221 -46.
Studies, 22( 1 ) : 25-37. Galbraith, Jay R. ( 1 973) Designing Complex Organiza­
Donaldson, Lex ( 1 986) 'Size and bureaucracy in East tions. Reading, MA: Addison-Wesley.
and West: a preliminary meta-analysis', in S.R. Galunic, R. Charles and Eisenhardt, Kathleen M.
Clegg, D. Dunphy and S.G. Redding (eds), The ( 1 994) 'Renewing the strategy-structure-perfor­
Enterprise and Management in East Asia. Hong mance paradigm', in B.M. Staw and I.I. Cummings
Kong: University of Hong Kong Press. (eds), Research in Organizational Behavior, 1 6: 2 1 5-
Donaldson, Lex ( 1 987) 'Strategy and structural 55.
adjustment to regain fit and performance: in Gerwin, Donald ( 1 977) 'Relationship between structure
defence of contingency theory', Journal of Manage­ and technology', in Paul Nystrom and William
ment Studies, 24( 1 ) : 1 -24. Starbuck (eds), Handbook of Organizational Design.
Donaldson, Lex ( 1 990) The ethereal hand: organiza­ Amsterdam: Elsevier.
tional economics and management theory', Academy Gerwin, Donald ( 1 979a) 'Relationships between
of Management Review, 1 5(3): 369-8 1 . structure and technology at the organizational and
Donaldson, Lex ( 1995a) American Anti-Management job levels', Journal of Management Studies, 1 6( 1 ):
Theories of Organization: a Critique of Paradigm 70-9.
Proliferation. Cambridge: Cambridge University Gerwin, Donald ( l 979b) 'The comparative analysis of
Press. structure and technology: a critical appraisal',
Donaldson, Lex ( 1 995b) Contingency Theory, vol. IX, A cademy of Management Review, 4(\ ) : 4 1 - 5 \ .
in D.S. Pugh (ed.), History of Management Thought. Gerwin, Donald and Christoffel, Wade ( 1 974)
Aldershot: Dartmouth Press. 'Organizational structure and technology: a compu­
Donaldson, Lex ( 1 996) For Positivist Organi:ation ter model approach', Management Science, 20(1 2):
Theory: Proving the Hard Core. London: Sage. 1 5 3 1 -42.
Donaldson, Lex, Child, John and Aldrich, Howard Ghoshal, Sumantra and Nohria, Nitin ( \ 989) 'Internal
( 1 975) 'The Aston findings on centralization: further differentiation within multinational corporations',
discussion', Administrative Science Quarterly, 20: Strategic Management Journal, 1 0(4): 323-37.
453-60. Gouldner, Alvin ( 1 954) Patterns of Industrial Bureau­
Drazin, Robert, and Van de Ven, Andrew H. ( 1 985) cracy. Glencoe, IL: Free Press.
'Alternative forms of fit in contingency theory', Greenwood, Royston and Hinings, C.R. ( 1 976)
Administrative Science Quarterly, 30: 5 1 4-39. 'Centralization revisited', Administrative Science
Dyas, Gareth P. and Thanheiser, Heinz T. ( 1 976) The Quarterly, 2 1 ( 1 ) : 1 5 1 -5.
Emerging European Enterprise: Strategy and Struc­ Gresov, C. ( 1 989) 'Exploring fit and misfit with multiple
ture in French and German Industry. London: contingencies', Administrative Science Quarterly, 34:
Macmillan. 43 1 -53.
Egelhoff, William G. ( \ 988) Organi:ing the Multi­ Gresov, c., Drazin, R. and Van de Ven, A. H. ( 1 989)
national Enterprise: an Information Processing 'Work-unit task uncertainty, design and morale',
Perspective. Cambridge, MA: Ballinger. Organi:ation Studies, 1 0( 1 ): 45-·62.
STR UCTURAL CONTINGENCY THEOR Y 67

Grinyer, Peter H. and Yasai-Ardekani, Masoud ( 1980) 'Divisionalization, decentralization and perfor­
'Dimensions of organizational structure: a critical mance of large United Kingdom companies',
replication', Academy of Management Journal, 23: Journal of Management Studies, 23( 1 ): 26-50.
