Printer Friendly

Some developments in the study of organizations.

The Problem

Modern organizations can no longer be depicted in the entrepreneurial terms used by classical economists. Working capital may be obtained from numerous sources, and management is often dissociated from ownership. So the picture of the entrepreneur-proprietor who assumes the risk, takes the decisions, and disposes of the profit, no longer applies to the larger businesses today. Similarly, the "Principles of Management" of the Scientific Management School were deduced from premises which often contained faulty assumptions about human motives and behaviour. Possibly some of the more recent theories of social scientists make equally erroneous assumptions about finance, technology, and production. For example, participative or democratic management may be advocated on social grounds in circumstances where it could not possibly produce the results intended. Moreover, the theorists do not speak with one voice. All of them have made valuable contributions to the development of organization theory, but we need some means of resolving the inconsistencies between the various approaches.

Four Basic Assumptions

In view of these difficulties, how may progress be made in understanding the nature of organizations so that those who manage them may be better equipped for their task? The work of the Industrial Administration Research Unit at The University of Aston in Birmingham indicates a possible approach. This work is based on four main assumptions:

1. In order to find which organizational problems are specific to particular kinds of organizations, and which are common to all organizations, comparative studies are needed which include organizations of many types.

2. Meaningful comparisons can only be made when there is a common standard for comparison -- preferably measurement.

3. The nature of an organization will be influenced by its objectives and environment, so these must be taken into account.

4. Study of the work behaviour of individuals or groups should include study of the characteristics of the organization in which the behaviour occurs.

Comparative Data and Methods

To satisfy our first requirements, for comparative studies, our initial sample of 52 organizations contained not only manufacturing businesses but also service organizations of various kinds -- government, local authority, retail, and commercial. To begin with, and for convenience, we limited ourselves to work organizations in the Birmingham area employing more than 250 people. A work organization is defined as one that employs (i. e. pays) its members. The products and services represented included motor car bumpers, milk chocolate buttons, insurance policies, road repairing, research, passenger transport, and many others.

We wrote first to the chief executive of the organization, who might be its chairman, area superintendent, works manager, chief officer, or whatever, and we began by interviewing him at length. There followed a series of interviews with department heads of varying status, as many as were necessary to obtain the information we required. Interviews were conducted with standard schedules listing what had to be found out. Since this was descriptive data about the organization and its immediate environment, and was not personal to the respondent, we made no attempt to standardise interview procedure. Wherever possible documentary evidence was sought to substantiate verbal accounts.

Common Standards of Measurement

To satisfy our second requirement, for a common standard, we used the scaling techniques which psychologists have found useful in measuring attitudes or analysing performance on different mental tasks. The major difficulty in this connection is, of course, the demonstration that the items forming a scale "hang together", that is: they are in some sense cumulative. To this end, we used mean item analysis values, generalised biserial coefficients, and random split-half reliabilities. The use of such methods in the study of organizations is relatively new.

Common standards of measurement enable us to record more accurately how variations in one aspect of an organization's functioning are related to variation in other aspects. We can relate environmental differences to internal differences, and so on. For this purpose the statistical techniques of multiple correlation and factor analysis are used. Results so far indicate that the method has some promise in helping us to predict many characteristics of an organization from knowledge of a few really important features.

The Organization's Objectives and Environment

Our third requirement was that the objectives and environment of an organization should be taken into account. To classify objectives we asked a number of questions about the policies adopted by the organization in relation to its products or services. The responses gave us a range of positions on a number of dimensions. For example, did the organization make few or many products, or offer few or many services? Did it offer a standard product or service, offer a choice of modifications, or supply entirely to customer or client specification? Was it aiming at a fixed clientele or named customer, at a limited range of outlets (e. g. manufacturers of engineering tools), or did it supply the public at large? Was the organization aiming to expand, maintain, or contract its range of outputs? And so on. The information obtained enabled us to scale objectives in such terms as multiplicity of outputs, standardization of outputs, client selection, etc.

