AUTHORITY, AUTONOMY AND

Dr. Miguel Angel Asomoza Bosque

CENTRALIZATION g&;,,:y IN MEXICAN Rodríguez

Graduate Section E.S.C.A.

ORGANIZATIONS. I.P.N. (National Polytechnic Institute)

The srudy of the structure of an organization show its importance by means of its determiners which are products or services rendered as well as its global effects on society which include servival, development and evolution.

When studying the structure of an organization, the administration theorists divide the structure in two vital parts: 1. Organizational Design and 2. Labor or Work Design.

An organizational desing is an integral model of duly formalized labor relations with specific trans- mission to the areas or departments for the particu- lar tasks which make up each unit of the organiza- tion, providing that they contribute to the attain- ment of predetermined objectives. It is essentially:

Every well-structured organization has duly for- malized standards set out in manuals and routines which facilitate the operation of the diverse de- partments. Organizational charts and rules also help reach such goals. Nevertheless, organizations which have no formalized standards exist and base themeselves simply on skills. acquired by the most senior members of the department or operative section. The skills are passed on without being do- nlm~nrwl and the resulting organi-ration structure

based on either convention or custom.

Between these two extremes rhere is a great variety in the degree or level of the formalizaion oforgani- zational standards.

The reasons for conceptuali;ring and abstracting an organizaion designare numerous. Fintly, there is the need for clear, well defined and functional arder in the organizaion so that a great deal of uncertainty and confusion can be avoided, this sub- sequently makes the work to be done more attrac- tive, safe and efticient and futhers interaction wit- hin and among the working groups. Better han- dling and identification of data systems is another result of precise operation and distribution of channels of cummunicarion. It can be said that an organizational design makes an organization more ordered, predictable and manageable: it also facili- tates organizational learning.

The degree of development and evolution rea- ched by an organizaion over the years is used to judge its leve1 of consolidation or soundness which is reflected in its prosperity orweakness. Prosperity and weakness are largely due to the structure and. especially to the established organizaion and labor design. Organizations adapt their designs to suit circumstances in such a way that if a working table is efficient. it should be respected and kept in use: however, when the upposite happens, its contents should bechecked and ifit is inefficient,itshould be dropped.

An organizational design softens the intense and powerful influente of a person in a supreme mana- gerial position and permiw functional continuity in the areas and department which make up an orga- nization. Besides helping in the definition of the

fields of activity to be developed and in the specifi- cation of courses of action to be taken. it establishes an indispensable level of influente for rhe smooth running of the organization.

Bearingin mind the organization’s projection ina social and administrative context. we can say that it has a clear, precise and even high degree of exce- Ilence when it is capable of solving its problems without manipulating the conflict to the detriment of its members and can reach its objetives with ma- ximum efficiency.

Thus WP cannot disconnect the concepts of an organization’s structure from those corresponding to an organizational and labor design.

We can establish that the wucture of an organi- xation is a model of labor conduct which has been duly regulated for the interrelation of work func- tions that have been indicated and previously esta- blished in the attainment of the organiration’s goals. It is best understood by studyingit in terms of task specialization and the leve1 of coordination en- tailed. This specialization includes a diuision of work for the total of tasks and activities encompas- sed by uniform and manageable units of operation. These units should be duly coordinated ordistribu- ted in areas of speciali-red work within aher highly representative lieldswhich form the essence of the departamental divisions of the organization.

The balance between specializxion and coordi- nation by means of stimulative combinations not only qakes the organization’s structure an opera- tion model, but also the designers’ efforts in the attainment of the goals are objectivized.

From an abstract point of view in administrative science we can say that organizational design is the normative or reglamentary relation between .the most diverse working unitr and the elementswhich make them up in arder to reach an indispensable balance between the leve1 of specializati& and coordinadon. This balance is needed to reach the goals by means of defined conduct which lets the organi7ationschangeand last. Inshort, thismeans:

.4uthority is the official pacer to establish rela- tions between the rules and regulations imposed in an organizttion.

The eupress or euident relation between the components of an organization and its goals.

The ways emplõyees becnme part of the organi- zation, their relationship to their work and how the organization responds to its labor force’s develop- ment needs.

These were the basic reasons why 1 decided to go into the scientific knowledge about authority, cen-

tralization and autonomy in the Mexican labor con- text.