405-2l . Hill, Charles W.L., Hitt, Michael A. and Hoskisson,
Grinyer, Peter H . and Yasai-Ardekani, Masoud Robert E. ( 1992) 'Cooperative versus competitive
( l 98 I a) 'Research note: some problems with structures in related and unrelated diversified firms',
measurement of macro-organizational structure', Organization Science, 3(4): 501 -2l .
Organization Studies, 2(3): 287-96. Hinings, CR. and Lee, Gloria ( 197 1 ) 'Dimensions of
Grinyer, Peter H. and Yasai-Ardekani, Masoud organization structure and their context: a replica­
( 1 9 8 1 b) 'Strategy, structure, size and bureaucracy', tion', Sociology, 5: 83-93.
Academy of Management Journal, 24(3): 47 1 -86. Hinings, CR., Ranson, S., and Bryman, A. ( 1976)
Grinyer, Peter H., Yasai-Ardekani, Masoud and AI­ 'Churches as organizations: structure and context',
Bazzaz, Shawki ( 1 980) 'Strategy, structure, the in D.S. Pugh and CR. Hinings (eds), Organizational
environment, and financial performance in 48 Structure: Extensions and Replications: the Aston
United Kingdom companies', A cademy of Manage­ Programme. Farnborough, Hants: Saxon House.
ment Journal, 23(2): 1 93-220. Hirsch, Paul ( 1 975) 'Organizational effectiveness and
Hage, Jerald ( 1 965) 'An axiomatic theory of organiza­ the institutional environment', Administrative
tions', Administrative Science Quarterly, 10(4): 289- Science Quarterly, 20(3): 327-44.
320. Hirst, Mark H. ( 1 984) 'A partial meta-analysis of the
Hage, Jerald and Aiken, Michael ( 1 967) 'Program relationship between organisational structure and
change and organizational properties: a comparative production technology', A ustralian Journal of
analysis', American Journal of Sociology, 72: 503- 19. Management, 9(2): 35-42.
Hage, Jerald and Aiken, Michael ( 1 969) 'Routine Horvath, D., McMillan, CJ., Azumi, K. and Hickson,
technology, social structure and organizational DJ. ( 1 9 8 1 ) 'The cultural context of organizational
goals', Administrative Science Quarterly, 1 4(3): control: an international comparison', International
366-76. Studies of Management and Organization, 6: 60-86.
Hage, Jerald and Dewar, Robert ( 1 973) 'Elite values Inkson, J.H.K., Pugh, D.S. and Hickson, D.J. ( 1 970)
versus organizational structure in predicting innova­ 'Organization context and structure: an abbreviated
tion', Administrative Science Quarterly, 1 8(3): 279- replication', Administrative Science Quarterly,
90. 1 5( 1 3) : 3 1 8-29.
Hall, DJ. and Saias, M.D. ( 1 980) 'Strategy follows Jelinek, Mariann ( 1 977) Technology, organizations
structure!', Strategic Management Journal, I : 149- and contingency', Academy of Management Review,
63. 2(1 ) : 1 7-26.
Hamilton, R.T. and Shergill, G.S. ( 1992) The relation­ Jensen, Michael C and Meckling, William H. ( 1976)
ship between strategy-structure fit and financial 'Theory of the firm: managerial behavior, agency
performance in New Zealand: evidence of generality costs and ownership structure', Journal of Financial
and validity with enhanced controls', Journal of Economics, 3(4): 305-60.
Management Studies, 29(1 ) : 95- 1 1 3 . Khandwalla, Pradip N. ( 1 973) 'Viable and effective
Hamilton, R.T. and Shergill, G.S. ( 1 993) The Logic of organizational designs of firms', A cademy of
New Zealand Business: Strategy, Structure, and Management Journal, 1 6(3): 48 1 -95.
Performance. Auckland, New Zealand: Oxford Khandwalla, Pradip N. ( 1 974) 'Mass output orienta­
University Press. tion of operations technology and organizational
Hannan, Michael T. and Freeman, John ( 1 989) structure', Administrative Science Quarterly, 19(1):
Organizational Ecology. Cambridge, MA: Harvard 74-97.
