Management and Cognitive Hierarchies: What Is the Role of Management Information

Systems?
Authors(s): John A. Martin and E. Sam Overman
Source: Public Productivity Review, Vol. 11, No. 4 (Summer, 1988), pp. 69-84
Published by: Taylor & Francis, Ltd.
Stable URL: http://www.jstor.org/stable/3379904
Accessed: 28-03-2016 06:36 UTC
REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.org/stable/3379904?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.

Taylor & Francis, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to Public
Productivity Review

http://www.jstor.org

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

Management and Cognitive

Hierarchies: What Is the Role of

Management Information Systems?

John A. Martin, E. Sam Overman

To be successful, the management information system must

integrate information needs, as well as cognitive expectations,

with management activities and behavior.

By now it is hardly a secret that computing and its more sophisticated

offspring, the management information system (MIS) often have not ful-

filled their promise of being the "deus ex machina that transforms gov?

ernment organizations" (Kraemer and King, 1986, p. 494). The reasons

offered for their inadequacy are numerous and varied. They include every-

thing from poor data and inappropriate technology to entrenched policy-

maker values, bad timing, reluctant users, faulty organizational design

and structure, and even the weakness of frameworks for MIS research

(Bozeman and Bretschneider, 1986; Banks and Rossini, 1987).

These reasons alone, however, do not explain why management

information systems have not met expectations. We argue here that the

origins of MIS inadequacy lie in their failure to explicitly integrate infor?

mation needs and cognitive expectations with the management activities

and behaviors present in organizations. Information and management

are not yet a single coherent system. Our presentation begins with a

theoretical description of the management and cognitive hierarchies.

Then, using two case studies from a public organization, we examine the

proposition that for the MIS to be successful, there must be a balance

between cognitive expectations and managerial use. We conclude our

presentation by suggesting how an MIS can facilitate this balance and

thus help ensure its own success.

Management and Cognitive Hierarchies:

Theories and Frameworks

Today, the notion of hierarchy as the primary structuring force within

organizations may be obsolete or in its "twilight" (Cleveland, 1985). A

PUBLIC PRODUCTIVITY REVIEW, VOL. XI, NO. 4, SUMMER 1988 69

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

70_ Martin, Overman

hierarchy may be rigid or robust (Bolman and Deal, 1984; Yates, 1985).

Its persona may differ considerably among organizations (Schein, 1985;

Frost and others, 1985). Yet, despite its many and perhaps new forms,

hierarchy remains a fundamental attribute of organizations that must be

accommodated by different types of managerial activity. Often these levels

of management activity correspond to levels of hierarchy. Stout (1980),

for example, emphasizes control as an appropriate management response

to the routine, operational, lower-level tasks within the organization,

while management is appropriate for less routine, higher-level organiza?

tional activities. Similarly, Anthony (1965) describes three levels of man?

agement activity: strategic, tactical, and operational. In our framework,

these notions of management activity corresponding to levels of hier?

archy have been merged to form the three-tiered management hierarchy

presented in Figure 1. The first level focuses on control and operations,

the second on decision making or management, and the third on orga?

nizational action and change.

The management hierarchy emphasizes the division of organi?

zations into subunits. It stresses the distribution or parceling out of

"things," such as responsibility, power, status, compensation, or author?

ity, among "units" (for example, positions and divisions) within the

organization (Anthony, 1965). The management hierarchy also empha?

sizes ordering or structuring organizations along variations in quantity?

more authority, responsibility, power, compensation, and so on. "Varying

levels in the hierarchy correspond to variations in authority, with each

higher office carrying more decisional responsibility than lower ones"

(Graham and Hays, 1986, p. 74).

Organizations also include a second, cognitive hierarchy. Building

from contemporary conceptions of cognitive science, we see diis cognitive

hierarchy as focusing on that part of organizational activity concerned

with the nature of knowledge widiin the organization, "its components,

its sources, its development, and its deployment" (Gardner, 1987, pp. 6-8).

Four levels form the cognitive hierarchy (see Figure 2). The first

focuses on data, the second on information, the third on knowledge, and

the fourth on wisdom. Note that in our conception, information, the

focal point of much MIS thinking, is but one part of the larger hierarchy

of cognitive activity.