AUTHORITY IN AN ADMINISTRATIVE DE- SIGN

When studying authority as a phenomenon wit- hin the conlines of Administrative Science, and when studyingitas an abstraction of the knowledge of-an organization’s structnre, it is possible to find some bases of the management’s function and fo situate the dimensions of this concept. Thus it is possible to determine the reason for elaborating under what conditions labor relations are establis- hed and also to set ant the nortns for said relations. Furthermore, it is possible to understand the me- chanisms needed to reach the goals and objetives in accordance with a specilic design. It is through aut- hority that employee performance guidelines are set and the use of resources is justified within the objectivization of a model. Authority harmonizes conflicting interests and establishes a consensus of individual and group values. It is here that the dimensions of hierarchic levels. the division of work and superior-subordinate relations become evi- dent.

Brown (1980, 91) sees authority as being able to enforce measures to carry out specilic courses of action in organizations, i.e. an institutionalized form of power. The person in whom this power is invested has the right to arder the fulfillment of tasks or plans to be carried ant in a course of action. Bernard’s definirion, according to Grimes, is that authority is a condition of an arder within a formal organization to which they contribute. The critical difference is implied in the phrase “In virtue of the farr that said arder is accepted” in such a way. that the under!ying rationality of accepting authorlty as an arder m the organiration and the fact that its members voluntarily carry out the orders leads fo the undersranding that the four basic aspects of authority are:

Van de Ven (1980, 121) sees authority as being the right to act and analyze the dimensions given by Weber. He also associates authoritu with “in-

fluente” which denotes the capacity of ene person to make another do something in a way in which it would not have been done. Presthus (1962. 138) and Fillet, House and Kerr (1976, 94) made an imponant contribution by distinguishing the thew retical differences between authority, power and influente and established the thesis that power and influente cwxist. After extensive studies in nume- rous enterprises Van de Ven did not find differen- ces between these variables. Nevertheless. this ob- servation shows the need to increase the dimensions of the Organization Assesment Instruments (OAI) to develop new measuring systems to show the fine distinctions established by other authors. The same author (1980.123) made three basic adaptations to Tannembaum’s,( 1968. 33) control gmph: (1) aut- hority is spread out through hierarchic levels and also horizontally over distinctive operative units: (2) all of the organization’s members were taken into account: (3) authority, as an extension of an exer- cise in an organization, depends on how it is percei- ved by the memben.

AUTONOMY

The differences in personal interest and the fulfi- llment of goals as seen tit with the power to act electively and reflexively when executing an action allow the individual to feel satisfaction; in fact, he feels a mingled sensation of pleasure and superio- tity. This leads to a cycle of self-satisfaction where pleasure is replaced by more status, respect and power.

In the culturaliition process, the human species has had to face the basic pmblem of finding the essence of its freedom and of outlinin~ freedom’s limits as well as having to know the &t relevant forms which chatacterize it. Politicians, philosop- hers, the dergy, sáentists and writers, irrespective of their epoch or geographical location, have used various terminologies ta conceptualize human behavior as a phenomenon. Nowadays, the tenns tend to converre due to circumstances or contin- gencies. Fromm, according to Gross (1964, 327), considers the “Freedom of man” to be the highest and sanest goal in the world. When man finds him- self, his awareness makes kim more productive and his relationship with his environment is more aut- hentic in concrete realitv in a positive and shared . sense.

Man’s knowledge and actions can be separated or isolated into otxrative units which facilitate their

to determine the conditions under which

they are linked together so as to reach maximum organization. Put differently, an attempt is made to reach their significance as cognitive or behavioral elementswhich are representadve ofa culturalcon- text within a systetn of values. The highest leve1 of learning is seen as a growth and metamorphosis process of a person who undergoes changes which make him different in succesive phases provided that each phase is an elevation of knowledge and ethic values.

This representaion only takes place within a framework of freedom since the forms shown by a person over a wide range of specific interests are spontaneous and illustrate growth from something elementary to a complex literary or scientitic work.

The interaction of these groups allows us to un- derstand the diffculty a human being has in stri- ving for freedom; freedom being a free choice of action or poner, elective and reflexive thought and action done out of conviction and without the inter- nal or environmental pressures which usually inter- fere. The interest of these thoughtsis moreevident when the individual is seen within an organization in the light of his need for self-development and creativity which are subject to the normsof a group and the interaction of interests within the group.

CENTRALIZATION

Aiken and Hage (1966. 427) view centralizaion as being the extent ta which paver is concentrated in a social system. From an administrative point of view, maximum centralizaion is when al1 paver is held by one person (OI. a small gmup of people) in an enterptise: inversely, minimal centralization is when power is held equally by all members of an organiration. All organizations, as can be expected, fall somewhere between maximum and minimum centrali7ation.