The environment includes all those aspects which impinge on the organization's functions. Thus, for example, it is possible to construct an environmental scale for location, based on the number of separate operating sites the organization has; or for external memberships based on the number of relevant trade associations, research associations, employer's associations, etc., to which the organization belongs. We have used the word CONTEXT to denote the combination of objectives and environment.

Three of the most important aspects of context which we have measured are size, technology, and dependence. From knowledge of these we can estimate how many internal specialisms an organization is likely to have and how centralised its decision taking is likely to be.


Size is not difficult to measure because there are readymade units of employees or capital which can be counted.


The impact of technology on organization has come to be recognised during the past decade as of prime importance. Starting with the familiar ideas of the differences between unit and small batch, large batch and mass, and process production, and also taking account of the continually increasing mechanisation and automation of production technology, we have found five separate scales for examining the production process. For example, one scale consisted of items measuring the dimension of rigidity -- adaptability. It is shown in Table I. A high score on this scale indicates that the technology is more rigid.


It will be noted that every item in this scale can apply to the work of both manufacturing and non-manufacturing organizations. It is meaningful, on this basis, to speak of the adaptable technology of a shoe repair shop (score 0 out of 8). One can say that the workflow is less rigid than that of motor vehicle assembly (score 7 out of 8), or for that matter, than that of a swimming baths (score 6 out of 8). The shoe repair shop is adaptable because it has multipurpose equipment and can hold buffer stocks, etc., whereas the swimming baths is less adaptable because it has single-purpose equipment, and continuous flow operation.


The degree of dependence of a particular organizational unit on some larger corporate body was derived from a number of separate scales, given in Table II., which were then subjected to factor analysis. A large first factor taking out 55 per cent of the variance was obtained which indicates that these scales are tapping a basic dimension of organizational context.


On this basis a single department of a local authority might be scored as dependent since it may employ only a small proportion of the total number of local authority employees; its chief executive officer does not have a seat on the council; it has branch status; and many of the services it uses may be provided by central service departments, and not by its own personnel.

On the other hand, a firm which is not part of a larger organization at all will be highly independent since it employs 100 per cent of the total employed by the organization, it has complete representation on the controlling body (since its board is the controlling body); it is the only -- and therefore the principal -- unit; and it is self-sufficient as far as specialist services are concerned.

There are many variations between these two extremes of dependence and independence.

Internal Measurements

The internal measurements were constructed in a similar way to the ones already described. Two major areas were examined. These were:

1. the organising of activities, i. e. the degree to which tasks are formalised, specialised, and standardised

2. the centralization of authority -- i. e. the proportion of decisions taken at the upper levels of the hierarchy.

The Organising of Activities: Formalization

We have developed a number of basic measures of the degree to which activities are organised: formalization, specialization, standardization. As an example, we shall show in some detail, how the analysis was carried out for the first of these dimensions: formalization.

Formalization is a major variable of organizational structure, distinguishing how far communications and procedures are written down and filed. Formalization can include both statements of procedures, rules, and roles, and documents involved in their operation; for example, applications, requisitions, agendas, minutes, records and memoranda.

The method employed to measure formalization is in the nature of a count of actual as compared to possible occurrences. Definitions of thirty-eight documents were assembled, which may be used by any work organization: for example, record of direct worker's time, written reports submitted to "workflow" (production) meeting, petty cash voucher. After adding in some instances an assessment of the range of personnel to whom a document applied, a total of fifty-five items were generated. The organization was scored in binary fashion on each item, and the aggregation of positive responses gave the organization's "score" on the overall formalization scale.

A range of scores from 4 to 49 was obtained, and the scale was shown to have a mean item correlation value of 0.63. A principal components analysis of a slightly modified list of 41 documents gave a first factor, overall formalization, accounting for 34 per cent of the variance; there were no other meaningfully distinctive factors.