The cognitive hierarchy emphasizes that the cognitive capacities

and requirements of an organization exist at several levels. It also stresses

integration; that is, it emphasizes the combining of elements to form a

comprehensive whole. It stresses that to promote efficiency, organizational

tasks and functions may be divided and assigned to units, positions, and

individuals, yet long-term organizational effectiveness requires a common

or corporate direction. The cognitive hierarchy also recognizes the impor?

tance of context. Information needs to be made useful. Information with-

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

Management and Cognitive Hierarchies 71

Figure 1. The Management Hierarchy

Level of

complexity Explanation of Level

High

Strategic change focuses on facilitating new or fundamentally differ?

ent organizational activity, direction, purpose, procedure, structure,

image, or orientation. Managers are part of organizational transfor?

mation (Brunsson, 1986, pp. 2-10).

Decision making/management focuses on the nonroutine, the discre?

tionary activities within organization. Managers are confronted by

uncertainty but still must make choices. Often planning and evalua?

tion help them make choices.

Control/operations focuses on the routine "maintenance" functions

within organizations, such as "letting contracts, paying bills, issuing

voter notices, keeping accounts, making up payroll, and keeping

track of taxes" (King and Kraemer, 1984, p. 103). Here, managers can

largely determine events and outcomes. They manipulate and control

the "technical core" (Stout, 1980, pp. 1-17).

Low

Figure 2. The Cognitive Hierarchy

Level of

Complexity Explanation of Level

High

Wisdom is understanding that can be used to guide appropriate

action. Theory and its use as an integrator of information are espe?

cially important components of wisdom. "Wisdom is integrated

knowledge, information made superuseful by theory, which relates

bits and fields of knowledge to each other, which in turn enables me

to use the knowledge to do something" (Cleveland, 1985, pp. 21-23).

Knowledge/understanding is organized information that is "inter-

preted, processed according to a point of view, preparing the receiver

for appropriate action" (Kochen, 1975, p. 5). It is information that

has been given a context, internalized by the receiver, and integrated

with experience, study, or intuition.

Information is organized data made useful to somebody, but not nec-

essarily to the producer. Information connotes removal of uncertainty.

It need not convey meaning to the receiver. It merely removes doubt

about which message, out of a set of coded messages, is likely to have

been sent.

Data are undigested or unvarnished facts that are available to be

organized.

Low

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

72 Martin, Overman

out the context provided by interpretation and theory?that is, informa?

tion without a point of view?is not all that useful. Information needs to

become knowledge and ultimately wisdom.

The cognitive hierarchy, in addition, is clearly cumulative. Wis?

dom cannot occur without understanding, just as there cannot be under?

standing without information. Moreover, as is not necessarily the case in

the management hierarchy, value judgments explicitly are an important

part of the cognitive hiearchy. Wisdom is understanding that guides

appropriate action. Wisdom requires distinguishing the appropriate from

the inappropriate, the proper from the improper, the right from the

wrong.

Together, the management and cognitive hierarchies form an essen?

tial organizational tension: More often than not, there is an imbalance

between managerial activity and cognitive expectations. For example, an

MIS that provides data cannot by itself be used for strategic change.

Similarly, but less frequently, a sophisticated computer information sys?

tem, such as an expert system, is inappropriate for organizational control.

The optimal role for a management information system in this concep-

tion of organization is to link or bridge the management and cognitive

hierarchies in the organization. The MIS role within an organization is

to integrate managerial activity with cognitive tasks. As we will show in

detail, the successful MIS should facilitate a synthesis of the two hier?

archies. Thus, management information systems should encompass the

interactions between the cognitive and management hierarchies, as well

as interactions among levels within each hierarchy.

This conception of organization suggests a role for the MIS some-

what different from those roles associated with other frameworks. Most

views of the role of the MIS do not explicitly acknowledge both the

complexity of the cognitive hierarchy and the need for it to be coordi-

nated with the management hierarchy. Instead, frameworks stress the

integration of management and "type of problem" (Gorry and Scott-

Morton, 1971). In stressing problem type, they greatly understate the

complexity of the cognitive hierarchy and its multiple components (see

Figure 3).