University Press. Khandwalla, Pradip N. ( 1 977) The Design of Organiza­
Harvey, Edward ( 1968) Technology and the structure tions. New York: Harcourt Brace Jovanovich.
of organizations', American Sociological Review, 33: Kuc, B., Hickson, D.l. and McMillan, CJ. ( 1 98 1 )
247-59. 'Centrally planned development: a comparison of
Hickson, David J. and McMillan, CJ. (eds) ( 1 9 8 1 ) Polish factories with equivalents in Britain, Japan
Organization and Nation: the Aston Programme I V. and Sweden', in D.J. Hickson and CJ. McMillan
Farnborough, Hants: Gower. (eds), Organization and Nation: the Aston Pro­
Hickson, David 1., Pugh, D.S. and Pheysey, Diana G. gramme IV. Farnborough, Hants: Gower. pp. 75-9 l .
( 1 969) 'Operations technology and organization Kuhn, Thomas S . ( 1 970) The Structure of Scientific
structure: an empirical reappraisal', Administrative Revolutions, 2nd enlarged edn. Chicago: University
Science Quarterly, 14(3): 378-97. of Chicago Press.
Hickson, David J., Hinings, CR., McMillan, CJ. and Lammers, C J . and Hickson, D.J. (eds) ( 1 979)
Schwitter, J.P. ( 1 974) 'The culture-free context of Organizations Alike and Unlike: International and
organization structure: a trinational comparison', Inter-Institutional Studies in the Sociology of
Sociology, 8( 1 ): 59-80. Organization. London: Routledge and Kegan Paul.
Hill, Charles W.L. and Pickering, J.F. ( 1 986) Lawrence, Paul R. and Lorsch, Jay W. ( 1 967)
68 STUD YING ORGA NIZA TIO:V

Organi::ation and £nrironmellt: ,\!Il1nugillg Di[(i:rell­ Orgalli::atiollal £llrirollllll'lltS: Ritual and Rationality.
tiatioll and Integration. Boston: Di\'ision of Be\ erly Hills. CA: Sage.
Research, Graduate School of Business Administra­ Meyer. Marshall W. ( 1 979) Change ill Public Bureau­
tion, Harvard University. cracies. Cambridge: Cambridge University Press.
Leatt. Peggy and Schneck. Rodney ( 1 982) 'Technology. Miller. Danny and Droge. Cornelia ( 1 986) 'Psycholo­
size. environment, and structure in nursing subunits'. gical and traditional determinants of structure'.
Organi::arion Studies, 3(2): 22 1 -42. Adlllinistrati!'e Science Quarterly. 3 1 (4): 539-60.
Likert, Rensis ( 1 96 1 ) Nell' Pallems of Managemclll . Miller. Danny and Toulouse. Jean-Marie ( 1 986) 'Chief
New York: McGraw-Hill. executive personality and corporate strategy and
Lincoln. J.R., Hanada, M. and McBride. K. ( 1 986) structure in small firms', Managellleni Science,
'Organizational structures in Japanese and U.S. 32( I I ) : 1 389-409.
manufacturing', Administrative Scienet' QUlIrterlr. Miller. Danny. Droge. Cornelia and Toulouse, Jean­
3 1 : 338-64. Marie ( 1 988) 'Strategic process and content as
Lioukas, S.K. and Xerokostas, D.A. ( 1 982) 'Size and mediators between organizational context and
administrative intensity in organizational diyisions·. structure'. Academy of Managemenl Journal. 3 1 ( 3 ):
Management Science, 28(8): 854-68. 544-69.
Lorsch, Jay W. and Allen, Stephen A. ( 1 973) Mallagillg Miller. George A. ( \ 987) 'Meta-analysis and the
Diversity and Il1ler-dependenct': all Orgalli::ariollul culture-free hypothesis'. Organi::ation SllIdies. 8(4):
Studv of Multidivisional Firms. Boston: Division of 309- 26.
Research, Graduate School of Business Administra­ Mintzberg. H. ( 1 979) The Structuring of Organi::ations:
(/ Srl1lhesis of the Research. Englewood Cliffs. NJ:
tion, Harvard University.
Prentice-Hall.