In arder to measure centralizadon, various sys- tetns can be used. One of these is the concept of “power” which means decision taking, another in- volves delving into the knowledge of power in gene- ral. Both are obtained by meaos of perceptual ins- tmments which reflect the subjectivity of the ans- wers. In Aiken and Hage’s work the dimensions given to centralization were the degree of the parti- cipation in decision taking which has two sub- dimensions: global organi-ration, and the degree of control held by those who carry out the tasks.

RESEARCH MODEL:

The design of this study of the problem of autho-

rity, autnnomy and centralizaion in some impar- tant Mexican oruanization can be summerised as

0

fOll*X~S:

.4uthority as an administrative phenomenon is associated with socialiration and iob satisfaction. -I-he de ree of freedom oï autono&); that a puso” should

f ave ts ltrmted, reduced or changed accor-

dingto the task +gned: especially in mid oì upper kvels of the Or&mni7atiOn structure.

This situation is not conducive to relations which favor a good organizational climate and has reper- cussions on the satisfactory integraion of the pr- duction system. The above mentioned reappears to the dettiment of worker autonomv and freedom when it links the authority with similar or concom- mitant phenomena. Thisstudy intends toshow that the autonomy variable is modified or neutralized in its action by the authority and centralization varia- bles. Furthermore. authority has suflicient power tu make itself evident through the varying comple- xity of organization. The study will be done accor- ding to the following hypotheses:

1. Authority (X) and centralization (2)modifywxk autonomy (Y).

2. The function oi authority (X) can he differentia- ted in the hierarchic lev& of the organization (WI.

The sample group of 123 enterprises to be stu- died was chosen from 500 large companies with assets of over, 2,500 million pesos and a binomial sanple formula of 0.5 probability and 0.9 reliability was used. Each organization ws asked to replay through three of its managers: a total of 322 res- pondants us obtained.

RESULTS

Table live the stîucture of the four factors which are represented by the following variables: autho- rity. socialization, centralization and complexity. These factors are orthogonally distributed and their coefiicicnts. communality and eigenvalues can be studied in depth.

Table six has the three dimensions which make up the authority variable in this study.

Table seven represents the two dimensions which make up socialization.

Table eight has the resultson the reliability of the inrtruments thruugh correlation coefftcients co- rrected by the Spearman-Brown formula with its respective levels of significance.

Table nine shows the resulw of the discriminant analysis tor the tunctlon ot authonty by means oi

the items corresponding to factor number ene (Authority) as well as the second factor (Socializa- tion). Theseare also contrasted with the five hierar- chic levels of companies in accordance with the complexity variable which is being studied. The canonical correlation had a value of 0.22, a Wilks Lambda of 0.95 and chi square value of 22.6 at a aignificance level of .03 for function number ene corresponding to authority. The coefticients for the centroid groups were: (1) (-.25): (2) (-.25): (3) (.07): (4) (.19): (5) 3,57).

Table ten has the prediction results in groups where 25% of the cases were shown to be correctho classified. The chi-square value was 4.7 ata signili- cance level nf .03.

Table eleven illustrates a territorial map which indicates the distribution of the centroid gtwups and their lay out while taking the functions of aut- hority and socializaion into account.

Table twelve is the result of the Path analysis in the study of the three variables shown at causal closure: the standardixd Beta coefficients and the value of the latent variables.

Table thirteen is the bivariant covariation break- down.

Table fourteen shows a final F’ath analysis for the three variables under study.

DISCUSSION

The autonomy UT laboral freedom variable was, with its respective cwfficients and eigenvalue, pre- sent as a factor in the pilot tea. The test was to Iint evaluate the functionability and viability of the live variables being studied. it was for this reason that they have been included in tbe study.

Table Ii\;e shows the factor structure of the four variables: (1) Authority; (2) Socialization: (3) Cen- tralization: (4) Complexity. Autonomy or laboral freedom is not included since this variable was not factorized in the base group (with the varimax met- hod after rotation with Kaiser normalization) and therefore the itemsofthis variablewere not”loaded on a factor” orare not “saturated in a factor”. It was observed that aher factors (1 and 3) were saturared in the final result of the complex factor test (four factors) but thisisnotwhat is beingreferred to here. These factors, 1 and 3, which were saturated by autonomy mere the authority and centralization va- riables whose factorial loads were visibly similar for both factors and for the five items of the autonomy variable.

Due tu what has been stated, authority and cen- tralization, when presenr at the same time, modify

the values which individually make evident the con- tinuous of the autonomy or laboral freedom varia- ble which is proof of the first hypothesis. This fin- ding, in my opinion, should not go unnoticed since it permits us tocontemplate a new approach within the science of administration through a bipolar re- presentaion of the phenomena of authotity and laboral freedom at work. The interplay between the intermediate values will depend on the magnitude or intensity with which one phenomenon domina- tes the values of another. When there is less auto- nomy in work and greater freedom, authority is seen fo have less influente.