More frequently, other conceptions of the MIS role emphasize rela?

tionships among actors (top management, users, designers) within each

hierarchy in dealing with the relationships among a few elements across

the lower levels of the two hierarchies. Legions of technocrats, for exam?

ple, have focused on how technology can process data more efficiently to

generate better information. They stress the relationship between data

and information. Adherents of the policy sciences have been especially

concerned about management information systems and the relationships

between information and control, and between information and decision

making, (King and Kraemer, 1984, pp. 102-104; Hadden, 1986). A few

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

Management and Cognitive Hierarchies 73

Figure 3. Organizational Wisdom as a Synthesis of Hierarchies

Cognitive/

Management Hierarchy Information Hierarchy

Strategic change Wisdom

Decision making/management Knowledge/understanding

Control/operations Information

Data

policy scientists have also entered the heady realm of the MIS role in

the relationship between knowledge and decision making, examining

how it differs from the relationships between information and control,

and between information and decision making:

An MIS can be described as an interconnected set of proce?

dures and mechanisms for data accumulation, storage, and

retrieval, which is designed to convert organizational data

into information appropriate for managerial decision mak?

ing. Management information systems generally summarize

data produced by transaction-based systems (data on clients,

facilities, employees, salaries, services provided, inventories,

etc.) and stored in organizational databases for analyses by

operational, middle, and upper level management and sup?

port staff. . . . [In contrast, a decision support system] is an

interactive computer-based system that is structured around

analytic decision models and a specialized management

database directly accessible to managers, that can be used to

assist management at all levels of an organization with

decisions about unstructured and unroutinized problems

[Rubin, 1986, p. 541; see also Sprague and Carlson, 1982].

The relationships among information, understanding, and wis?

dom have been the special concern of cognitive scientists in psychology,

linguistics, anthropology, and neuroscience, but the potential role for

management information systems in these two relationships has yet to be

fully developed. For now, the most important message offered by cogni?

tive science to MIS designers simply may be that relationships among

levels in the cognitive hierarchy are less linear, less mechanistic, and less

easily quantified than was once believed (Gardner, 1987).

Management information systems' potential role in the relation?

ship between decision making and strategic change within the man?

agement hierarchy is also not very clear. Brunsson (1986) indicates that

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

74 Martin, Overman

in part the management information role in this relationship may be

unclear because "change" does not follow the "reasonable" decision pro?

cess implied in models of "rational decision making" (see also Starling,

1982, pp. 210-233). Contrary to rational decision-making dogma, in real

life, alternatives are not routinely sought, let alone evaluated, and conse?

quences are not assessed. Instead, change occurs, period, or change occurs

and is justified after the fact. Often it is justified with the language of

rational decision making (Brunsson, 1986, pp. 21-24). Change, then, the

pinnacle of the management hierarchy, frequently does not require the

so-called essential information that MIS designers attempt to produce,

but it certainly requires some greater level of knowledge and wisdom.

It is not that these other views of the MIS organizational role are

wrong; but, because they focus on only a few relationships within and

between the two organizational hierarchies, they are simply incomplete.

They rarely explicitly address the general relationship between the two

hierarchies or the role the MIS might play in this relationship. In addi-

tion, contemporary views of management information systems tend to be

too centered on technology. Machine capabilities?what computers and

their software can and cannot do?routinely define the limits (Hadden,

1986, pp. 577-578).

In contrast to these approaches, our conception suggests that the

optimal organizational role of the MIS may be to integrate, explicitly,

the management and cognitive hierarchies. Ideally, management infor?

mation systems should fully inculcate knowledge and understanding to

facilitate meaningful organizational activity, as well as facilitate (as sug?

gested by other frameworks) the smooth interactions among elements in

the two hierarchies. Organizational wisdom, then, encompasses (1) the

general relationship between the two hierarchies, (2) the specific rela?

tionships among levels within each of the two hierarchies, and (3) the

specific relationships among elements between each hierarchy.

Examining the MIS and Organizational Hierarchies:

Context and Study Methods

Interaction between the organizational hierarchies, the importance of

working toward organizational wisdom, and the potential role of the

MIS in supporting organizational wisdom are revealed in what went

on during the development of two public management information sys?

tems. The first MIS case examined in this study was an automated cross-

jurisdictional wage and salary analysis system. The second case was a

gender-focused affirmative action MIS for monitoring city government

performance. The two cases are described and analyzed in this section;

both descriptions use the general framework presented in the previous

section, especially the material presented in Figures 1 and 2.