Mahoney, Joseph T. ( 1 992) 'The adoption of the
Mone. Mark A. and McKinley. William ( 1 993) 'The
multidivisional form of organization: a contingency
uniqueness value and its consequences for organiza­
model', Journal of Management SllIdies. 29( I ) : 49-
tion studies'. Journal of Management Inquir)', 2(3):
72.
284-96.
Mahoney, Thomas A. and Frost, Peter J. ( 1 974) 'The
Palmer. Donald A .. Friedland. Roger. Jennings, P.O.
role of technology in models of organizational
and Powers. Melanie E . ( 1 987) The economics and
effectiveness', Organi::ational Belli/rior und HUIlli/1I
politics of structure: the multidivisional form and
Performallce. I I : 1 22-38.
large U.S. corporations'. Administrative Science
Mansfield, Roger (\973) 'Bureaucracy and centraliza­
Quarta/r. 32: 25-48.
tion: an examination of organizational structure'.
Palmer. Donald A .. Jennings, P. Devereaux and Zhou,
Administrative Science Quarterly. 1 8 : 477 - 88.
Xueguang ( \ 993) 'Late adoption of the multi­
Mansfield, Roger and Poole. Michael ( 1 98 1 ) Il1lema­
divisional form by large U . S . corporations:
tional Perspectives on Management and Orgalli::atioll.
institutional. political. and economic accounts ',
Aldershot, Hampshire: Gower. A tllIlillislralire Science Quarterly. 38: 1 00-3 1 .
Marsh. Robert M. and Mannari. Hiroshi ( 1 976) Parsons. T. ( 1 95 1 ) The Social Srstem. Chicago, IL: Free
Moderni::ation and the Japanese Facto/T. Princeton. Press.
NJ: Princeton University Press Parsons. T. ( 1 964) 'Suggestions for a sociological
Marsh. Robert M. and Mannari, Hiroshi ( 1 980) approach to the theory of organizations'. in Amitai
'Technological implications theory: a Japanese Etzioni (cd.). Complex Organi::atiolls: a Sociological
test'. Organization Studies, 1 (2): 1 6 1 -83. Readt'/". New York: Holt.
Marsh, Robert M . and Mannari. Hiroshi ( 1 98 1 ) Pavan. Robert J. ( 1 976) 'Strategy and structure: the
'Technology and size as determinants of the Italian experience'. Journal of Economics and
organizational structure of Japanese factories'. Busilless. 28(3): 254-60.
Administrative Science Quarter/r. 26( I ) : 33--57. Pennings. Johannes M. ( 1 992) 'Structural contingency
McMillan, Charles J . . Hickson. David J . . Hinings. theory: a reappraisal', in B.M. Staw and 1 . 1 .
Christopher R. and Schneck. Rodney E. ( 1 973) 'The Cummings (eds). Research in Organi::ational Beha­
structure of work organizations across societies' . vior. 14: 267-309.
Administrative Science Quarterlr. 16: 555-69. Perrow. Charles ( 1 967) . A framework for the
Meadows, Ian S.G. ( \ 980) 'Organic structure and comparative analysis of organizations', American
innovation in small work groups'. HUlIIlIn Rcili tiom . Sociological Reriell·. 32(3): 1 94--208.
33(6): 369-82. Perrow. Charles ( \ 986) Complex Organi::ations: a
Merton. R . K . ( 1 949) Social Thl'or r alld Socilll Critical £.1'.1' 01'. 3rd edn. New York: Random House.
Structure. Chicago, IL: Free Press. Pfeffer. Jeffrey ( 1 982) Orgalli::ations and Organi::alion
Merton. R.K. ( 1975) 'Structural analysis in sociology'. Theo/"r. Boston: Pitman.
in P.M. Blau (ed.). Approaches to Iht' SllIt ll" o{Sociul Pfeffer. Jeffrey ( 1 993) 'Barriers to the advance of
Structures. New York: Free Press. organizational science: paradigm development as a
Meyer, John W. and Scott, W. Richard with the dependent \ ariable', Academy of Management
assistance of Rowan, B. and Deal. T.E. ( 1 983) R<'I'iell·. 1 8(4): 599-620.
STRUCTURAL CONTINGENCY THEOR Y 69

Pfeffer, Jeffrey and Salancik, Gerald R. ( 1 978) The sions of structures in certain Finnish companies',
External Control of Organi::ations: a Resource lournal of Management Studies, 22(5): 498-522.
Dependence Perspective. New York: Harper and Rumelt, Richard P. ( 1 974) Strategy, Structure and
Row. Economic Performance. Boston: Division of
Powell, Walter W. and DiMaggio, Paul J. ( 1 99 1 ) The Research, Graduate School of Business Administra­
Nelf Institutionalism in Organi::.ational Analysis. tion, Harvard University.
Chicago: University of Chicago Press. Sathe, V. ( 1 978) 'Institutional versus questionnaire
Pugh, D.S. and Hickson, D.1. ( 1 976) Organizational measures of organizational structure', A cademy of
Structure in its Context: the Aston Programme I. Management, 2 1 : 227-38.
Farnborough, Hants: Saxon House. Schoonhoven, Claudia Bird ( 1 98 1 ) 'Problems with
Pugh, D.S. and Hinings, CR. ( 1 976) Organi::ational contingency theory: testing assumptions hidden
Structure: Extensions and Replications: the Aston within the language of contingency "theory" ',
Programme II. Farnborough, Hants: Saxon House. A dministrative Science Quarterly, 26(3): 349-77.
Pugh, D.S., Hickson, 0.1., Hinings, CR., Macdonald, Scott, W. Richard ( 1992) Organizations: Rational,
K . M . , Turner, C and Lupton, T. ( 1 963) 'A Natural and Open Systems, 3rd edn. Englewood
conceptual scheme for organizational analysis', Cliffs, NJ: Prentice-Hall.
Administrative Science Quarterly, 8(3): 289-3 1 5. Selznick, P. ( 1 957) Leadership in Administration. New
Pugh, D.S., Hickson, D.1., Hinings, CR. and Turner, York: Harper and Row.
C ( 1 968) 'Dimensions of organization structure', Shenoy, S. ( 1 98 1 ) 'Organization structure and context:
Administrative Science Quarterly, 1 3( 1 ) : 65- 105. a replication of the Aston study in India', in D.1.
Pugh, D.S., Hickson, D.J., Hinings, CR. and Turner, Hickson and C.1. McMillan (eds), Organi::ation and
C ( 1 969) 'The context of organization structures', Nation: the Aston Programme I V. Farnborough,
Administrative Science Quarterly, 14( 1 ): 9 1 - 1 14. Hants: Gower. pp. 1 33-54.
Silverman, David ( 1 970) The Theory of Organizations.
Randolph, W. Alan and Dess, Gregory G. ( 1 984) 'The
London: Heinemann.
congruence perspective of organization design: a
Singh, Jitendra V. ( 1 986) 'Technology, size, and
conceptual model and multivariate research
organizational structure: a reexamination of the
approach', Academy of Management Review, 9 ( 1 ) :
Okayama Study data', Academy of Management
1 1 4-27.
lournal, 29(4): 800- 1 2.
Reed, Michael ( 1 985) Redirections in Organizational
Starbuck, William H. ( 1 98 1 ) 'A trip to view the
Analysis. London: Tavistock Publications.
elephants and rattlesnakes in the Garden of Aston',
Reeves, Kynaston T. and Turner, Barry A. ( 1972) 'A
in A. Van de Ven and W. Joyce (eds), Perspectives on
theory of organization and behaviour in batch
Organization Design and Behavior. New York: Wiley.
production factories', Administrative Science Quar­
Stopford, J.M. and Wells, L.T. Jr ( 1 972) Managing the
terly, 1 7( 1 ): 8 1 -98.
Multinational Enterprise. New York: Basic Books.
Reimann, Bernard C ( 1 973) 'On the dimensions of
Suzuki, Y. ( 1 980) 'The strategy and structure of the top
bureaucratic structure: an empirical reappraisal', 1 00 Japanese industrial enterprises 1 950-1 970',
Administrative Science Quarterly, 1 8: 462-76. Strategic Management lournal, 1 (3): 265-9 / .
Reimann, Bernard C ( 1 977) 'Dimensions of organiza­ Tai, Elizabeth ( 1 987) 'Adaptability and organizational
tional technology and structure: an exploratory effectiveness: a study of thirty manufacturing firms in
study', Human Relations, 30(6): 545-66. Singapore'. PhD thesis, University of New South
Reimann, Bernard C ( 1 980) 'Organization structure Wales, Sydney, Australia.
and technology in manufacturing: system versus Tauber, I. ( 1 968) 'A yardstick of hospital organization'.
work flow level perspectives', Academy of Manage­ Diploma thesis, University of Aston, Birmingham,
ment lournal, 23( 1 ): 61 -77. England.
Reimann, Bernard C and Inzerilli, Giorgio ( 1 979) 'A Thompson, James D. ( 1 967) Organi::ations in Action.
comparative analysis of empirical research on New York: McGraw-HilI.
technology and structure', Journal of Management, Van de Ven, Andrew H. and Delbecq, A.L. ( 1 974) 'A
5(2): 1 67-92. task contingent model of work-unit structure',
Richards, V.G. ( 1 980) 'Research note: the Aston Administrative Science Quarterly, 1 9(2): 1 83-97.
Databank', Organi::ation Studies, 1 (3): 2 7 1 -8. Van de Ven, Andrew H. and Drazin, Robert ( 1 985)
Roethlisberger, Fritz J. and Dickson, William J. ( 1 939) 'The concept of fit in contingency theory', in B.M.
Management and the Worker. Cambridge, MA: Staw and L.L. Cummings (eds), Research in
Harvard University Press. Organizational Behaviour, vol. 7. Greenwich, CT:
Rousseau, Denise M. ( 1 979) 'Assessment of technology JAI Press.
in organizations: closed versus open systems Van de Ven, Andrew and Ferry, Diane L. ( 1 980)
approaches', Academy of Management Review, 4(4): Measuring and Assessing Organi::ations. New York:
5 3 1 -42. Wiley.
Routamaa, Vesa ( 1 985) 'Organizational structuring: an Van de Ven, Andrew H . , Delbecq, A.L. and Koenig, R.
empirical analysis of the relationships and dimen- Jr. ( 1 976) 'Determinants of coordination modes
70 STUD YING ORGANIZA TION

within organizations', American Sociological Review, Innovalive Polenfial. Cambridge, MA: Harvard
4 1 : 322-38. University Press.
Vroom, Victor H. and Yetton, Philip W. ( 1 973) Williamson, Oliver E. ( 1 985) The Economic Inslitutions
Leadership and Decision-Making. Pittsburgh: Uni­ of Capilalism: Firms, Markels, Relational COnfract­
versity of Pittsburgh Press. ing. New York: Free Press.
Weber, Max ( 1 968) Economy and Sociely: an OUlline of Woodward, Joan ( 1 958) Managemenf and Technology.
Inlerprelive Sociology, edited by Guenther Roth and London: HMSO.
Woodward, Joan ( 1 965) Industrial Organizalion:
Claus Wittich. New York: Bedminster Press.
Theory and Praclice. London: Oxford University
Whittington, R. ( 1 989) Corporale Slralegies in
Press.
Recession and Recovery: Social SlruclUre and
Woodward, Joan ( 1 970) Induslrial Organization:
. Slralegic Choice. London: Unwin Hyman.
Behaviour and COnfrol, London: Oxford University
Williamson, Oliver E. ( 1970) Corporale COnfrol and
Press.
Business Behavior: an Inquiry infO Ihe Effecls of Zeffane, Rachid M. ( 1 989) 'Organization structures and
Organizalion Form on Enlerprise Behavior. Engle­ contingencies in different nations: Algeria, Britain,
wood Cliffs, NJ: Prentice-Hall. and France', Social Science Research, 1 8: 3 3 1 -69.
Williamson, Oliver E. ( 1 9 7 1 ) 'Managerial discretion, Zwerman, William L. ( 1 970) New Perspectives on
organization form, and the multidivision hypoth­ Organi::alion Theory: an Empirical Reconsideralion
esis', in Robin Marris and Adrian Woods (eds), The oflhe Marxian and Classical Analyses. Westport, CT:
Corporate Economy: Growth, Compelilion and Greenwood.