From a practica1 point of view of the results ob- tained in this study -given that the study represen& the most select groups of the private sector in Me- xico- we can infer that the powerful action of aut- hority left the lowest values of autonomy for the development of the administrative taskTbis can e~- plain the reason for loading six items on factors 1 and 3 and why autonomy did not factorize.

The statement that authority is shown with “po- werful action” is backed by the results of concrete reality through the alreadyexplained factorizati- on test. The saturation of this factor was done in tb ree dimensions. The first dimension has been “0 - bjetive authority” since it is the object of authority in itself when it is shown ormadeclear, irrespective of its orientation (upwards, downwards and horizon. tal). The ZICU of control are included in this dimensi- on. The second dimension,or “Subjective authority” represents the internalization of authority and is the formation of certain criteria by each person. The third and last dimension is “Utilitarian autho- rity” which is the expression of usefulness in au _ thority.

Having considered the eigenvahtes obtained in the factorial analysis it was thought suitable to create discriminating function by meana of the sta- dstic andysis of the original values of the authority variable (first function of the discriminant analysis). The wcialization variable was a function. The groups in which fo

plied as the second th functions acted

were represented by hierarchic levels of the enter- prise into which the answers to complexity were divided.

The results of the discriminant analysis for the authority function were: a canonic mrrelation coef- ticient of 0.22; Wilks Lambda 0.9s; a chi sqtire 22.60 ata leve1 of significance of 0.13. The centroid groups were adequatly spread out on the map.

The above mentioned results show the maximum ~pr~ssion and dimension of the descriminatory

of authority for the large Mexican enter-

prises where the chain of command or hiera-chic levels cover tifteen levels in the most important branches at operational level. This means that aut- hority as an administrative phenomenon has suffi- cient conceptual strength and rhat it is capable of differentiation through the various hierarchic le- veis which make up the power or command struc- ture of an organizarion. The test proves its value asa continum ladder represented by thisvariable which is called “Administrative authority”, thus it is stu- died, not only asa basic element of the organizatio- nal structure design. but alsoasan essential compo- nent of work design and redesign. 1 have not found the later described or observed by the organization or administraion theoreticians: nevertheless, it is important since it can modify the structural model of both designs in their integral conceptualization.

Table eleven shows the horizontal axis of the territorial map, the function of authority and the magnitude of its behavior in relation to the centroid groups. The vertical axis represents the discrimina- tory function of socialization.

Authority as a theoretic body and vital principie of administration practice is the meeting point of interdepartmenml interaction in an organization and the site where decisions for the attainment of goals and objectives should be made. However, authority is also respect and it is here where its relation to administrative freedom is best. Autho- rity should be in agreement with the fundamental convictions. codes, beliefs and habits of the indivi- duals and action groups. It is for this reason whoe- ver exercises power always does so in a limited way. Traditionally authority has been thought of as lxing without limits and the fact that it has no limits has often made it an instrument of corruption. Over the centuries man has imposed limits on the actions of those who have authoritv for this verv raso”.

The second hi pothesis of this paper which states that: “The function of authoritv fX1 can be diffe- rentiated in the various hiera& levels of organi- mtions (W)” wasshown with the first discriminatorv function (authority). which was split into hierarchi; level groups of the organizational structure of the sample (table seven. central group distribution).

On evaluating the findings of this research paper. it is desirable to delve into the study of the authority and centralization variables. For this reason the causal closure for the three variables under study was designed. This was done by using the Path analysis. Table twclve shows its findings.

Table thirteen gives the bivariant covariation breakdown data which is germine or causal between

43

authority (X3) and socializaion (Xz). The covariant between authority (X3) and centrali-ration (XI) is two thirds direct causal in origin only one third indirect: however, the covariation is, in its totality, of causal or genuine origin.

The relation between socialization (X2) and cen- traliration (X3) is broken down into causal and spu- rious. The bivariant causal closure relation between (X2) and (X1) shows tbat 85% is of causal origen and 15% is spurious.

The analysis of Cij coefficient afiirms that the causal closure between authority (X3) and centrali- zation (X1) is undoubtedly of causal origen. This concept is in agreement with the theoretic context and with administradve practice.

Socialization (X2) and centralizadon (Xi) have a double relation which is on ene band of causal mi- gin, and on the other. a spuriou+ relation.

This model is in agreement with the reality of theory-practice and shows that while some kinds of socialiration facilitate the phenomenon of centrali- ration, sorne of its other dimensions are detinately incompatible.