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

Management and Cognitive Hierarchies 75

Case 1: Determining Wages and Salaries

Automating an existing information system was the focus of the wage

and salary project. For well over a decade, before salary negotiations with

employee bargaining units, city personnel department staff had compared

wages and salaries with those in neighboring and other comparable juris?

dictions throughout the state. The end product of their efforts was a

series of tables, which followed the format presented in Figure 4, provid?

ing baseline information for wage and salary negotiations. All of the

data compilation, storage, retrieval, and analysis was done without the

help of computers; it was done by hand. Personnel department staff even-

tually felt that the task of preparing the tables had become too complex.

Each year new types of jobs, and often new forms of compensation, such

as dental insurance, were being added. The time required for the wage

and salary analysis was increasing. Also increasing were fears that greater

opportunities for data and computational errors might result in inaccu-

rate information being presented during salary negotiations. Computers,

it was felt, could dramatically increase both the speed and accuracy of the

wage and salary analysis.

Essentially, then, MIS designers were asked to make an existing

system more efficient. Increased efficiency was to be achieved by conver-

sion from a manual to an automated information system. Both before

and after automation, project data were "facts" about salary and fringe

benefits for forty-six positions, collected from fifty-seven local govern?

ments by use of a standardized survey instrument. These data were con?

verted to information when they were organized into tables displaying

comparable compensation data for comparable jurisdictions. In turn,

this information became knowledge and understanding as it was inter-

preted from the point of view of personnel department officials. Finally,

knowledge and understanding became wisdom, once used by personnel

staff and bargaining unit representatives in contract negotiations.

Case 2: Examining Women's Organizational Status

Creating a new management information system, rather than simply

automating an existing manual system, was the focus of the gender

project. Once again, however, as with the wage and salary project,

demand for increased efficiency was the catalyst for the gender MIS

project. The investigators for a city government-sponsored study of

working women's status spent weeks assembling information about

municipal employees. Some city officials, especially those from the

social services, felt strongly that gender-focused employee information

should be readily available. They lobbied for a gender-focused affirma?

tive action MIS.

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

(Example:

Planner II)

Average for comparison

employers

Our city figure

aTables were prepared for minimum, maximum, and mean salary levels for each position.

bCompensation elements could include dollar amounts for any combination of the follow?

ing: salary; retirement contribution paid by employer; health, dental, life, and/or disability

insurance premiums paid by employer; and longevity pay.

Data for the gender MIS were facts about the position, salary,

tenure, department location, and gender of municipal employees. The

data became information when they were organized into tables with the

characteristics displayed in Figures 5 and 6. Once it was interpreted and

reported in a comprehensive report about the status of the community's

working women, the information became knowledge and understanding

(Martin and Smith, 1986). As this knowledge is used or integrated into

policy debates, it too may become wisdom. With regard to die manage?

ment component of organization, the information initially produced by

the MIS was directed to the management hierarchies' upper levels. It was

directed at nonroutine and non-status quo aspects of the organization.

Findings

On the surface, at least, the two MIS projects had some common charac?

teristics. Both were responses to demands for increased efficiency. Both

used computer technology. Both involved condensing great quantities of

data into a few brief informational matrix tables. Both involved personnel

issues. Despite these surface similarities, however, only one of the two

projects contributed to organizational wisdom. Information produced in

the wage and salary MIS project resulted in knowledge and understand?

ing that facilitated meaningful organizational activity, while until now

the information produced by the gender project has not. Five factors

explain much of the two projects' impact disparity.

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

?g-s

1g8

=2-2 ?

^o?

? Jfe> -^

^ c *>

.a

65 65

1.1

65

65 65

<v

-s
C

c/5

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

78_Martin, Overman

First, the availability of context-defining concepts differed consid-

erably for the two projects. For the wage and salary project, the personnel

management literature provided a comprehensive conceptual package

(Shafritz, Hyde, and Rosenbloom, 1981; Thompson, 1979; Patten, 1979).

The wage and salary analyst's toolbox is full of such time-honored

notions as the one stating that organizations have positions filled by

incumbents who have specific duties and responsibilities. According to

this notion, these incumbents are selected and assigned to positions on

the basis of qualifications. Advancement within organizations along

career ladders is determined by merit. Moreover, within an organization,

individual positions can be assigned to job classes on the basis of objec?

tive criteria. In addition, for decades organizations have conducted salary

surveys and used odier methods to elicit the basic data that, combined

with rich personnel management conceptual packages, can become mean?

ingful information. In short, well-established personnel "principles" pro?

vided (1) concepts to help organize data so that it became information,

and (2) concepts to help users interpret diis information so that it could

become knowledge and understanding. In contrast, a well-developed

conceptual package was not available for information generated by the

gender MIS. Relatively few are context-defining concepts, such as occu-

pational sex segregation and job segregation. Also limited is the accom-

panying vocabulary for describing and analyzing gender in organization

(Ferguson, 1984, pp. 30-82; Aaron and Loughy, 1986; Gerson, 1985). Per?

haps somewhat ironically (as suggested in the conceptual vocabulary

presented in Figures 5, 6, and 7), the gender MIS borrowed heavily from

the conceptual package offered in the traditional personnel management

literature. As one consequence, the gender project risked misunderstand-

ing. Worse, as a result of "conceptual confusion" (Ammons, 1985), the

project risked serious tainting or biasing data in the process of turning it

into information; for example, "reform efforts to implement job evalua?

tion, even in the context of a strong affirmative action program, may not

overcome the salary disadvantages of female jobs, and, most perniciously,

may inadvertently serve to legitimate traditional values under the mask

of scientific procedure" (Rosenbaum, 1985, p. 117).

Second, concept familiarity among city management staff and

policymakers differed greatly between the two projects. They had mas-

tered the numerous concepts inherent in the comparative wage and salary

analysis, yet they had not mastered the relatively few associated with the

gender project. "I don't know what this means" was a common response

among managers and policymakers when such essential concepts as occu-

pational sex segregation were examined. Perhaps somewhat surprisingly,

this fundamental lack of concept familiarity was primarily apparent with

gender-specific concepts. For example, even though the notions occupa-

tional sex segregation and job sex segregation were unfamiliar to senior

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

Management and Cognitive Hierarchies_79

managers and policymakers generally, other concepts describing these

essential concepts' possible consequences were more readily understood.

Few senior managers and policymakers had difficulty grasping the mean-

ing of such concepts as wage disparity, susceptibility to unemployment,

occupational and status mobility, and occupational prestige (Reskin and

Hartmann, 1986, pp. 13-16).

Third, information generated by the wage and salary analysis was

generally seen within the organization as enhancing the status quo, while

that generated by the gender project was not. Wage- and salary-generated

MIS information was perceived as a tool of management to be used

during salary negotiations. The gender-generated MIS information was

viewed as a potentially disruptive force that required damage control. It

was also viewed as a tool that could be used by policymakers and staff to

critique senior management. To senior management officials, especially

those associated with the personnel department, the gender-based MIS

project brought up the whole "comparable worth" controversy. In addi-

tion, such concepts as occupational and job segregation sounded too

negative, too much like discrimination, and suggested a wrong impres-

sion. One senior official remarked, "Of course we support carefully

describing the status of working women in our organization and town.

But what will the headline read? We don't want to shoot ourselves in the

foot." The two MIS projects differed substantially along two other "status

quo" dimensions: Who controlled the design of the MIS, and who would

have access to the information it would generate? The two projects dif?

fered greatly in who determined the variables or attributes that should be

included in the data base and how data would be aggregated, analyzed,

and reported. Both the design and use of information generated by the

wage and salary MIS were controlled by traditional personnel depart?

ment staff. In contrast, the attributes of the gender-based MIS?that is,

the data it included and how these became information?were determined

by other government staff, outside interest groups, and elected officials.

In addition, the information was included in other documents, including

a detailed report of the status of the communities' working women, and

was readily available to any "interested citizen" (Martin, 1987; Martin

and Smith, 1986). Kraemer and King's (1986) comments about the role

of computing in organizations support this aspect of a status quo-

enhancing/detracting role for the MIS: "Computing is not in itself a

powerful and influential force in organizations. But it does provide the

opportunity to reinforce prevailing policy and attitudes toward larger

organizational issues. These policies and attitudes are typically shaped

by those already in powerful positions, and computing naturally is

absorbed as a tool by this elite" (p. 492).

Fourth, the level of managerial use differed considerably for the

two projects. The wage and salary MIS focused on providing information

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

80_ Martin, Overman

for maintaining the organization on a steady course of traditional func?

tions. It emphasized functions associated widi the middle levels of the man?

agement hierarchy. Although requiring choices to be made in a general

environment characterized by some uncertainty, the wage and salary MIS

information contributed to a standard and accepted organizational task.

Information generated by the gender-based MIS did not. Instead, infor?

mation it produced suggested new organizational activity, direction, and

image. Given the municipal government's current position within a his-

torical context?what it had and had not done over the last decade or

so?the gender-based MIS produced information that was directed toward

action for change, the upper level of die management hierarchy.

Fifth, and perhaps most important in explaining the projects'

differing impacts, the wage and salary MIS complemented and was sup?

ported by a very powerful "management" ideology; the information it

produced fit into this broader ideology. The gender-based MIS produced

information that did not; in fact, it produced information that could

challenge some fundamental tenets of management ideology. Although a

detailed discussion of management ideology is beyond the scope of this

article, we suggest that contemporary management, especially personnel

management, prescribes a way of looking at the world?as it "is" and

"should be" (Zeitlin, 1968; Bluhm, 1974; Merkle, 1980; Harmon and

Mayer, 1986; Paris and Reynolds, 1983). For example, management ideol?

ogy tells us that the organizational world is a world of hierarchy, and

that this hierarchy is largely determined by complexity and level of diffi?

culty; it is a world of competition that rewards and punishes on the basis

of merit and performance. As part of that description, management ideol?

ogy also explains why the world "is" the way it is ("hierarchy reflects

humankind's natural tendency toward competition and selfishness"). As

one consequence, collective or corporate efficiency requires "control" of

human passions (Merkle, 1980, pp. 287-294). With this description of the

world, management ideology also justifies the need for managers and for

the managed, and it tells why there is and must be a disparity between

them (Martin, 1983). Similarly, management ideology provides pre-

scriptions for what ought to be, a vision of "the good." It is a "good"

characterized by order, efficiency, a rational approach to problems, and

limited politics in administrative matters (Garson and Overman, 1983,

pp. 93-96). Clearly, as illustrated in Figure 4, the wage and salary analysis

MIS produced information both supportive of and supported by these

prescriptive and descriptive aspects of management ideology, but with?

out substantial interpretation, the gender-based MIS cannot. Instead, it

reveals disparity among employees, possibly attributable to gender, an

ideologically unacceptable criterion.

In addition, as noted previously, the wage and salary MIS was sup?

ported by the conceptual vocabulary of personnel management; informa-

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

Management and Cognitive Hierarchies 81

tion produced by the gender-based MIS was not. This also reflects one

function of management ideology, that of combining diversity and sim-

plifying complexity. The organizational world is extremely complex; to

be comprehended, it must be simplified into a more cognitively manage-

able package. Management ideology, like any ideology, provides "a partic?

ular conceptual vocabulary, . . . a set of concepts and terms by which men

interpret the world around them" (Rostow, 1953, p. 32). Moreover, to have

meaning, the world's contradictory, conflicting, and diverse pieces must

somehow be combined into a coherent whole. Thus, an ideology "ties

together into a coherent whole the various assumptions about human

nature, the nature of relationships, and the nature of society itself"

(Schein, 1985, p. 79). It also ties togedier levels of the cognitive hierarchy.

Finally, the wage and salary MIS information supported and was

supported by two other components of management ideology. It helped

justify the interests of believers in management ideology by providing a

rational and scientific view of organizational wage and salary distribu-

tions, relative to other jurisdictions. It also helped justify a broader social

order. Again, the gender-based MIS produced information that did nei?

ther of these things; in fact, it was seen as a challenge to the personal

integrity and performance of personnel and other senior management

officials, as well as a challenge to the municipal government or the

broader social order.

Conclusion

The wage and salary MIS helped integrate the management and cognitive

hierarchy in the study jurisdiction, while the gender-based MIS did not.

This "success" disparity between the two MIS projects was not due to

differences in technological sophistication, data availability, or methodo?

logical eloquence. Instead, much of the two projects' disparity can be attri-

buted to differences in ideological coherence, contextual fit, and how well

the managerial and cognitive hiearchies were balanced. As a result of the

host of reasons examined in the previous section, the wage and salary MIS

complemented the organization by providing knowledge and understand?

ing for management's decision making. Moreover, it was supported by

context-defining concepts. In turn, the information it produced further

validated those concepts. The wage and salary MIS had meaning for its

users, too, particularly for senior management staff. It generally supported

the status quo and was immediately useful as an enhancement to routine

activities. The wage and salary MIS also fit into the prevailing manage?

ment ideology. In short, it helped synthesize the management and cogni?

tive organizational hiearchies, while the information and data produced

by the gender-based MIS, for the reasons described previously, did not

satisfy management's needs for decision making or strategic change.

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

82_Martin, Overman

The findings presented here provide diree important lessons for

those involved with management information systems. First, the notion

of a management information system should be viewed more broadly. We

suggest that organizations include both a management hierarchy and a

cognitive hierarchy, each having many components. MIS designers also

need to consider, explicitly, all the components and their interactions in

designing systems. Viewing MIS broadly includes recognizing that tech?

nology, and particularly hardware and software, often may be the least

rather than the most important MIS component. An MIS that includes

sophisticated computers but does not fit into an organizational context

(perhaps because of impediments we described in connection with the

gender-based MIS) can be truly meaningless.

Second, those involved with an MIS should acknowledge the poten?

tially powerful role of management ideology. The meaning of data is not

self-evident. Meaning?data becoming information?requires interpreta-

tion and context. Much of the context may be supplied by management

ideology, but when it is not, data and information may fail to produce

strategic change. The content of the "point of view" that enables infor?

mation to become knowledge and understanding, and ultimately wisdom

within an organization, may largely be provided by management ideol?

ogy. When this ideology is incomplete or inappropriate, little can be

expected from an MIS. It was not simply that the results of the gender-

based MIS were not what senior managers wanted to see; that gender

considerations may be part of organization at all is problematic. For the

study organization, and perhaps for management as a field, the topic of

gender remains a challenge, an anomaly as yet unincorporated into the

rational world of management ideology.

The results of this study also provide a third important message.

It is not a new one. "MIS research is not well grounded in organizational

theory, nor have MIS research results been widely diffused in the organi?

zational literature, even though the organization is a concept central to

the definition of MIS" (Culnan, 1986, p. 170). Today, information man?

agement is as central as personnel and financial management to contem-

porary organizations. We believe that organizational theory holds the key

to better management information systems. We have chosen one particu?

lar view of organization (with strategic, management, and operational

levels) and its relationship to the cognitive characteristics of data, infor?

mation, knowledge, and wisdom in order to explain successful MIS

implementation. Clearly, this picture of the organization is not in itself

complete. Other theories of organization may also provide valuable ways

to understand the role of the MIS. For example, are there cultural deter-

minants of MIS success in the organization? Building better organiza?

tional theory requires coming to grips with the cognitive capacities and

potential products of the MIS. New organizational theory needs first to

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

Management and Cognitive Hierarchies_#3

acknowledge both the presence and the full complexity of the cognitive

hierarchy within organizations and then to develop appropriate roles for

the MIS. We are confident that the proper role of the MIS is to be found

through the integration or balancing of management purposes and

cognitive capacities in the organization. Finding this balance, however,

may require searching well beyond the confines of contemporary MIS

research.

References

Aaron, H., and Loughy, C. The Comparable Worth Controversy. Washington,

D.C: The Brookings Institution, 1986.

Ammons, D. N. "Common Barriers to Productivity Improvements in Local Gov?

ernment." Public Productivity Review, 1985, 9 (4), 293-310.

Anthony, R. Planning and Control Systems: A Framework for Analysis. Cambridge,

Mass.: Harvard University Press, 1965.

Banks, J., and Rossini, F. A. "Management Science Failures in the Public Sector."

Public Productivity Review, 1987, 10 (4), 15-23.

Bluhm, W. Ideologies and Attitudes: Modern Political Culture. Englewood Cliffs,

N.J.: Prentice-Hall, 1974.

Bolman, L., and Deal, T. Modern Approaches to Understanding and Managing

Organizations. San Francisco: Jossey-Bass, 1984.

Bozeman, B., and Bretschneider, S. (eds.). "Public Management Information Sys?

tems." Public Administration Review, 1986, 46, 488-496.

Brunsson, N. The Irrational Organization. New York: Wiley, 1986.

Cleveland, H. The Knowledge Executive. New York: E. P. Dutton, 1985.

Culnan, M. "The Intellectual Development of Management Information Systems,

1972-1982: A Co-Citation Analysis." Management Science, 1986,32 (2), 156-172.

Ferguson, K. The Feminist Case Against Bureaucracy. Philadelphia: Temple Uni?

versity Press, 1984.

Frost, P, Moore, L., Louis, M., Lundberg, C, and Martin, J. Organizational

Culture. Newbury Park, Calif.: Sage, 1985.

Gardner, H. The Mind's New Science: A History of the Cognitive Revolution. New

York: Basic Books, 1987.

Garson, G, and Overman, S. Public Management Research in the United States.

New York: Praeger, 1983.

Gerson, K. Hard Choices: How Women Decide About Work, Career, and Mother-

hood. Berkeley: University of California Press, 1985.

Gorry, G, and Scott-Morton, M. "Framework for Management Information Sys?

tems." Sloan Management Review, 1971, 13, 55-70.

Graham, C, and Hays, S. Managing the Public Organization. Washington, D.C:

U.S. Government Printing Office, 1986.

Hadden, S. "Intelligent Advisory Systems for Managing and Disseminating Infor?

mation." Public Administration Review, 1986, 46, 572-578.

Harmon, M., and Mayer, R. Organization Theory for Public Administration. Bos?

ton: Little, Brown, 1986.

King, J., and Kraemer, K. "Information Systems and Intergovernmental Rela?

tions." In T. Miller (ed.), Public Sector Performance: A Conceptual Turning

Point. Baltimore, Md.: Johns Hopkins Press, 1984.

Kochen, M. Information for Action: From Knowledge to Wisdom. New York: Aca-

demic Press, 1975.

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms

Martin, Overman

Kraemer, K., and King, J. "Computing and Public Organizations." Public Admin?

istration Review, 1986, 46, 488-496.

Martin, J. Enhancing Working Women's Status: Community Issues, Community

Responses. Boulder, Colo.: Division of Research and Evaluation, 1987.

Martin, J., and Smith, K. The Status of Working Women in Boulder. Boulder,

Colo.: Division of Research and Evaluation, 1986.

Martin, S. Managing Without Managers. Newbury Park, Calif.: Sage, 1983.

Merkle, J. Management and Ideology: The Legacy of the International Scientific

Management Movement. Berkeley: University of California Press, 1980.

Paris, D., and Reynolds, J. The Logic of Policy Inquiry. New York: Longman,

1983.

Patten, T. (ed.). Classics of Personnel Management. Oak Park, 111.: Moore Publish-

ing, 1979.

Reskin, B., and Hartmann, H. Women's Work, Men's Work: Sex Segregation on the

Job. Washington, D.C: National Academy Press, 1986.

Rosenbaum, J. "Jobs, Job Status, and Women's Gains from Affirmative Action:

Implications for Comparable Worth." In H. Hartmann (ed.), Comparable

Worth: New Directions for Research. Washington, D.C: National Academy Press,

1985.

Rostow, W., "A Note on the Diffusion of Ideologies." Confluence, 1953, 2 (1),

32-36.

Rubin, B. "Information Systems for Public Management: Design and Implemen?

tation." Public Administration Review, 1986, 46, 540-552.

Schein, E. Organizational Culture and Leadership: A Dynamic View. San Francisco:

Jossey-Bass, 1985.

Shafritz, J., Hyde, A., and Rosenbloom, D. Personnel Management in Government:

Politics and Process. New York: Marcel Dekker, 1981.

Sprague, R., and Carlson, E. Building Effective Decision Support Systems. Engle-

wood Cliffs, N.J.: Prentice-Hall, 1982.

Starling, G. Managing the Public Sector. Homewood, 111.: Dorsey Press, 1982.

Stout, R. Management or Control. Bloomington: Indiana University Press, 1980.

Thompson, F. Classics of Public Personnel Policy. Oak Park, 111.: Moore Publish-

ing, 1979.

Yates, D., Jr., The Politics of Management: Exploring the Inner Workings of Public

and Private Organizations. San Francisco: Jossey-Bass, 1985.

Zeitlin, I. Ideology and the Development of Sociological Theory. Englewood Cliffs,

N.J.: Prentice-Hall, 1968.

John A. Martin is senior research associate, Division

of Research and Evaluation, City of Boulder, Colorado.

He has written about the relationships among information,

management, organizational structure, and performance

for a variety of American and British publications.

E. Sam Overman is associate professor and director of doctoral

studies at the Graduate School of Public Affairs, University

of Colorado at Denver. He is the author of several articles on

information management in public organizations. His other

research and publications focus on public management,

decision making, and scientific inquiry.

This content downloaded from 121.52.154.36 on Mon, 28 Mar 2016 06:36:27 UTC
All use subject to http://about.jstor.org/terms