An attempt was nevertheless made to represent conceptual distinctions within the overall scale and to give information about the orientation of the organization's formalization.


Three sub-scales were created from assemblies in the items in the original scale, concerned with formalization of role-definition, of information passing, and of role-performance recording. The three sub-scales gave mean item correlation values of 0.82, 0.68, and 0.67 respectively, and intercorrelate highly. As an example of the measurement of formalization, a role-definition scale is shown in Table III.


The degree of centralization is another topic which has long interested organization theorists. We asked, for a range of 35 recurrent decisions, who was the most junior person with authority to initiate action. (If a superior merely gave routine confirmation after action, the subordinate had authority. If reference to a superior had to precede action, then the superior had authority.) The job positions of those with authority were related to a standard six-level hierarchy to make comparisons possible across the entire range of organizations studied. The levels are given below.

A high score means highly centralised, and a range has been found from the most decentralised organization with a score of 108 (a large vehicle manufacturer) to an extremely centralised organization, where a considerable majority of the decisions are taken right at the top which has a score of 173 (a small food factory which is part of a large group).


The Results of the Research

We anticipated that the size, technology, and degree of dependence of an organization would affect its internal structure. Many writers have said as much. We have, however, been able to progress beyond this. We have measured how strong the relationship is, so that we can predict the extent to which activities will be more organised (i. e. specialised, standardised and formalised), or authority more concentrated.

The diagrams below show this result graphically. The figures represent scores in relation to a mean score of 50. Two-thirds of the sample have scores on any measure which lie between 35 and 65. Scores outside this range are relatively extreme. To make visual comparison easier, the external and internal measures have been paired so that the higher scores are always farther from the centre of each diagram. The thickness of the bands should not be taken into consideration. We show two instances where the results of our predictions were very close to the actual position, and one instance where the prediction was less accurate. If we could make perfect predictions in every case it would mean that circumstances entirely determine the shape of organizations, and that the views of top managers are irrelevant. This is obviously not so. Perfect prediction would be attained with a correlation of 1.0. In fact, the correlation of the external measures with organising activities is 0.75 and with concentration of authority is 0.81.

Table V

The Measurement of Centralization: Examples of the 35 Recurrent Decisions

Which level in the hierarchy has authority to:

1. decide which suppliers of materials are to be used

2. decide methods of work to be used (not involving expenditure) i. e. how a job is to be done

3. decide which machinery/equipment is to be used

4. decide allocation of work to be done among available workers

5. decide what and how many welfare facilities are to be provided

6. decide the price of the output

7. alter responsibilities/areas of work of functional specialist department

8. alter responsibilities/areas of work of line departments

9. create a new department (functional specialist or line)

10. create a new job (functional specialist or line, of any status, probably signified by a new job title)

The Implications of the Results

It has long been realised that size and technology are important determinants of organization structure. Nor does the relationship between external dependence and internal centralization come as a surprise. What is surprising is the magnitude of the correlation between business context and the way activities are organised and authority concentrated or dispersed. People often speak as if the personalities of the founder and directors of a business had been the most important influence in creating the present organization. Other people point to historical crises, or to the vagaries of government policy as being the stimuli which caused the business to develop in a particular way. Though we would certainly expect personality, events, and policies to play their part, the fact that information relating solely to an organization's size, technology and dependence enables us to make predictions indicates that these aspects of business context have an even greater importance than was previously realised. They set the guidelines within which the management of an organization has to operate.