For a better understanding of the negative aGo- ciation of the causal origen between authority and centraliration it is advisable to study E. Resenos’ concept given be&:

“Authority is a speciftcation of the function’s hie- rarchic scope. Tbe more the speciticati& the func- tions, the greater tbe integrarion of authority and the less the tendancy to centralize.”

Table fourteen shows the final model of the Path analysis in accordance with the final results of the causalogic closure as found in this study.

CONCLIJSIONS

1. The autonomy variable did not factorize, Le. its items mere not saturated by ene factory only. The saturaion variable carne into effect in the factors corresponding ta the authority and cen- tralhation variables. Tbis leads to the conclusion that these two variable modify the values of au- tonomy or labor freedom.

2. Authority and autonomy can be represented bi- polarly where tbeir values will have an indirect relation, i.e. the greater, the authority, the less work autonomy, and vice versa.

3. Authority is a powerful phenomenon in its theo- retical and practica1 conceptualization in the field of administrative science and has tbree di-

mension: Objetive authority Subjective authority Utilidarian authority

4. It was shown through the factorial and discrimi- nant analysis that administradve authority is a “discriminatory fimction” before the hierarchic levels of an organization.

5. The causal closure between authority and socia- lintion was achieved by meaos of the Path analy- sis. The breakdown of the bivatiant covaration between authotity and centralization wasofcau- sal or genuine origen and the relation between socialimtion and centraliration was 85% causal and 15% spurious.

6. The socialiration variable is presented in this study a’s having a bipolar nature: at ene pole there are the superficial relationships between people, and at the other, their feelings and affec- tations.

7. It is shown in the study that the ptivate sector. according to the sample obtained. is putting ad- ministrative decentralization into effect.

SUMMARY

The study deiines its basic objects and purpose which are to find the relationship between autho- rity. autonomy, centralization. socialiration and complexity variables.

The sample group is made up of “gigantic and large” enterprises of rhe private sector with 322 responden taken from high levels of manage- ment.

Theoretic concepts on authority and autonomy and their application to administrative practice are evaluated.

The analysis if the answers in the empiric test was done by applying factor analysis, discriminaúon and the Path analysis. The results obtained are commented on and discussed by interpreting the findings within the theoretic context ofadministra- tive science.

The conclusions are made and further reading is given in the bibliography.

44

List of Tables

Table No. 1 Analysis of Organizational De@ Variables

Table No. 2 Variable Intmaction Graph Table No. 9 Bipolar Structure of the Relations

Between Authority and Autonomy Table No. 4 Roposed Theoretic Model Table No. 5 Structure of Four Factors Table No. 6 Three Dimensional Structure of

Authority Table No. 7 Socialization Stmcture in Two

Dimensions Table No. 8 Instmment Reliabilitv Table No. 9 Analvsis of Authorit; and

Hier+rchic Lev& Table No. 10 Group Prediction R&ults Tahle No. ll Territorial Map of Discriminatory

FUtlCtiOns Tahle No. 12 Path Analysis Results for Thrm

Variables Table No. 13 Bivariant Covanation Breakdown Tahle No. 14 Path Analysis: Final Model

REFINENCES

Asomora, 8.. M. A. Validmión de la Escala Inventario de Caracte*ticm del Trabajo (ICl) en el Media Mexicano. Renisla Serie Trabajos de Invertigación, I.P.N., 1978

Asomm, B., M. A. Innouacti, Tecnología y Complejidad en un CmpO de Orgmiraciones Me.&ana.x C.I.C.A. No. 7, E.S.C.A., I.P.N., 1980

Asmza, B., M. A. Alineación, So&liirac&Ín y Autmreoliza&n en el Au- se* Laboral. C.I.C.A. No. 10, E.S.C.A., I.P.N., 1980.

Asomoza, B., M. A. Lo S&liración como UM Caracterí&a del Trabaio y su Asociación Negaliva en Grupos de Profesionistas dentro de las Organizaciones. C.I.C.A. No. II, E.S.C.A., I.P.N., 1981

Binu, P., Schomhen; R. The Slructure on Organizatimu; BnSnc Books, Iru., Pubiishers, 16, 1971.

Biuu. Peter M., mui W. Richard Scott 01 OrgmGzarionr (San Fmncisco: Chondler,

Brown, W.B. Organiration Throry and Mamgemnt Ed. Joha Wile) and sons, Inc. 1980.

Bucklq, Walter Sociolo~ and Modmn Sytm Tkeov (Englmood Ci#s, N. J. Prentice Hall, 1967)

Evan, WiU& M. Indica ofHiemrchica1 Stmcture of Indwtil Orga- nirahns, Management

Filley, Alan C., Robert T. House, and Steve Kerr Manapial Process and Organiurtional Behauior, 2nd. Ed. (Glenview, Illinois: Scott Foreman, 1976)

Gamwn, William A. Power md Dixontent (Homewood, III. Doney Press 1968)

Authority, Powr, In$luence and Social Control: A Thewical Synthesti, Academy of Mmagement Re- uiew, Oct., 1978.

Gmx, B.M. The Managing of Orgoniralions, The‘Free Press of Glmoe, 1964.

Hackman, J. R. and LmLer, E.E. Employee Reactiom to Job Chnract&tics, Journa! of A@ed Psgchology, 55, 1971.

Homy, Karen Our Inner Con-flicts, Nm York: Norton, 1955

Kaufmann, Hany Psicología Social, Editorial Interame?i~am, Méxim, 1977.

Masiow, A.H. Motiuo&on and Personalit7, Neu. York: Harper 1954

Nk, Norman H., Hull, C, Hadlni, Jenhim, Jean G. Stdislicol Package for the Social Sciences (SPSS), Ed. Me. Graw HiU, Inc. 1975.

Presth~uus, Robert V. The Orgmizationnl Society, New York: Knopf, 1962

Pugh, D.S., Hickson, D.J. A Conceptual Scheme for Organizationat Anaiyti,

45

FIGURE IV’ 1

VARIARLES IN ORGANIZATION DESIGN

AUTHORITY POWER

TASK DIVISION

L CENTRALIZATION

I - DEPARTMENTATION

Functional departmentation C Purpose departmentation Mattix ~tructut-es

COORDINATION l-b RULES, REGULATIONS AND PROCEDURES AND CONTROL PATTERNS

L

4

SPAN OF CONTROL

______, STANDARIZATION

-----------------, ,’ ADAPTIVE

I

INNOVATIVE , ‘, -D~SIGNS DESIGNS 1

-w-s- --------’

JOB DESIGN

EFFICIENCY

Invcrtigaci6n Administradva

FIGURE No 2

VARIABLES INTERACTION IN TASK DESIGN

Deckion making

JOB DEPTH

FIGURE No 3

BIPOLAR STRUCTURE OF THE INDIRECT RELATIONSHIP AMONG AUTHORITY AND AUTONOMY

AUTHORITY

AUTONOMY

RELATIONSHIP AS DEFINED BY MANAGEMENT THEORIES

ALJTHORITY

AUTONOMY

RELATIONSHIP FOUND OUT BY THE AUTHOR’S MODEL

FIGURE No 4

PROPOSED TECHNICAL MODEL

r-r Authority

Autonomy -

ORGANIZATION DESIGN JOB DESIGN

TASK CHARACTERISTICS

- Job depth

Centralizaúon

I

- Job scope

GURE RTO 5

FACTOR!3 COMMUNALITY MEAN STD DEV

FACTOR 1

02 ~ .42 ~

os - .52 -

09 .42 -

g ll- 31 -C-J

2 14- .50 -‘-)

8

;:,

.41 __

5 37 -

4 23 .33 --6)

26 ~ .56 __

27 - .45 -

29 ~ 31 ~

.22 -

.oo -

.25 -

.02 -

.05 -

.12 -

.03 -

.09 -

.06 __

.12 -6)

.05 ~

.21 -

32

.Ol -‘-)

.14

.08 -

.16 -(-)

.12

.25 -

.19

.04 ---c-j

.14 -

.13 -

.26 -

.IS -

.02 ~

.08 ~

.Ol -

.17 -

.16 -

.lO -

.13 -

.05 -

.29 -

.43

.26 -

.12 -

.28 ~

.21 -

.36 __

.21 ~

.36 __

.23

.12

3.6 -

3.4 ~

4.2 ~

3.8 ~

3.4 __

4.2 -

3.7 ~

3.7 -

3.6 -

3.8 -

3.7 __

.90

.98

.73

.84

.91

.64

.80

.78

.85

.74

.84

51

‘CURE No 5.2 N

ITEMi FACTORS COMMUNALITY MEAN STD DEV

FACTOR IV

k R

;y- 42- 4:;

.09 -(-)

::2 .22

-(-) .07 - 31 - .lS 2.1 1.3

3 1;; - __ - - .02 .04 - - .56 .70 - - 27 3.4 3.0 1.7 1.3 -(-) .Ol - .15 __ .02 - .60 __ .si 2.2 1.4

FACTOR EINGENVALUE PCT OF VAR CUM PCT

1 6.3 50.9 50.9 II 2.5 20.4 71.3

III 2.0 16.8 88.1 IV 1.4 11.9 100.

lIwm@ihAdmin 53

FIGURE No 6

Thme Dimensions of auihority

ITEMS FACTORS COMMUNALITY MEAN STD DEV

FACTOR 1 1 II III

02 - .50 - .ll .25 - .28 3.6 90 08 - .64 b .18 .12 - ,45 3.4 .98 14 - 37 - .28 23 - .26 3.4 .91 20 - .66 .13 .ll - .47 3.7 .80 26 - .59 .03 36 - .47 3.6 .85

FACTOR II

ll - .03 .58 .ll - .35 3.8 .84 .13 - .3? 4.2 .64

-(-) .12- .25 3.7. .78 .19- .16 3.7 .84

i; - - .13 .16 .07 .18 .61 .47 - 4.2 .73 -

.40 .28 3.8 .74

FACTOR EIGENVALUE PCT OF VAR CUM PCT

1 2.71 69.5 69.5

II .65 16.7 86.3

III .53 13.7 lOO.-

FIGURE No 7

Two dimensions of kuxial~o~

ITEMS FACTORS COMMUNALITY MEAN STD DEV

FACTOR 1 1 II

19 .71 44 .50 3.9 .90

24 .29 .ll 0.9 3.9 .89

35 50 .43 .55 3.9 .89

39 .62 .29 .46 4.0 .81

FACTOR II

01 .24 .56 .37 3.2 .95

45 .17 .45 .23 3.2 1.01

31 .03 .51 .26 3.3 1.03

38 .19 .37 .18 3.9 .90

FACTOR EIGENVALUE PCT OF VAR CUM PCT 1 2.54 83.7 83.7

II .49 16.3 lOO.-

FIGURE No 8

BELIABILITY

Correlation coefficients (Spearman- Brown)

AUTHOBITY 191 s = .OOl

SOCIALIZATION .94 s = .OOl

CENTBALIZATION _ 84 s q .OOl

COMPLEXITY _ 85 s = .OOl

XJRE No 9 i;

--scriminant Analysis

Authority and hierarchichal levels (complexity)

DIRECT SOLUTION METHODOLOGY

PRIOR PROBABILITIES - EQUAL

NUMBER REMOVED

0

:

GROUP 1

.20

EICENVALUE

.05

.02 Do

GROUP 2 GROUP 3

20 .20

CANONICAL PERCENT WILKS C~RRRU;PION OF TRACE LAMBDA

.22 70.0 .93

.lS .04 26.8 3.2 .98 99

GROUP 4 GROUP 5

.PO .20

CHI-SQUARB

22.6 6.8 .7

D.F. SIGNIFICANCE

12 -03 6 3s 2 69

STANDAR DIZED DISCRIMINANT FUNCTION COEFICIENTS 1 2 3

V. Authority V. Sucialization V. Centralization

.64 .73 .40 -.52 -.41 .82 8

74 .64 33 ? 5 0

57

FIGURE 9.1

CENTROIDES OF GROUPS IN REDUCED SPACE.

GROUP 1

GROUP 2

GROUP s

CROUP 4

GROUP 5

-.05 -.OS .02

-.25 .09 -.o‘l

.07 -.Ol .lO

19 -.40 .-06

.57 .19 -.03

FIGURE No 10

ACTUAL GROUP No of Cases PREDICTED GROUP MEMBERSHIP

CODE Group 1 Group 2 Group 3 Group 4 Group

Group 1

Group 2

Group 3

Group 4

Group 5

1 141 1. 3 PCT

2 86

61. 18. 18.9 PCT 5.6 PCT

2. 47. .6 PCT 14.6 PCT

3 38 1. 12. 23 PCT 3.7

0 6. 0 PCT 1.9 PCT

0 9. 0 0 PCT 2.8 PCT 0 PCT

4 25

5 32

24.8 Perceht of “known” groups correctly classified CHI - Square = 4.724 significance = éO30

5. 1.6 PCT

6. 1.9 PCT

4. 1.2 PCT

31. 30. 9.6 PCT 9.3 PCT

16. 16. 5.0 PCT 5.0 PCT

9. 2.8 PCT

11. 3.4 PCT

8. 2.5 PCT

10. 3.1 PCT

4. 1.2 PCT

15. 4.7 PCT

58 Enerc-Marm 1984

FIGURE IV’ ll

PLOT OF DISCRIMINANT SCORE 1 (HORIZONTAL) VS DISCRIMINANT SCORE 2 (VERTICAL)

4 INDICATES A CENTROID GROUP

3wh 2 2 2 2 2 2 2 2 2 25 5 5 5 5 5 5 5 5 5 5 5 5

2 25 5

25

2,250- ;

5

25 5

2 25

5

2 25

5 1,500

2 95

5

2 25 5

2 25 5 0.750- ‘5 2

6

5

2 5 + 5 2 35

0 - 2 $21+1 5 Ib

5

2 21 13 95 5

2 21 l6 -v5 45

5

2 21 14 214 5

-0.75& 45

2 24 45

24 45 I 5

2 24 45 45

5 24

-1.500-, * 24 45

45 5.

2 24 45 4 24

2 24 4 24

-2,250- ’ 24 4

4 4

4 4

-3.000 444444444444444444444444

-3.000 -2.250 -1.500 -0.750 0 0.750 1.500 2.250 3.000

FIGURE No. 1.2

Results thought out Path analysis

showing the variables under study.

AUTHORITY X3 W) p 23 pp SOCIALIZATION Xz

(-. 16) P 12

CENTRALIZATION XI fg \/1;24=.81

FIGURE No. 13

Descomwsition of Bivariate Covariation

(A) ORIGINAL COVARIATION = rij

(B) bl = Causal-dimt

b4=Causal-indirect

Total causal=(bl) + (b2) = cij

(C) NON causal= (A) - (B) = rij-cij

(x2, x3) (x1, x3) (x1, x2)

.28 -.18

28 -.13 -.16

0 -.05 0

2s -.18 -.16

0 0 -.03

60 EnPro-Mano 1983

FIGURE No. 14

Final model of the Path analysis

0 De las suscripciones, intercambios y donaciones de Investigación Administrativa: Toda orden de suscripci6n y correspondencia relativa a sutipciones deberi ser remitida al Apartado Postal 4025, Sta. Mana la Ribera, Lkleg. Cuauhtémoc, 06400 México, D.F., que es la dirrrci6n establecida para recibir este tipo de notilicaci”“es. No se aceptan swctipciones por períodos mayores a un año, requrikndose los pagos por adelantado, en moneda nacional para el psis, remitidos a Investigacidn Administrativa en cheque 0 giro postal únicatuente. No entie efectivo. las auacripcioues del extranjero deberán pagarse en d6lares (“Sc), remitiendo Ias brdcner de pago a Investigacüm Administrativa. En el precio anual esta incluido el porte correspondiente para IU envio por ti atrea. El precio de la suscripción. asi corno de los mímeros sueltos o atrasados SC especifica en la página 2 de todos los ntímemî, estando sujeta a cambios sin previo aviso. Toda solicitud de intercambio o canje de publicaciones será decidida por cl Conn+ Editorial, asi como el periodo por el cual se establezca Er requisito indispensable para solicitar intercambio de publicaciones que se ctwle por lo menos un ejemplar actual de la publicacidn ofrtida en arije. para que el Consejo Editotil evahie su contenido e importancia. así como su periodicidad. Una vez establecido el intercambio. eSte podra renovarse

l De nuestros simbolos

WVESTIGACION EN CIRNCIAS ADMINISTRATIVAS

ADMINISTRACION PUBLICA

ADMINISTRACION DE NEGOCIOS

ADMINISTRACION DE LA RDUCACION

l De los ammcios.:

Para anuncios cn blanco y negro. cl anunciante enviará negativo y prueba del mkmo.

Tarifa de los anuncios pzw ejemplar dc apmicibn:

4 tinmí 1-2 números 3 4 nútuem 4~. de forros 4o.ooaoo 35.030.00

Blauco y negro 30,000.00 28,000.00

Sa. de forros p8gi”a interior 25,ooo.oo 23.ooO.000 112 ,+ina horizontal 15,00i.W 13.ooo.00

I cohmua 15,ooo.oa 13,cmo.oo 1/2 rol”mna 10,000.00 8,OM).itO

Laz f-has de cierre de los anuncios son 30 diti previos a la aparicibn de cada número: 31 de Marro. 30 de Junio, 30 de Septiembre y SI dc Diriembre. De Iãs corsideraciones de pago: la presentacibn de facturas se hati al publicarse el anuncio. Se otorgaá un 15% de descuento por pronto pago (30dias). Lascancelacionesse~naceptadassisepresentanantesdelas fechas de cierre para cada ntimen,. Cudquicr cambio en el material entregado por el diente se facturarA como Servicio Aditional. Al ordenarse originales se facturará por la elaboración de los mismos, previa autorizacicm del cliente. No se acepta responsabilidad alguna por errores en los números clave de identifieáón cn los anuncios. El anunciante yfo JU agencia de publicidad indemnizati y liberati a Investigación Administrativa de cualquier redarnación. queja o cargo de nivelación de derechos reservadoí. plagio, derecho de prioridad o cualquier otro cargo tasado en el coontenido o mncepto de los ammcio~.