Group Behaviour

Our fourth requirement given at the beginning of this paper stipulated that the study of the work behaviour of individuals or groups should include study of the characteristics of the organization in which the behaviour occurs. We are therefore designing a research programme which will enable us to relate information on how groups and individuals behave at work to the characteristics of the organization itself. Various advantages and disadvantages have been supposed to accompany a high degree of organization and of centralization. For example, where there is specialization and standardization, do people generally know what is expected of them, but do they also tend to develop sectional interests which may run counter to the overall objectives of the total organization? Where there is centralization are the problems of effective co-ordination more easily solved, but is this at the cost of a loss of initiative on the part of subordinates? The manager will want to know how far internal conflict is generated by faulty organization and what changes in structure may alleviate strife. Or he may wonder whether greater delegation of responsibility will lead to a decrease in apathy at the lower levels of his organization.

At present there are no unequivocal answers to such questions, though much experimental work has been done. For example, there have been laboratory studies of small group behaviour under competitive conditions where victory for one group means defeat for another. There have also been field studies of such things as the relationship between supervisory styles and worker attitudes and performance. What is needed is some way of establishing what are the conditions under which the findings of such researches are more generally applicable. Systematic comparative study relating organizational to group and to individual differences is therefore necessary.

Summary and Conclusions

The manager of the future will have available to him ever increasing amounts of information, and will be anxious to know what signals he should primarily attend to. If he knows what is crucial to organization functioning he can manage by exception. What types and amounts of environmental change can occur before internal adjustments must be made in order to maintain performance? How much internal change will be required, and in what direction? What sequence of compensatory changes is likely to be triggered off by any specific decision? The precepts of the classical management theorists have proved inadequate to deal with such questions as these. The empirical approach to organization theory may yield a more useful set of criteria to help the manager to take important decisions.

The Industrial Administration Research Unit at The University of Aston in Birmingham has developed over 130 major scales for measuring the external context and internal characteristics of organizations as diverse as breweries, banks, car manufacturers, and education authorities. Because numerical scores are used, the position of an organization can be separately plotted for each measure, and a graphic "profile" obtained which can be compared directly with the profiles of other organizations. Closer analysis of atypical organizations may lead to a better understanding of the patterns we have found. At least it should enable the development of future theory to be firmly grounded in fact.

The Industrial Administration Research Unit's programme is more fully described in:

D. S. Pugh, D. J. Hickson, C. R. Hinings, K. M. Macdonald, C. Turner, and T. Lupton: 'A Conceptual Scheme for Organizational Analysis', Administrative Science Quarterly, Vol. 8, No. 3, December 1963.

D. J. Hickson and K. M. Macdonald: 'A scheme for the Empirical Study of Organizational Behaviour The International Journal of Production Research, Vol. 3, No. 1, 1964.

D. J. Hickson and D. S. Pugh: "The Facts about 'Bureaucracy'" The Manager, July 1965.

D. J. Hickson: 'A Convergence in Organization Theory' Administrative Science Quarterly, Vol. 11, No. 2, September 1966.

D. S. Pugh: 'Modern Organization Theory: a Psychological and Sociological Study' Psychological Bulletin, Vol. 66, October 1966, 235--251.

C. R. Hinings, D. S. Pugh, D. J. Hickson, and C. Turner: 'An Approach to the Study of Bureaucracy' Sociology, Vol. 1, No. 1, January 1967, 61--72.

D. S. Pugh, D. J. Hickson, C. R. Hinings, and C. Turner: 'Dimensions of Organization Structure' Administrative Science Quarterly, 1968.
COPYRIGHT 1992 Gabler Verlag
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Organization and Personnel
Author:Pugh, D.S.; Pheysey, Diana C.
Publication:Management International Review
Date:Jan 1, 1992
Previous Article:Organizational theory and organizational practice.
Next Article:The human resources function.

Related Articles
Managing security around the world.
Headquarters, host-culture and organizational influences on HRM policies and practices.
Personnel costs and patterns in libraries.
HR practices differ by company size.
Office of the Under Secretary of Defense: 3000 defense pentagon Washington, D.C. 20301-3000; OCT 17 2003.
Tourism scholarships.
Run for WILPF's U.S. section board! Positions available for self-nomination.
Study to assess value of mixed land use.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters