Citation
An assessment of the impact of personal computer technology on employee productivity

Material Information

Title:
An assessment of the impact of personal computer technology on employee productivity
Creator:
Whatley, Roger
Publication Date:
Language:
English
Physical Description:
vii, 200 leaves : forms ; 29 cm

Subjects

Subjects / Keywords:
Employees -- Effect of Microcomputers on ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 169-187).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Doctor of Philosophy, Public Administration.
General Note:
School of Public Affairs
Statement of Responsibility:
by Roger Whatley.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
34233089 ( OCLC )
ocm34233089
Classification:
LD1190.P86 1995d .W43 ( lcc )

Full Text
AN ASSESSMENT OF THE IMPACT OF PERSONAL COMPUTER
TECHNOLOGY ON EMPLOYEE PRODUCTIVITY
by
Roger Whatley
B.S., Metropolitan State College, 1976
M.A., University of Northern Colorado, 1978
A. thesis submitted to the
Faculty of the Graduate School of the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Public Administration
1995


1995 by Roger Whatley
All rights reserved.


This Thesis for the Doctor of Philosophy
degree by
Roger Whatley
has been approved for the
Graduate School of Public Affairs
LawrencqJB. Marquez


Whatley, Roger (Ph.D., Public Administration)
An Assessment of the Impact of Personal Computer
Technology on Employee Productivity
Thesis directed by Professor Franklin James
ABSTRACT
Despite the proliferation of personal computers
(PCs) in organizations in recent years little has been
published on their effect on technology management or of
their effect on employees in the Federal public sector.
This study examines the effects of personal computer
technology on employee productivity, quality of work life
and attitudes toward computers in a Federal organization.
Generally, the research hypothesis states that the
infusion of technology will bring about changes in the
areas examined, and that the overall effect will be an
increase in productivity.
The research model is quasi-experimental
(before/after). Analysis is based on data obtained
before and after technology infusion from the same 75
randomly chosen employees involved primarily in
performing clerical duties as well as some data analysis.
Statistical survey techniques included a combination of
retrospective interviews and questionnaires. An
extensive review of the literature provided additional
views and concepts for interpreting the data.
IV


Data was analyzed using the-Statistical Program for
the Social Sciences (SPSS-PC) computer software package.
Pre- and post-implementation responses were measured
using the Student's t-test to determine whether change
occurred. Results were evaluated to determine the
technology's effects on productivity, quality of work
life and attitudes.
Results of the analysis resulted in modifying models
of technological impact as reported in the literature and
in challenging the assumption of one-way causation.
Results also support Erik Brynjolfsson's "time lag"
theory which suggests that productivity improvements do
not appear until some time after technology infusion.
Analysis of the research data suggests that the
introduction of personal computers had powerful but
varied effects on attitudes, productivity and quality of
work life. While employee productivity appeared to
improve soon after the introduction of the personal
computer, the technology's contribution to organizational
productivity was not realized until almost two years
later.
v


CONTENTS
CHAPTER
1. INTRODUCTION........................................ 1
The Productivity Paradox............................ 4
The Federal Experience..............................10
Improved Services Through Technology................15
Framework for the Study.............................17
The Job Before Automation.......................21
The Job After Automation........................24
Purpose of the Study................................26
Significance of the Study...........................28
2. LITERATURE REVIEW...................................30
Introduction........................................30
Computers A Brief History.........................31
The Personal Computer...........................33
The Business Side of Computer Technology........38
Office Automation...................................43
Office Applications of Computer Technology......45
Office Automation and Nature of Organization...49
Office Automation and Nature of Office Work....55
Productivity and the Personal Computer..............70
Office Automation and Worker Productivity.......79
Productivity in Office Work.....................85
What Can Be Learned From This Research?.............94
Conclusions.........................................98
vi


3. INVESTIGATIVE PROCEDURES ..........................103
Methodology .......................................103
Research Design....................................109
Data Analysis. ................................113
Data Collection and Respondents....................114
4. RESEARCH FINDINGS................................ 120
Measures of Productivity...........................120
Effects on Productivity........................... 121
Short-term Effects on Organizational
Productivity.............................. .122
Long-term Effects on Organizational
Productivity. ...............................129
Employee Productivity..........................134
Effects on Quality of Work Life....................142
Attitudes Toward Computers.........................148
One Way Causation.............................. 149
The Evolution of the Technological Configuration..149
5. CONCLUSIONS .......................................153
A Revised Model for Technology Impact.............157
Summary........................................... 163
BIBLIOGRAPHY..........................................169
APPENDIX
A. Cover Letters ..................................188
B. Questionnaire Outline...........................192
C. Questionnaire....................................196
vn


CHAPTER 1
INTRODUCTION
The Federal Government is the largest user of
computer technology in the United States. Of the more
than 1.6 million personal computers in the Federal
Government in 1990, approximately 25,000 are in use in
the Department of the Interior (Government Computer News
1990a). Approximately 1,200 are in use in the
organization examined. Spending on computer hardware,
software and services increases every year. In 1989, the
Federal Government spent approximately $17 billion, or
1.6 percent of the Federal budget, on information
technology. The Office of Management and Budget (OMB)
reported that in 1993 computer-technology- related
spending would exceed $25 billion, constituting
approximately an eighth of all government procurement
outlays (Government Executive 1993a).
Throughout the Federal Government, agencies are
pursuing a variety of approaches to improve
organizational productivity technological infusion and
management methods such as Total Quality Management are
examples of this. As the Federal Government enters the
decade of the 1990s, a fundamental dilemma is being faced
1


by the Federal establishment. On the one hand,
legitimate demands for Federal Government services are
sharply increasing a movement abetted in part by a
growing population seeking Federal services. On the
other hand, in the face of a staggering Federal deficit,
Federal managers are finding it exceedingly more
difficult to obtain adequate funding for essential
ongoing and newly needed programs, costs of which are
mounting through inflationary trends.
A practical solution to the dilemma may be for
Federal policy-makers and for Federal managers at all
levels (1) to assess their productivity-related
activities; (2) to consider very carefully positive and
negative experiences of applying computer technology to
increase productivity and (3) from the array of effective
methods, techniques and ideas, to establish and/or
augment a useful pattern of appropriate actions geared to
their responsibilities and designed to optimize
productivity.
Today's Federal managers have a virtual arsenal of
automation at their command. The ability to collate,
sort, disseminate and analyze data of all sorts has never
been better, thanks to billions of dollars' worth of
2


investments in information technology. Some of the
literature posits that worker productivity should
increase, to a certain extent, thanks to the computer
(Edwards and Snyder 1987; Scientific American 1987;
Reichard 1992; Rummer 1994). In the workplace, the
personal computer has evolved from being a luxury, to a
privilege, to a necessity. With the ability to network
with other offices and the benefits of sharing resources,
office productivity is further enhanced (Sharma 1993).
Taken together, one could almost deduce that the personal
computer has ushered in a new Golden Age of productivity.
But, Reichard reports, scattered skepticism remains. In
the words of economist and Nobel laureate Robert Solow,
"You see computers everywhere but in the productivity
statistics" (Reichard 1992).
In the past decade, the personal computer's effect
on productivity has stimulated interest worldwide.
Consequently, billions are being spent on personal
computers. However, the impact that this equipment is
having on performance is still being examined largely in
very general, qualitative terms. One difficulty in
measuring or evaluating productivity quantitatively is
the diversity of applications for which employees use
3


personal computers. The discussion that follows
discusses these and other related issues.
The Productivity Paradox
Productivity is the fundamental economic measure of
a technology's contribution (Brynjolfsson 1993). With
this in mind, managers have increasingly begun to
question their huge investments in computers and related
technologies. While major success stories exist, so do
equally impressive failures.
For more than a decade, economists and policy-makers
have been concerned about the sluggish growth in economic
productivity in the United States (Crawford 1988). A
fundamental weakness affecting much of American industry,
according to Baily and Chakrabarti (1988), is the failure
to diffuse new knowledge quickly and to refine existing
technology. The slowdown in productivity, they say,
occurs because we fail to incorporate new technology
effectively. The blame, therefore, must be shared by
those who choose not to employ available technology in a
way that would raise productivity.
The failure to realize substantial productivity
gains involves not just industry, but extends to the
4


white-collar service sector as well. Baily and
Chakrabarti state that while the United States has had
steady productivity improvements in the manufacturing of
computers, the computerization of the American work place
has not yielded similar results. They cite what they
call the "productivity paradox": difficulty in measuring
gains, a delayed response relating to learning how to
utilize the equipment efficiently, or, findings that
staffing cannot be reduced because equipment is not
readily substitutable for labor in the information sector
(Baily and Chakrabarti 1988).
Baily states that much of the difficulty in
measuring productivity in the office results from
inconsistent reporting throughout the organizations and
the general failure of the white collar sector to
establish measurement standards that assess the
effectiveness of information technology (Baily 1995).
Duncan and Chester (1990) go one step further by
suggesting that the problem may be deeper. They state
that within those companies employing personal computers
in their work, typical problems in achieving productivity
growth include the users' lack of knowledge about a
company-wide computing strategy and the fact that firms
5


generally do not attempt to measure how their technology
investments are translating into increased productivity
(Duncan and Chester 1990).
Roach (1985), a prominent critic of information
technology spending, states that within the white collar
sector, output per "information worker" has either fallen
or risen more slowly than output per worker in the non-
service sector. He states that in spite of 15 years of
"high teching," America is still trapped in a quagmire of
low productivity growth. "There are only scattered
examples where a strategy of technology acquisition can
be associated with improved productivity performance,"
Roach wrote in a study published in 1987. "[In] the broad
scheme of things, these and other success stories turn
out to be the exceptions rather than the rule." He
concluded by stating that computer technology has quite
simply not delivered its long-awaited productivity
payback (Roach 1987).
Roach cites statistics indicating that output per
production worker grew by 16.9% between the mid-1970s and
1986, while output per information worker decreased by
6.6%. He suggests that there is less productivity in the
6


white collar sector where the highest investment in
technology has occurred (Roach 1991).
In another study, Osterman (1986) found that
computers encourage companies to engage in more computer-
related activities rather than to reduce labor. He
concluded that information technology is not readily
substitutable for information labor. He also states that
technology requires more information workers as the price
of technology falls. For this reason, he suggests,
clerical employment increased in the 1980s after the
introduction of computers. Berndt and Morrison (1991)
confirm that information technology capital is, on
average, a complement for white collar labor, even as it
leads to fewer blue collar workers.
However, not everyone agrees with these assessments.
Other observers say that statistics alone do not present
a clear picture of the boosts in productivity computers
have brought to businesses. Sidney Harris, Professor of
MIS at the Peter F. Drucker Graduate Management Center in
Claremont, California, said that studies like Roach's are
too sweeping to have any meaning. "The problem with
studies of this kind is it's too easy for the effects of
computer technology to look like a wash. If the level of
7


aggregation is too high, the successes are pulled down by
the failures." He suggested that technology payoffs
emerge when businesses are studied on an industry-by-
industry basis, firm by firm, and even department by
department (Harris and Katz 1989).
By way of illustration, Harris cites a study he
conducted of the insurance industry, which he says
demonstrates a clear link between investment in computers
and productivity. Of the companies he studied, 97
percent of the most profitable were in the top quarter of
spending on computer technology. Of the least profitable
companies, however, only 10 percent were among the top-
spending quarter.
Brynjolfsson (1993) states that the relationship
between information technology and productivity is widely
discussed but little understood. He states that after
reviewing and assessing the research to date, it appears
that the shortfall of information technology productivity
in the white collar sector is as much due to deficiencies
in Our measurement of outputs and inputs as to
mismanagement by developers and users of the technology.
He also cites lags due to learning and adjusting to the
technology as a problem.
8


Mismeasurement points to shortcomings in research,
not practice. Denison (1989) makes a case that
productivity and output statistics can be very
unreliable. For example, it is particularly difficult to
account for intangibles in productivity statistics such
as increased quality, variety, customer service, speed
and responsiveness. Yet, these may be some of the very
benefits that managers attribute to technology. With
this in mind, it is probably no coincidence that white
collar productivity statistics show less growth than in
the manufacturing industry where measures are better.
Baily and Gordon (1988) conducted a study of the
finance, insurance and real estate areas. Computer
technology and information workers are particularly high
in this segment of the white collar sector. They
identified a number of measurement practices by the
Bureau of Economic Analysis which tend to understate
productivity growth. Their revisions add 2.3 percent per
year to productivity between 1973 and 1987 over previous
estimates to this sector.
Additionally, if lags between cost and benefit do
exist, then the productivity statistics would indeed be
poor in the short-term. As the technology adjustment and
9


learning curve shortened, the long-term statistics would
show an increase in productivity as the technology is
fully exploited. A recent study by Brynjolfsson et al.
(1994) found lags of two-to-three years before strong
organizational impacts of technology were felt.
Roach also seems to be changing his tune. In a more
recent study he predicts that we have turned the corner
and will soon begin to see substantial improvements in
productivity (Roach 1992). But, he cautions, the country
still has a managerial challenge ahead. What is needed
is a revolution in how firms apply the technology.
The Federal Experience
Charles Goodsell (1994) states that "bureaucracy
works" (p. 46). He points to Bureau of Labor Statistics
(BLS) data on Federal government productivity compiled
between 1967 and 1990 which measures labor resources
spent on, and program outputs achieved from, more than
2000 individual activities in over 300 organizations. He
suggests that when compared to the private sector during
the same period of time, Federal productivity was much
higher. He concludes that this should lay to rest any
assumptions that the productivity of the government never
10


goes up. Edward Lynch (1993) agrees and points out that
BLS statistics covering the 10 year period 1982 through
1991 indicate that output by Federal workers increased by
13.7 percent. The Federal gain exceeded the 11.2 percent
productivity increase BLS reported on the private sector
during the same time span.
A GAO study has calculated that for every five
percent increase in Federal productivity, a $4.5 billion
dollars could be saved (U.S. 0MB 1987). A major
initiative in the Federal government continues to be
productivity improvement. Early in 1986, President
Reagan signed an Executive Order which called for a
government-wide program to improve productivity by 20
percent in selected high priority functions. He called
on Congress to join in designating productivity
improvement a national priority.
Several other management initiatives are either
directly or indirectly tied to productivity issues. In
addition to President Reagan's Productivity Improvement
Program, other initiatives over the past several years
include Reform 88 (the Paperwork Reduction Act); Waste,
Fraud and Abuse; and Management Improvement (0MB Circular
11


A-76) which details procedures for contracting out
products and services.
There are a few reports that improvements in Federal
office productivity through technology have occurred and
that the Federal Government has benefitted somewhat from
productivity improvement efforts of the past several
years (Government Executive 1991). Burstein and Sedlock
(1988) suggest that the personal computer appears to be a
significant factor in productivity increases. Certainly,
the Federal government appears to have bought into this
premise. According to a 1990 study conducted by
Information Strategies Group of Vienna, Virginia,
agencies across the Federal Government are exploring the
adoption of new desktop technologies more rapidly than
industry and its observers expected (Federal Computer
Week 1991).
President Clinton's election signaled a new dawn for
information technology in the Federal government.
Clinton is clearly more comfortable and conversant with
technology than were his predecessors. In his campaign,
he called technology "the engine of economic growth."
His 1993 State of the Union speech was the first ever to
mention information technology. He has made clear that
12


he sees computers and related technologies as
productivity tools for improving government services and
cutting Federal expenses. Vice President A1 Gore, one of
technology's best friends on Capitol Hill as a senator,
is moving ahead with the National Performance Review
mandate to make a more productive government through
reinvention (Government Executive 1993b), which
emphasizes technology as well as purchasing and personnel
reforms.
Experience in the private sector suggests that
changes in organizational structure and size are
essential to reap the full benefits of technology
(Golembiewski 1985). With the exception of the current
mandate to reduce the Federal work force by 272,000, the
size of the Federal civilian work force has not changed
significantly in the past 20 years (OPM 1990). Does this
suggest that the Federal government is not being
productive? Certainly in this era of cutback management,
pervasive shortfalls in funds for the basic
infrastructure costs of providing government services
personnel, computers, training, capital equipment have
compromised the ability of almost every government entity
to accomplish its traditional mission. And this has
13


occurred at a time when the demands on the public service
have never been higher fGovernment Executive 1989b).
Most Federal managers are given little leeway when it
comes to implementing employee cutbacks. Agencies that
try to close offices and bureaus in response to budget
reductions or improvements in productivity face the ire
of Congress. Members typically reject any reduction of
jobs or federal presence in their districts (the
reductions in Defense installations are an exception to
this rule). But these reductions were driven by budget
constraints and not because of productivity improvements.
As Federal agencies struggle to do their work in the
face of tight budgets, they are continuing to invest in
technology that can automate processes to cut costs and
improve services to their customers. This is evident
from the Federal budgetwhich allots about $27 million
for information technology in fiscal year 1995, an
increase of seven percent over 1994 (Government Executive
1994b). Staff reductions aside, the Office of Technology
Assessment reports many instances of productivity
improvements, many of which are attributable to the
current technology (OTA 1995). But, considering the
difficulty in measuring productivity improvement, perhaps
14


Vice President Gore was not far off the mark when he
suggested that it would take eight to ten years to
complete the transformation to a more productive and
efficient government (Creating a Government that Works
Better and Costs Less 1993). The crucial step in
overhauling government in an effort to increase
productivity is to decide what it should and should not
do and then eliminate the programs that don't belong.
These are, however, political choices, not questions of
management efficiency.
Improved Services Through Technology
The steady advance of information technology brings
a demand for ever more sophisticated computer
applications and telecommunications improvements (Snyder
1986). This increased demand for information technology
applications is expected to outstrip the supply of
technically sophisticated Federal workers, thereby
increasing reliance on contracting out (Government
Computer News 1987). Additionally, Federal agencies will
increasingly purchase, rather than build, their
information systems (Federal Computer Week 1990).
Private sector systems integrators will provide
15


comprehensive information processing solutions to
government's service delivery problems by combining
expertise in off-the-shelf hardware, software, and
telecommunications technologies (Government Computer News
1990). Much more so than today, nearly all Federal
managers will become information technology managers
directly or indirectly.
Herbert Simon noted as early as 1976 that public
sector managers may find that the introduction and use of
information technology tools creates additional or new
demands on their time and attention and places additional
burdens on them in terms of resource utilization and on
organizational issues such as productivity and morale
(Simon 1976). According to Simon:
A major task ahead for science and technology
is to design effective information-processing
systems for making decisions in business and in
government. The design of such systems ...must
handle with equal care the information-
processing characteristics and capabilities of
the human members of organizations who
constitute the other half of the system ...and
will continue to be the most numerous and
critical components of organizations (p. 286).
Technological change introduced to an organization
frequently has implications for the entire organization,
often far more significant and widespread than originally
anticipated. Understanding the potential ramifications
16


thus takes on organizational importance. As in the
scenario described by this study, computer technology
continues to be applied to the automation of office tasks
and procedures in an effort to increase organizational
productivity, often without adequate advance planning.
Framework for the Study
In 1991, top management officials of the
organization examined in this study decided to further
automate the Fiscal Accounting Division to increase
efficiency and cut labor costs. Other factors driving
this decision were that management officials of the
organization reportedly perceived pressure from senior
Bureau officials to "enter the modern age." Discussions
with managers from other Bureaus implied that this
organization was "behind the technology curve." Reaction
to these perceptions of pressure resulted in inadequate
planning to successfully integrate the technology into
the organization and a general failure to reengineer the
work processes of the data analysts. As this study will
show, this made productivity improvements very difficult
to realize.
17


The managerial goals that initiated the
computerization of the record system reducing labor
costs while maintaining service quality placed only
loose constraints on the technology that was introduced.
The interactive record system using personal computers is
only one type of technology that could have met these
goals. However, this option was determined to be the
most cost effective and, in terms of training and time
required for implementation, more efficient.
Other options considered included expert systems
linked directly to the large mainframe computer using
newer, more sophisticated terminals and other tool-based
systems that would provide data analysts with more
information for making decisions or answering questions
(Cole 1986).
The Fiscal Accounting Division has primary
responsibility for all revenues collected from solid
minerals, oil and gas leases on Federal lands. This
includes determining the value of oil, gas and solid
mineral leases, collecting interest on late payments,
ensuring that royalties paid are on time and are the
correct amount, and disbursing collected royalties to
clients (i.e., Bureau of Indian Affairs, Treasury
18


Department, states, Indian tribes and individual Indian
allottees). An integral part of their responsibility is
ensuring that the data base is kept current and accurate
for obvious fiscal reasons, but also because it serves as
a reference data bank for other client Divisions whose
responsibilities include regulatory enforcement, mineral
production verification, auditing, and lease sales.
The focus of the research is on the data analysts
employed by the Fiscal Accounting Division. Personal
computers were installed to automate information that
data analysts had previously viewed on microfiche and
altered by completing paper transaction records.
Compared to the microfiche system it replaced, the
interactive computer system was to provide data analysts
with more recent revenue and disbursement reports in a
more convenient way and would allow interactive updating
of client records. By intention, no attempt was made to
alter the data analysts' work processes more extensively.
In particular, the technology altered neither the range
of tasks that they performed nor the importance of
interaction with clients while performing these tasks.
In this environment, personal computers appeared to
be the best alternative in terms of equipment investment
19


and software development. The cost of the PCs were
hardly more than the newer terminals would have been.
And, it was an alternative that was easy to sell to
senior Bureau officials who, local management believed,
wanted to see quick results. Additionally, personal
computers could serve long range plans for off-line
(i.e., non-mainframe based) development of future
applications. This could not have been done with the
single purpose terminals, nor could the single
application software have served a multiple purpose in
the future. While the initial budget outlay for personal
computers was higher than it would have been for the
other alternatives, in the long run the personal computer
alternative was considered to be more cost effective
because of its multiple capabilities and its potential
for doing jobs that otherwise would not be possible to
perform. Downloading data from the centralized data base
for further comparison, research or ad hoc report
generation and electronic spreadsheet analysis are
examples of this. The following summarizes the work
environment before and after automation.
20


The Job Before Automation
In this organization, the dominant characteristic of
customer service work was customer contact (i.e., payors
of mineral royalties): conversing with customers,
interpreting regulations, and solving the customer's
problems as well as those of the organization in this
study. The computerized record system was not designed
to affect the importance of customer contact in the data
analysts' job description or the way in which they
interacted with customers.
Data analysts had two major responsibilities: to
answer incoming calls from customers and to make outgoing
collection calls on overdue royalty payments or error
corrections on submitted payment forms. On a typical
incoming call, a data analyst would explain royalties due
on a producing lease to a customer and determine whether
the charge was accurate or could be appealed. On a
typical outgoing call, a data analyst would determine why
a customer had not paid royalties due or had submitted a
late or erroneous payment, ask the customer to pay the
correct amount, and determine whether to initiate
penalties for overdue payments. While these tasks
represent approximately two-thirds of their work
21


activities, data analysts were also responsible for a
wide range of other customer contact activities which
included preparing penalty assessment correspondence;
responding to complaints about annoying aspects of the
organization's service; suspending or canceling mineral
leases; answering queries about Federal oil and gas rules
and procedures; changing payor addresses and payor codes;
and transferring customer calls to other parts of the
Bureau.
To provide the informational support for these
customer contact tasks, each data analyst had a
microfiche reader and a set of microfiche copies of
recent bills for all accounts serviced. Older bills were
kept in a central storage area, and data analysts had to
leave their desks to retrieve them. In addition, each
data analyst had primary responsibility for a subset of
the accounts and kept trays of paper records regarding
those accounts on his or her desk.
Incoming calls were routed through a central call
distributor that connected a customer to the next
available data analyst. Upon receiving a call, the data
analyst retrieved and displayed the microfiche copy of
the customer's payment records. While looking at the
22


customer's record, the data analyst discussed the account
with the customer and answered questions. For about half
of the calls, a paper trail of the customer's problem had
already been initiated (this would be true, for example,
if the customer had been issued on overdue notice). In
such cases the data analyst put the customer on hold,
retrieved those paper records from the desk of whomever
was responsible for that account, and returned to
complete the call. Between calls or during routine
interactions with long-winded customers, data analysts
completed the clerical work generated by customer
contacts (e.g., completing paper transaction records or
refiling microfiche). This "load balancing" (filling
dead time with clerical work) and "overlapping"
(finishing paper work for one customer while talking to
another) were important skills that allowed data analysts
to work efficiently.
Outgoing calls were conducted in scheduled blocks of
time, during which data analysts did not answer incoming
calls. Data analysts generated a prioritized list of
customers who required a return telephone call or who had
an overdue account or had submitted an erroneous payment.
They then made calls to these customers, generally to
23


inquire about unkept payment arrangements or to discuss
and correct errors such as an over- or under-payment on
submitted royalty payments.
The Job After Automation
With the PC-based automated record system, the role
of the mainframe was reduced to comparative updates and
mass data base storage. While the data analysts' contact
with customers remained much the same, the clerical
aspects of the job changed substantially. Personal
computers replaced the microfiche readers, although a few
readers were still available in a central location for
accessing old accounts on terminated leases. Data
analysts retrieved a customer's record by entering the
customer's identification number (payor code) into the
personal computer and matching that against a master
record on the mainframe. Because data analysts typed
changes, lease arrangements and other comments directly
into the personal computer data model and then
transferred that data directly to the mainframe system,
they generated few paper forms and avoided a substantial
amount of filing and other routine clerical work. This
direct data entry also reduced their need to retrieve
24


paper records from others' desks, diminishing movement
(and social interaction) throughout the office.
When they were not scheduled to take incoming calls,
data analysts made their outgoing calls. The computer
rather than the data analysts generated the prioritized
work list of accounts requiring action.
Along with the introduction of the PC-based record
system, other changes in the technology and social
organization of the division were made. First, all of
the data analysts were moved to new quarters with
modular, ergonomically designed furniture. The office
layout also changed: square arrangements of modular
furniture separated by portable panels replaced rows of
desks, disrupting familiar seating arrangements and
friendship patterns, and making informal chatting more
difficult. Second, a new telephone system, with
increased reporting and monitoring capabilities, was
installed. The reporting capabilities were used to
schedule and manage work loads, rather than to evaluate
individual data analysts. Finally, some changes in
management policy were introduced. Most important was
enforcement of an existing policy that all clerical work
associated with a customer interaction be accomplished
25


while the customer was on the line. In large part, this
change occurred because all work on one customer record
had to be completed before a new record could be
accessed. These changes reduced opportunities for data
analysts to balance their work load or to overlap tasks
in traditional ways, and created tensions as the data
analysts raced to finish their work before hanging up on
a customer.
Purpose of the Study
Do employees doing predominantly clerical tasks such
as filing, data entry, or answering telephone queries for
account reconciliation become more productive using
personal computers? If not, why are so many business
entities, public and private, investing in this new
technology? The lack of a clear-cut answer to this
question is why this topic was chosen for research. It
should be a major point of discussion in business
circles, both public and private, today.
The purpose of this research, therefore, is to
determine the impact of personal computer automation on
employee productivity and quality of work life, and its
effect on employee attitudes, by analyzing collected data
26


regarding the conversion of a largely manual,
recordkeeping system to a PC-based system. The study
examines the short-term effect of automating the
organization's royalty revenues accounting department in
an effort to increase productivity and, potentially, cut
labor costs. The manual system consisted of clerical
functions which were converted to a PC-based system that
left the output and tasks of the job basically unchanged.
The technology introduced automated information that data
analysts had previously handled in paper form, viewed on
microfiche, and altered by completing paper transaction
records for input into the large mainframe computer.
This research augments knowledge of public Federal
sector productivity improvement efforts through
technology infusion. It explores the impact on
effectiveness, productivity and morale that personal
computers can have in the work place and on the quality
of work being performed by government employees. It will
provide a resource for the organizational planning that
should take place before making purchases of personal
computers. This research also provides a framework,
which similar organizations can examine, for measuring
the impact of personal computers upon productivity.
27


Significance of the Study
The issue of how employee productivity is affected
by the use of personal computers is significant because
productivity in the Federal public sector attracts the
focussed attention of Federal managers as well as of the
taxpaying public. Computer technology has long been a
useful tool for improving or increasing productivity in
many areas of industry such as finance, banking, and
manufacturing (Business Week 1986), but review of the
literature indicates that few studies have been completed
to determine the effect of personal computers on
productivity in an office environment in the Federal
public sector. Perhaps this is because the personal
computer as a useful office tool has only gained
recognition in the last few years. Or perhaps, as in the
private-sector white-collar jobs, adequate measures of
and controls on productivity do not exist, and these
deficiencies are magnified by the sheer size of the
Federal bureaucracy. In any case, there is a major gap
in our understanding of the effects of personal computers
on productivity in office work. Perhaps the results of
this study can help public administrators as well as
researchers review the potential for productivity changes
28


using this technology and the possible effects on human
resource management. It is possible that this technology
can offer organizations the potential to do things better
or to do things that were very difficult, if not
impossible, before (Courter 1994).
29


CHAPTER 2
LITERATURE REVIEW
Introduction
Technology makes the world a new placea concept
expressed by Fernand Braudel (1981) when he wrote:
It was only when things went wrong, when
society came up against the ceiling of the
possible, that people turned of necessity to
technology, and interest was aroused for the
thousand potential inventions, out of which one
would be recognized as the best, the one that
would break through the obstacle and open the
door to a different future.... In this sense,
technology is indeed a queen: it does change
the world (p. 435).
To fully grasp the way in which a major new
technology can change the world, as described by Braudel,
one must consider both the manner in which it creates
intrinsically new qualities of experience and the way in
which new possibilities are engaged by the often
conflicting demands of social, political, and economic
interests. Though intentions do not always predict
consequences, humans do attempt to proceed by
constructing meaning, assessing interests, and, with
varying degrees of awareness, making choices. As the
ceiling of the possible is newly defined, opportunities
for choice are multiplied. It is here in the realm of
30


choice that technology reveals its indeterminacy. Though
it redefines the possible, it cannot determine which
choices are taken up and to what purpose.
This chapter discusses the history of computer
technology including the evolution of the personal
computer and its application in the business environment.
Major applications and benefits of personal computers in
office automation, their impact on worker productivity,
hindrances and problems that arise from their use, and
their effect on quality of work life are also addressed.
Computers A Brief History
Computers as we know them have been a part of big
business as long as there have been computers. The first
modern computers were financed by the U. S. government to
be used in the 1890 census. The 1880 census took nine
years to count; the 1890 census (with machine) took
three. The man who invented the machine, Herman
Hollerith, renamed it "The Tabulating Machine," and
joined the Computing Tabulating Recording Company, which
later became IBM (McWilliams 1983).
The hardware end of computing was revolutionized
with the introduction of the transistor in the mid-to-
31


late 1950s. The first transistorized computer was built
in 1960. Technology advanced; engineers learned new and
better ways to make transistors. Transistors became more
reliable and were able to work at higher frequencies.
Packaging transistor circuitry took great strides with
the progression of printed circuity technology.
Multilayer printed circuit boards allowed new, smaller
packaging and higher-speed switching. In the early 1960s
a new technology was created, the integrated circuit.
With the integrated circuit it was possible to duplicate
not only the transistor, but also a resistor by using a
straight piece of silicon as an interconnection between
the two transistors (Huffman 1979).
Until a few years ago, most people viewed a computer
as a million-dollar monster occupying half of the third
story of a high-rise office building. Because of the
speed improvements that ICs offered, they were rapidly
pressed into service in the big mainframes. Huge
multimillion-dollar computers were made obsolete almost
overnight. At the same time, minicomputers were brought
to prominence. Minis made it possible to set up a
computer system for $100,000 to $200,000 and unlocked the
doors for medium-sized and smaller businesses to have on-
32


site computers. During the 1960s, medium scale
integration paved the way for improvement in size
reduction and performance. As technology continued to
advance, it was not long before an entire computer could
be combined on one chip, and thus truly was born the
personal computer (Utterback 1994).
The Personal Computer
Most attribute the beginning of the personal
computer industry to the Altair computer, which was made
available initially in 1975 through a kit that could be
assembled at home. Personal computer sales were fueled
by the subsequent introductions of the Apple and
Commodore PET computers and other microcomputers that are
no longer sold. The other side of the computer
industry mainframe and minicomputer manufacturers --
were also significant contributors. Programmable
calculators were gradually becoming more sophisticated by
the early 1970s and minicomputers were being scaled down
to fit on desktops. Computer development strategy, in
general, was either to increase functionality and keep
the price the same, or decrease the price and keep
functionality the same, a pattern that still remains.
33


Part of that functionality was decreasing computer size
while keeping computing power the same (Fleming 1984).
Clive Sinclair, a Cambridge entrepreneur, introduced
the first pocket calculator, in the late 1960s, at a
price in the $200 range. By early 1980, he was producing
a computer for the same price. Taking inflation into
account, the true cost is between one-seventh and one-
tenth of the original. Consequently, an employee can now
have, on his own desk, more computer power than the
entire computing department would have possessed a decade
ago (Bradshaw 1982).
Microcomputers, or personal computers as they are
now called, have come into widespread use in the
corporate environment of the 1990s because of their
increasing capabilities, the demands of large
organizations on the time of their employees, and the
human factors associated with the large-scale work
environment. The personal computer acquired its name
from its physically small size and (by present-day
standards) its somewhat limited memory size and speed.
Yet today's personal computers have as much, or even
more, capacity and capability than the most commonly used
machines of the late 1960s or early 1970s. Thus, for a
34


modest price, it is possible to have on one's desk a
computer capable of assisting with many common office
chores, i.e., word processing, financial planning,
modeling, mathematical calculations, note taking, and a
multitude of similar tasks (Madron 1983).
The "personal" aspect of personal computers cannot
be overemphasized. One of the primary reasons for these
machines' appeal is that they are a personal or
individual tool, one shared by many (Madron 1983). In
some ways, personal also means private. The user can
test out ideas without risking his/her "image" in the
organization. Most large companies seek to improve staff
productivity. One way to do this is to provide employees
with better tools. For the white collar workforce, one
of the major productivity tools of the 1980s and the
1990s appears to be the personal computer.
Because personal computers now have the power that
only large-scale computers once had, these small machines
can be applied to the automation of many small jobs
previously done manually. Mansfield (1983) suggests that
they also offer advantages over the central-computer
approach:
35


1. Because programs are independent of the
mainframe system, there is no need to evaluate the
effect of a change on those subsystems.
2. Development can be done in the users
department, as the programming task has usually been
simplified.
3. Because of the location and simplicity of
programs, implementation can normally be done much
more quickly.
4. Time can often be saved through in-department
development, in that the programmer/user already has
a full understanding of the problem and is able to
arrive at a solution without several abortive
attempts.
5. Finally, the small computer approach is more
flexible, because the programming task is less
complex. When changes occur, new programs can be
written quickly to deal with the new set of problems
brought on by these changes.
These advantages, suggests Mansfield, often translate
into productivity improvements to the organization
through time savings and redirected efforts of their
employees.
36


An article written by Erwin Frand (1983) contrasts
the development of the personal computer with that of the
personal transporter the automobile. According to
Frand:
When transportation moved from horses and/or
feet to some form of mechanical power as the
prime mover, the emphasis was on large units
capable of hauling many people or large
quantities of freightthe equivalent of a
mainframe computer. As prime movers became
more efficient, buses and streetcars became the
form of municipal transportation. This is
analogous to the smaller mainframes and task-
oriented minicomputers. With the invention and
subsequent refinement of the gasoline engine,
personal transportation (PT) became a reality
(p. 23).
Frand cleverly compares the initial high price and
unreliability of the PT with that of the PC and suggests
that, like the PT, the PC has taken a while to develop
because when initially introduced they were high-priced,
relatively unreliable, required a great deal of expertise
to use and maintain and software (gasoline for the PT)
was sparse or unavailable. To continue the analogy,
Frand states that early users had to write their own
programs (build their own roads) because PC manufacturers
did not; and documentation and user instructions (road
maps) were practically nonexistent. Frand concludes by
suggesting that since, like the automobile, the personal
37


computer provides a high degree of personal freedom,
almost everybody will have one or use one.
The Business Side of Computer Technology
When personal computers first arrived on the scene
in the early 1980s, a large segment of the corporate
world viewed the new technology as personal productivity
tools that had little to offer in a business environment.
PCs were seen as little more than glorified typewriters
and interesting toys; the real computers remained the
purview of the data processing department (Piturro 1989).
However, few developments in the recent past have
had a greater impact on the business community than the
advent of personal computers. With these computers,
executives, managers, professionals, and clerical workers
can perform many sophisticated information-handling
functions that previously required either manual handling
or referral to a data processing department.
John F. Rockart, director of the Center for
Information Systems Research (CISR) at the MIT Sloan
School of Management, has stated that information is an
ingredient vital to good management. He suggests that as
the cost of computer technology continues to drop and
38


rapid improvements in the application of this technology
to office automation continue to rise, opportunities to
improve the effectiveness of business and nonprofit
management will abound. Rockart also states that these
opportunities are not easy to exploit and, if anything,
it has become more difficult to do so. The need to
prioritize resources committed to information systems
development, not integrating the various pieces of
information technology, difficulties that nonspecialists
have with the technologies, conflicts between users of
information and information specialists and top
management's lack of education for identifying
opportunities are problems Rockart identifies in the
information systems field (Synnott and Gruber 1981).
It is obvious that change has not always been a bed
of roses for those business organizations that have had
to make the switch from manual to computerized functions.
Many new challenges presented themselves that required
reorienting and retraining employees into performing
their old tasks in a new and unfamiliar manner, which
often caused a certain amount of dissention and
hostility. Productivity generally dropped off at first
and new training was required initially only of select
39


employees. Consequently, there was often a very wide gap
between management and employees, with management
jealously guarding their newly gained knowledge in the
interest of job security. This behavior exacerbated
employees' negative and hostile attitudes towards
managers and employers because the new technology was
perceived as generally intimidating and a threat to the
employees' job security. Only later, as more and more
employees were trained, and as management took a less
hostile approach toward employees, did attitudes towards
managers improve (Bradley 1989).
Freedman (1992) states that the more science and
technology reshape the very essence of business, the less
useful the concept of management itself as a science
seems to be. He suggests that the majority of new
managerial ideas like crossfunctional teams, self-
managed work groups and the networked organization are
either direct or indirect responses to the inadequacies
of Frederick Taylor's original model which advised
managers to never deal with workers as a group.
According to Gunilla Bradley, for organizations that
already had a design in effect, the switch to
computerization strengthened rather than weakened it.
40


She outlines three phases in the development of computer
technology which she believes affected organizational
structure in connection with automation.
Beginning in the late 1950s and increasingly in the
early and mid 1960s, the dominant philosophy was to
centralize the power of the data process. This was the
age of the mainframe computer. Larger and faster
computers were designed with relatively high memory
capacities, They were used more for purely
administrative applications than for scientific or
technical applications.
Toward the end of the 1960s all this changed when
minicomputers were introduced. They had neither the
memory nor the calculation capabilities of the mainframes
but were able to handle more technical applications and
tasks. Additionally, they were cheaper and easier to use
than the multipurpose mainframes, and even though they
resembled the mainframe in the way they performed, they
were much cheaper.
In the third phase of development, the 1970s, the
microcomputer played a major role as "component
technology." Microcomputers resulted in
decentralization because compact components were built
41


directly into the desktop system, making it both highly
complex and at the same time providing the user with a
wide spectrum of multitasking operations (Bradley 1989).
Two groups of business microcomputer users emerged.
The first type were users who saw the potential to make
their business more efficient by spending a few thousand
dollars on a microcomputer. These buyers typically would
purchase Radio Shack TRS-80, Commodore PET or Apple
computers. The second group of early users was found in
companies or departments within companies that needed a
computer and considered the $10,000 price tag of a small
business computer a bargain compared to minicomputers or
time-sharing services, the only alternatives available.
It became evident that microcomputers could actually be
functional. The microcomputer industry really started
escalating when IBM introduced its Personal Computer in
August 1981 (Fleming 1984).
The early 1980s saw new applications of information
technology for professionals, managers, and technical
workers. The introduction of small, stand-alone word
processors and personal computers made it easier for
nonspecialists to use data bases, to manipulate text and
quantitative data, to generate tables and graphic
42


displays, to utilize analytical software, and to
communicate with one another through a computer network.
Today, personal computers are being linked to one another
as well as to central computers so that they can be used
independently or in conjunction with corporate data
bases. These emerging systems also can be linked to
external data bases and communications networks; further,
they can cross organizational and national boundaries.
Office Automation
The term "office automation" is generally used to
refer to the use of integrated computer and
communications systems that support administrative
procedures in an office environment. Office automation
looks at the process of what people do in offices and
provides computer-based tools to allow them to handle
that entire process with as little effort as possible.
It also means that the tools have to be available to
every level of worker not just to some of them.
Office automation is process-oriented rather than task-
oriented. It supports the things managers and
professionals do, as well as the things clerical workers
do (Sharma 1993).
43


Automated office systems represent structured
methods of handling business text processing and
communications through an integrated network that may
include electronic spreadsheets, word processing for
generating correspondence, electronic message systems for
person-to-person communication, teleconferencing
services, facsimile transmission, electronic filing
systems, on-line calendar systems, and links to corporate
files and outside services (Shaiken 1984). The driving
force of technology forces us to change the way we work.
Now that this wave of information technology has worked
its way into practice, it is time to think about where we
are headed next (Applegate 1988).
Currently we are at the point where the office
automation movement has gained a lot of recognition. It
has become an item on Federal agendas as well as
corporate agendas. Executives in the public and private
sector alike are concerned about productivity. They
believe office automation gives them an opportunity to
improve productivity. They are concerned about the costs
of running their offices and the percentage of their
employees involved in what they consider basically
nonproductive work. People have begun to realize that
44


office automation is something that can help them out of
these problem situations (Cornell 1992).
In the automated office, not only can office work be
performed more efficiently, but the concept of office
work itself is altered (Strassman 1982; Garson 1988).
However, the greatest potential of office automation is
not expected to be from the improvement of clerical and
administrative tasks, but from the ability of managers to
gain increased control over their operations (Meyer
1982). Of all the new skills demanded of effective
managers in the 1990s, confidence and comfort in handling
information technology are at the top of the list (Keen
1991).
Office Applications of Computer Technology
Although there have been forecasts for the past 10
years that the "paperless office" was to become a reality
by now, it is estimated that 95 percent of records
continue to be paper-based. Automation has changed
people's work lives, but these changes are occurring very
differently from the way the office critics predicted.
The most obvious change has taken place at the desktop
level with the use of personal computers (Burgetz 1990).
45


A wide variety of business professionals engaged in
diversified tasks are being supported by personal
computers in a wide spectrum of applications. Although
these machines are, in fact, miniature data processors in
their own right, they do not process data in the
traditional manner that one normally associates with
mainframe systems. The functions provided by personal
computers might categorized into three large groupings:
(1) text and data management, (2) communications, and (3)
data collection and control (Madron 1983). Specific
applications within these functions might lie wholly
within one function or they might span two or more
functions. For example, a decision support system that
includes such features as query, analysis, reports and
graphics could require collection of data, organizing and
managing the data, and then communicating the results to
another entity (Meah 1984).
Some of the business-oriented applications that a
microcomputer should be able to perform are:
1. Text-editing and word processing
2. Mailing label printing
3. Inventory control
4. Accounts receivable/payable
46


5. General ledger accounting
6. Investment analysis
7. Report generation
8. Payroll calculation
9. Security functions
10. Cost projections
11. Marketing trends and analyses
12. A myriad of similar business uses (p. 16)
Peter A. McWilliams (1983) who has written on
personal computers and their uses at home and in the
business environment, delineates the major uses within
large business establishments into the following groups:
1. Word processing
2. Accounting
3. Electronic spreadsheets
4. Graphics
5. Data bases
6. Data banks
7. Electronic mail
8. Teleconferencing
9. Networking (p. 110)
Technological developments have brought about a
second industrial revolution based upon simple mechanical
47


activities characteristic of the first industrial
revolution. Developments central to this breakthrough
have been, first, the invention of machines that can
convert observations into symbols and can feed those
symbols back into machines in such a way as to regulate
their behavior in "real time" and, second, the invention
of the electronic digital computer. In the early days of
work-related computer technology, people assumed that the
technology associated with the work itself, the job,
could not be altered except to increase the efficiency
with which it was performed. But increased knowledge and
rapid technological change have brought many
unprecedented problems to business and public
organizations such as, how to employ new techniques
for managing information through the use of a computer
system and the effects of these and other techniques on
management (Kraemer and Pinsonneault 1993).
Computer technology has had a profound effect in the
Federal public sector. But, as noted in Chapter 1, the
growing concern in government is that improving service
quality is incompatible with maximizing employee
efficiency. Government personnel policies and civil
service systems are frequently accused of stifling
48


incentive and ignoring excellence, which can lead to
alienation, frustration and poor morale among government
workers, as well as lower productivity. As a result of
this growing concern over worker efficiency and
effectiveness, the Federal government has become
increasingly interested in new methods of motivating its
employees.
In recent years the Federal Government has spent
millions of dollars on improved computer technology in
the hope of improving employee productivity (Government
Executive 1994b). Many new strategies have been
considered and implemented some of which include: (1) the
use of new and improved technology (i.e., the personal
computer), (2) the improvement of operational procedures,
(3) an improvement in organizational structures, (4)
cultivating the work skills of employees and the
managerial skills of supervisors, and (5) intensive
efforts to motivate employees.
Office Automation and the Nature of the Organization
It has been suggested that technology is the most
significant factor in redesign of organizations through
office automation (Keen 1982). Computer technology may
49


be used initially in an attempt to make the current
organization function more efficiently. However, a
decision to adapt the structure to the technology in
order to realize productivity gains may well come after
the technology has been installed for quite some time.
When new technology is introduced, there are three
types of effects. One is the intended technological
effects. In the case of computer technology, these
effects are faster processing, fewer errors, and so on.
The second type is transient effects. These involve
organizational adjustments that gradually disappear. In
the case of the introduction of the telephone, for
example, one transient effect was the loss of privacy due
to the use of party lines. But this effect gradually
disappeared as party lines disappeared.
The third type is unintended social effects; these
are permanent and are generally unanticipated. Based on
in-depth studies of organizations, Sara Kiesler of
Carnegie-Mellon University (1986) observed that
hierarchical and departmental boundaries tend to break
down when people do not see others in their social
context. In face-to-face meetings, introductions are
performed as necessary, and each person is mentally
50


assigned to an organizational niche by the others. But
computer-mediated communications, such as messaging and
conferencing, do not provide this kind of social context
information. The cure for this may be video
conferencing, a new and expensive technology that is
beginning to enjoy increasing use in corporate circles.
Also, it takes longer to reach a consensus with
computer-mediated communications than with face-to-face
meetings. One reason is that computer communications are
asynchronous. That is, users read their mail and prepare
their replies at the times convenient to them.
Another reason is that there is less social pressure
on a minority to go along with the majority view. In
face-to-face meetings, social status of participants is
apparent. The person highest in the management hierarchy
tends to dominate, and other members generally defer to
the views of that person. In computer-mediated
communications, there is no reminder of status, so
minority views are more openly expressed (Kiesler 1986).
Although some technologies can improve
organizational communications, they have their limits.
Several researchers have argued persuasively that the
organizational trust and interpersonal context necessary
51


to achieve a true networked organization are not based
solely on electronic networks. Rather, relationships
must be initially constructed through face-to-face
meetings (Nohria and Eccles 1992).
Thomas W. Madron (1983), in a book on microcomputers
in business, makes the following observation: "The
extended use of high-performance microcomputers by the
end of the 1980s will contribute to major changes in the
way large organizations function" (p. 155). Madron
states that some of the changes will be benign and
beneficial while others will be problematic and
controversial. He suggests that those individuals who
can embrace the new technologies and use them effectively
will continue to make an impact on their organizations;
outdistancing those who will not or cannot accept change.
The fact that technological change rises in an
exponential growth pattern makes it difficult, if not
impossible, to achieve orderly progress, he concludes.
According to an article in Business Week (1994), a
few leading-edge companies have started to report
significant structural changes that personal computers
and other office automation technologies have made
possible. For example,
52


Firm
Hercules,
Inc.
Technology
Word Processing
Electronic Mail
Voice Mail
Videoconferencing
Communications
Impact
Reduction in
levels of
management between
the chairman and
plant foremen
"from a dozen to
six or seven."
FMC Corp. Voice mail for
salesmen and
sales managers
Amalgamation of
sales districts
and reduction in
levels.
Citicorp
Sophisticated
information system
to improve customer
service and provide
better information
for corporate
clients
Reduction in staff
from 2,650 to
2,150. Seventy
percent were
clerical; now 60
percent are pro-
fessionals. Ninety
percent were in
New York; now less
than 50 percent
are there.
Andrew
Politics of
M. Pettigrew (1973), in a book entitled The
Organizational Decision Making, has stated
that power in an organization comes to those who
understand how the structure operates and then make it
work to their advantage. He describes the "technical
gatekeeper" as a person who, in seeking support for his
demands and a power base in the organization, uses
information and technology to maximize the advantage that
these factors offer (Pettigrew 1973).
53


Two other authors have stated that the politics of
organizational decision making has three important
implications for providers of information services:
first, the success of the implementation is contingent
upon minimal effects on the current social structure;
second, planned changes in the social structure can be
carefully made; but third, enhancement of the relative
power of the information systems group can create
disaster (Bariff and Galbraith 1978).
Lynne Markus (1981), in research done at MIT, has
observed that technological innovation in the provision
of information can have clear effects on the power
distribution in an organization. She notes that
organizational power structures have a great deal of
influence on the behavior of individuals, groups, and
subunits contained in them; therefore, power structure
changes introduced by computer-based systems comprise an
efficient and fruitful starting point for identifying the
organizational impacts of systems and the causes of
resistance to them (Markus 1981).
The implications for managers and for the
organization are tremendous. The growing use of
technological tools is not the same as proper deployment
54


of resources. We are only beginning to learn how to use
these tools, and only a few organizations or individuals
have begun to assess the impact of computers and
communications for their way of doing business. But it
usually takes time for organizations to realize how
powerful technology can be. The walls set up in
organizations between managers and workers, between
functions constrain change, and managers are slow to
dismantle them (Jaikumar 1991).
Personal computers, however valuable a tool, are
still just a tool. New technologies alone will not
change anyone's behavior. Changing the organization's
information culture is the best way to implement
information technology, but it is also the hardest to
carry out. Information managers must begin by thinking
about how people use information, not how people use
machines. Too many managers still believe that once the
right technology is in place, appropriate information
sharing will follow (Davenport 1994).
Office Automation and Nature of Office Work
Information technology is the wave of the future,
and management must be prepared for its good and bad
55


aspects. Offices are flooded with personal computers,
video display terminals and other technological tools
designed to improve productivity. However, these tools
are largely products of mass-production methods, which
emphasize ease of manufacturing, not the user's health
and comfort (Harvey 1991).
In order to obtain maximum benefits from personal
computer use, implementors must consider the sociology of
office work. Human factors may preclude using systems to
their utmost advantage because doing so would upset
office hierarchies, etiquette, work methods, and even the
ergonomics of the office with accompanying levels of
stress (Goldfield 1982; Horowitz 1992).
Since this research explores the impact of
technology on quality of work life as well as on
productivity, it is appropriate to review the pertinent
literature which suggests that, while information
technology may increase productivity, it can also degrade
the work lives of those who use it.
The introduction of personal computers into the work
force has had mental and physical impacts on employees.
Along with every new innovation comes a "down-side," and
the personal computer is no exception. The introduction
56


of the personal computer into the office setting, while
appearing to be the panacea for improved office
productivity and performance and thus winding its way
through the entire organization, has brought with it new
problems which lie outside the formula of organizational
success. One new malady known as "technostress" has
managed to inhibit productivity and performance in a
large part of the work force. It is a computer-generated
form of physical and emotional burnout which comes about
as a result of the inability of an individual or
organization to adapt to the introduction and operation
of new technology (Braud 1982; McDonnald 1983).
Another related symptom is computer anxiety, which
results in dizziness, shortness of breath, and sweaty
palms. This phenomenon is referred to as cyberphobia.
Research has shown that stress and anxiety reduces job
effectiveness. It has also shown that this is related to
the use of a computer and thus can negatively impact
computer-related job performance, resulting in an
individual's completely avoiding the use of the computer
(Elder, Gardner and Ruth 1987). According to studies
conducted at George Mason University, 3 percent of the
people surveyed are affected by cyberphobia and 11
57


percent have suffered from technostress (Gardner, Render,
Ruth and Ross 1985).
There are other real dangers associated with
technostress and cyberphobia that affect productivity and
can lead to serious problems in the work place. Some of
these include sabotage, a drop in motivation, the quality
of work performed, a drop in morale, more errors,
absenteeism, conflicts with fellow employees, as well as
management and job turnovers. Elder et al. (1987)
indicate that technostress and cyberphobia lead to
resistance to computer technology, which can result in
scheduling delays and inhibits innovative creativity in
computer applications. All in all, these pathologies
would result in reduced productivity.
As indicated earlier, the use of personal computers
might also affect "office sociology," the web of
interpersonal relationships within the office
environment. In a number of ways, the PC can be very
threatening to the middle manager. Roles can change when
the computer is available to the manager and the employee
alike. Questions of social status and job displacement
can get in the way of efficiency (Beaufort 1983).
58


The very act of putting a professional online to
information previously not accessible to him/her could
change the nature of his/her job. Most white collar jobs
can be defined as "knowing more about a specific area
than anyone else in the department;" thus, knowledge
workers spend a great deal of their time collecting,
distilling, and communicating information in their areas
of expertise (Haeckel 1983). To the extent that the
office of the future provides bosses, subordinates, and
peers with access to the same information, knowledge
workers' jobs will have to change from handling
information to exploiting it (Drucker 1991).
Shoshana Zuboff (1988), assistant professor at the
Harvard Business School, interviewed some two hundred
employees, supervisors, professionals, and managers from
several different organizations in three countries. She
researched ways that people at distinct organizational
levels respond to their work when it has been
fundamentally reorganized by information technology. She
observed that as computers moved into offices:
1. Many of the managers and professionals were
wary of the technologies that they perceived
were encroaching on their judgement, their
59


freedom/ or the "artistry" of their
professional assessments.
2. Instead of feeling that increased information
augmented their power, people resisted the
information systems that they saw as limiting
their freedom or increasing the measurability
of their work.
3. People's reactions to computers were different
from those anticipated. People reacted
negatively to having to handle their work the
way the machines prescribed. The range of
choices in how to handle their work became
limited.
4. Information technology challenged old
procedures. Computers altered the social
structure of the organization. The machine
itself became the primary focus of interaction.
This caused people to feel isolated in an
impersonal environment.
5. Managers became reluctant to make decisions on
the basis of information that their supervisors
received simultaneously. They felt that they
might be "second-guessed" by their bosses.
60


Change can be a threat or an opportunity. It can be
anticipated or reacted to. Although technology is
critical to the reengineering process going on in today's
offices, it cannot drive change only enable it.
Hence, technology should be added only after work
processes have been redesigned (Corbin 1993).
Information technology and change are now synonymous. In
many organizations, taking charge of change is impossible
without taking charge of technology (Keen 1991a).
Personal computers may endanger the way things are done
in the organization. This could upset comfortable,
familiar routines, and might force changes in management
style, role, or function. Much of this negative impact
can be avoided through the knowledge that it could occur,
and by taking corrective action during the actual
implementation of the personal computer into the work
environment. A thorough program of familiarity and
training can go a long way in reducing many of these
initial fears (Schnack 1983).
Although there is no doubt that personal computers
can have their negative side effects, respondents in a
study by Elder et al. (1987) also indicated that
computers did, in fact, save them time and improved
61


overall office productivity (though not necessarily the
productivity of every individual in the office). Most
also indicated that, since computers have now become an
important aspect of the office, they would not want to
part with them.
While much of the literature has focused on
individual productivity (Blair, Cohen and Hurwitz 1982;
Engel et al. 1979; White 1977), little has been said
about potential changes in role definitions or in quality
of work life. How can the pressure for higher
productivity be balanced against concerns about the
quality of work life? How can the dull job of data
processing be enriched? And how can the people who hold
these jobs be made to feel good about themselves and
about their work? Employees need to feel that their own
needs have been met within the organization before they
can become enthusiastic about meeting the needs of
customers (Schneider and Bowen 1993). But all too often
technology is applied in ways that make office work dull,
lacking in challenge or appeal. The machines need to be
used to enhance the job of workers, not the workers
twisted to conform to false ideals of what is good for
the machines (Lasden 1983). Without a clear idea of what
62


must be done to handle the human problems of automation,
undesired results such as unwanted turnover, absenteeism,
lower productivity, bitterness or formation of anti-
management sentiments may result.
John Naisbitt, in his bestselling book Megatrends
(1982), describes the way people relate to new technology
in what he calls the "high tech/high touch" formula. He
suggests that whenever new technology is introduced into
society, there must be a counterbalancing human
response that is, high touch or the technology is
rejected. The more high technology around us, the more
the need for the human touch.
Basically, the "high tech/high touch" concept is no
more than common sense. Still, many managers implement a
"high tech/no touch" approach, resulting in a
demotivated, poorly performing work force.
The impact of automation on the human element does
not result from the features of technology, nor from the
characteristics of people and organizations. Rather, it
arises from the way particular system features interact
with a particular organizational setting or how they
function within the organizational climate or culture.
Managers can improve their organization's effectiveness
63


by changing their organization's climate and culture, but
the process for doing so is slow and difficult. This is
because managers must modify the practices, procedures,
and behavior by which they manage before there will be a
change in the climate. And, climate change precedes
culture change (Schneider et al. 1994). Herbert Simon
believes that the central problem that managers will have
to deal with in an information society is not how to
organize to produce efficiently, but rather how to
organize to make decisions (Simon 1973).
As information technology advances and knowledge
increases, the Federal government also will enter into
new programs and assume new responsibilities, and in
consequence, its organizational structure will evolve to
accommodate the new environment (Kaufman 1987). A
shortage of trained personnel will also result as the
demand for technical expertise increases (Government
Computer News 1994b). This supply-and-demand equation
will result in increased labor costs and could become an
impediment to the use of computer technology or, at the
very least, cause a slowing in organizational acceptance
(King 1982).
64


With both public and private employers seeking to
reduce labor costs, a viable solution may be to
substitute less experienced, educated, and skilled
employees for more skilled employees by using technology.
This could also indirectly assist in job enhancement by
providing the more experienced employees with other
opportunities within the organization.
Information technology can aid this process in at
least two ways. First, it can incorporate some skill and
knowledge directly, requiring that users have less. Much
special-purpose expert-system software avowedly has this
aim (Allen 1988). More generally available technology on
both the shop floor and in the office can also be used in
this way. Second, information technology can complement
Taylorism the fragmentation of an integrated job into
relatively autonomous components, with different laborers
performing different components by aiding in the
reassembling of information that is generated by separate
individuals. Case studies and analyses by Glenn and
Feldberg (1977), Nussbaum and Gregory (1982), and
Murphree (1984) document occasions where information
technology has been used to incorporate workers' skills
65


and to reintegrate information produced through
fragmented jobs.
On the other hand, some researchers and analysts
have argued that computerized work is manifestly more
fulfilling than conventional work (Giuliano 1982; Spinrad
1982; Gibson and Jackson 1987). According to this
position, the most routinized work is most likely to be
automated, eliminating jobs requiring low skill and
eliminating the most boring and repetitive tasks within
more skilled jobs (Adler 1983). Poppel1s analysis of the
benefits of office automation for sales and information
workers is typical (Poppel 1982). After studying 15
large U. S. organizations, he concluded that a
salesperson's time is wasted on travel, missing contacts,
finding out information, and filling out forms, while the
time of many managers and professionals is similarly
wasted on meetings and clerical work. According to his
analysis, office automation technology can rescue some of
that wasted time and make jobs more rewarding.
In a more sophisticated and balanced analysis,
Attewell (1985) has shown how computerization has
eliminated some of the routine work of insurance
examiners, such as calculating deductibles or identifying
66


potential duplicate payments, while at the same time
leaving examiners more time to make decisions about
dubious claims. Zuboff (1988) has argued that
computerization has added intellectual content to work by
making activities more abstract. Baron (1987) also
discusses ways in which intellectual content is being
reintroduced into white collar work.
What leads to these contradictory results? The
domain is far more complex than the idealized rhetoric of
deskilling or upgrading imply. As Attewell and Rule
(1984) point out, the large number of variables
moderating the effects of technology imply that both
deskilling and upgrading are occurring within white
collar occupations. To determine which tendency
predominates would require quantitative studies that
sample representatively from firms and workers.
Thus, the advent of the computer has given rise to
thoughtful and valuable estimates of the changes they
have brought into organizations and labor deployment. It
has led to promises, albeit somewhat far-fetched ones, of
a consequent escalation in productivity increases,
leading shortly into a world without work; and to fears,
not unfounded but possibly excessive, of the consequent
67


change in the status of the individual. To a central
computer, a person becomes a collection of registered
attributes. Additionally, vast amounts of electronic
communication and computing equipment have already been
installed for business data processing. Yet, as was the
situation in this study, the equipment is used almost
entirely for tasks of the type that were previously done
manually. Emphasis has been on doing more data
processing within the earlier patterns, or on reducing
the cost of work already being done. This is especially
true of clerical functions. Peter Sassone, an economist
at Georgia Tech, suggests that in many offices,
technology such as personal computers actually reduced
productivity because managers and professionals had to
use their personal computers too often on lower-level
work (Aley 1992).
Technology prognosticators seem to agree that the
general trend toward an automated work place will see the
development of microcomputer-based systems tailored to a
particular professional, manager, or secretary. Much has
been written on how human and technological resources
will interface. For example, telecommuting has been
described in some circles as the substitution of
68


communications capabilities for travel to a central work
location (Minahan 1994). And, in fact, the Federal
government is currently conducting limited tests in this
area. The computerized work place permits many office
workers to be potential telecommuters in that their work
can be performed remotely with computer and
communications support without ever leaving home. But
the impact of this technology on the work force deserves
further consideration. Yet to be examined are the
behavioral, organizational, and social issues surrounding
the office of the future. While some limited research
has been done with regard to impacts on human resource
management in the private sector, virtually none has been
done in the Federal public sector. But what is certain
is that the changes computers have brought to the
workplace are here to stay. Information Technology is
transforming how organizations operate. Managers who
stick their heads in the sand and "leave it to the
experts" run the risk of being left far behind in the
times ahead (Keen 1991b).
With the continued developments in technology, every
executive and professional will be dealing more and more
with computers and other tools of office automation. To
69


make the most of the new technology, managers and their
employees need to be team players more than ever before.
Every single tool we have in the office is undergoing
radical change. We are moving quickly toward integration
of many of these tools into a framework that allows fully
integrated office automation. One would hope that
organizations that use the right office automation tools
today will find that within a very short time, their
people are accomplishing much more.
While in the 1970s the emphasis was on making the
clerical and support functions more efficient, in the
1980s the trend in office automation has been toward
increasing the effectiveness of the organization by
increasing the effectiveness of managers. This trend is
expected to continue into the 1990s but as a more
balanced quest for increased productivity opportunities
from both employees and management alike (Kraemer and
Pinnsonneault 1993).
Productivity and the Personal Computer
How does computer use relate to productivity gains
in the work place? In order to answer this question, one
needs to review the importance of productivity and how it
70


relates to employee computing. There are many ways to
assess the productivity of the United States workforce;
some of these were discussed in Chapter 1. The Bureau of
Labor Statistics measures the productivity of U.S.
workers quarterly. It employs a variety of measures to
assess overall productivity. These include overall
output, which is a measure of the total output of goods
and services produced; output per hour worked, which is a
measure of the efficiency of the workforce; and unit
labor costs, which are a measure of the cost of labor to
produce a single marketable item. During much of the
past decade, U.S. business ranked last among the twelve
leading industrial nations in just about every measure of
productivity (U.S. Department of Labor 1989; U.S.
Department of Commerce 1991). As a result, a renewed
emphasis has been placed on improving the productivity of
U.S. businesses through more effective production
practices. While the preponderance of the effort has
been devoted to streamlining the manufacturing process to
better compete in international markets, such efforts are
often so expensive that in the short term they exacerbate
the situation. Despite renewed efforts, the productivity
of the U.S. worker in manufacturing industries is not
71


expected to increase significantly in the future
(Workforce 2000 1987).
The overall key to improving U.S. productivity is
increasing the efficiency of the service sector
workforce. One of the greatest challenges in managing
white collar work is to measure productivity in a way
that adequately captures either the value being created
by specific work activities or their real cost (Hori
1993). A study supported by the United States Department
of Labor concluded that economic prosperity in the next
decade will depend more heavily on how fast worker output
increases in health care, retailing, government and other
service industries than on gains in manufacturing
(Workforce 2000 1987). Since the size of the Federal
government exceeds the largest U.S. corporation by a
factor of 2.5, and government accounts for 25 percent of
all economic activity in society, the onus will rest with
the Federal government to provide leadership for
restoring economic prosperity (Scott 1987).
How will government become more efficient?
Significant productivity gains can be made through the
application of computer and information technology.
Japan, which ranked first in productivity among the
72


twelve leading industrial nations during the past decade,
is well known for aggressively developing information
technology. The Japanese believe that such technology
holds the key to future competitiveness across a whole
range of industrial sectors (robotics, machine tools,
telecommunications): not only high technology, but also
old-line industries (steel, automobiles, chemicals) and
even the services (banking, insurance, distribution)
(Okimoto 1989).
However, even Japan, which built its economic
powerhouse on blue-collar productivity, has experienced
problems in white-collar productivity. Even where
technology has been installed, worker productivity is low
across the full spectrum of service industries. At gate
checkin for domestic flights at Haneda Airport, for
example, boarding passes are inserted into a machine that
retains the airline's portion and returns the passenger
stub. This simple transaction, which could easily be
operated by passengers themselves, is handled by two
workers. The first takes and inserts the pass, and the
second hands back the stub (Hori 1993).
By the year 2000, economic growth in the United
States is expected to depend more on the application of
73


automated technology to deliver government and business
services more efficiently than on improvements in
manufacturing. According to a study conducted by Peters
and Waterman, U.S. companies and government agencies are
being stymied by their information systems which, they
state, inhibit timely action (Peters and Waterman 1982).
McFarlan (1984) suggests computer technology can provide
a competitive advantage to businesses which, over time,
will significantly cut costs.
Reasons for the crisis in the manufacturing sector
are clear: high labor costs, antiquated and inefficient
production technology, markets closed to American
manufacturing, and often an over-valued dollar. The
situation in the United States information sector is more
confusing. Information systems and personal computer
technology could cut costs and improve worker
productivity; they are limited, however, by a serious
challenge from the very managers and employees for whom
they were acquired.
In the 1980s, information resource managers were
seduced by promises of what technology could do for their
agencies. Federal budgets sprang open for the latest in
technology that was supposed to perform automation
74


miracles. But as government productivity trends remained
anemic, it became clear that state-of-the-art hardware
and software could never substitute for good management
and organization (Government Executive 1992).
Now, the Federal sector is beginning to embrace
business-process reengineering in an effort to meet
challenges brought on by budget and staff cutbacks (NPR
1993). Reengineering is centered around the
organizational structures that govern the flow of work.
To succeed, therefore, it must engage the energies of key
managers in the organization, not just the information
technology chiefs. Although technology plays a key role
in reengineering efforts, it is being added only after
processes have been redefined. "We've learned that
technology for technology's sake does not increase
productivity or mission effectiveness," says Henry
Philcox, chief information officer at the Internal
Revenue Service. "If you start with a mess and simply
add technology, you end up with an automated mess"
(Government Computer News 1992).
The U.S. General Accounting Office (GAO) states the
most frequent criticisms of organizational computing are
in three general areas, individual problems,
75


organizational concerns, and management problems
(Datamation 1983). Individual problems, such as the lack
of effective training for employees and the lack of
employee persistence, are the most commonly cited and
intractable problems (Levine 1991). Organizational
concerns plague many employees. These concerns include
the incompatibility of hardware and software components
often competitively procured from the lowest bidder,
insufficient numbers of technical support staff to assist
in trouble-shooting difficulties, and the overloading of
newly acquired computer resources by employees who are
finding new and unanticipated ways to use their personal
computers (Anthes 1992). Finally, management problems
exist when organizations fail to establish measurement
standards for assessing productivity and determining the
proper balance of resources between employee computing
and traditional data processing activities (U.S.
Department of Commerce 1983). Often, documents created
on one workstation cannot be interchanged with other end-
users without the loss of information. As a result of
the myriad of problems, some researchers have concluded
that the personal computer can introduce more problems
76


into an organization than it solves (Business Week 1985;
Lacey 1990).
Another GAO report (1994) concluded that many
Federal information technology systems were not being
well managed. The report cited massive cost overruns,
low performance levels, unreliable systems, schedule
delays and poor access to data. GAO criticized managers
for not properly weighing the costs and benefits of
systems before procuring them. Among the many agencies
berated in the report was the Federal Aviation
Administration, which GAO said could not adequately judge
whether new computers ordered for the Los Angeles-basin
area would have sufficient capacity. The Immigration and
Naturalization Service was lambasted for having computer
systems that could not share data, resulting in agents
not being able to retrieve information needed to
identify, apprehend and deport illegal aliens.
The Department of Education was criticized for not
having adequate procedures for checking the accuracy and
completeness of data in its loan databases.
Consequently, under the Stafford Student Loan Program,
students who had defaulted on previous loans obtained
$109 million, and students who had reached their legal
77


loan limits also received millions of dollars. The GAO
also said that poorly managed systems at the Department
of Justice meant that the security of highly sensitive
informationsuch as names of defendants, witnesses,
informants and undercover law-enforcement officials
could not be ensured (U.S. General Accounting Office
1994).
All these agencies and, perhaps, dozens more are
candidates that could benefit from a thorough analysis of
all tasks and for determining how the application of
technology might benefit process reengineering. This is
not to say that there are not already some measures of
success in the public sector. For example, the Veterans
Administration has reorganized its records and placed all
files under each veteran's name so when a veteran's
social security number is typed into the computer, a
complete record instantly is made available. And
Californians can pay traffic tickets by typing their
credit-card numbers into computers located in their
neighborhoods. Automated teller machines in Minnesota
dispense welfare payments, while food stamps are
available through point-of-sale terminals in Maryland
Government Executive 1995). The IRS is also achieving
78


productivity gains by allowing taxpayers to file their
tax returns using their personal computers, and IRS
auditors are being equipped with lap-top computers to
help them become more productive on the road f Government
Computer News 1994a). In these examples, business
processes have been subdivided into many components and
quality management principles and technology have been
combined to achieve success.
Davenport (1992), in discussing process innovation,
illustrates how the IRS has increased collections by 33
percent with half the staff and one-third the branch
offices. He also indicates that IBM has reduced the time
to prepare price quotations on buying from seven days to
one while preparing ten times as many quotations. He
suggests that by fusing information technology with new
ways of managing people, organizations have achieved
major reductions in process cost or time, as well as
major improvements in quality.
Office Automation and Worker Productivity
One of the major functions of personal computers in
the office is to make white collar workers more
productive. At least part of the key to white collar
79


productivity is to provide people in these occupations
with the technology to do their tasks well. In a 1981
speech, John W. White, assistant vice president for
information systems at Texas Instruments, indicated that
the average cost for a one-page message communicated
through Texas Instruments' PC network was two to three
cents. This cost should be compared to the average cost
of preparing and sending a standard business letter: $6
to $7 (Madron 1983). This indicates that the benefits of
work stations using PCs include efficiency as well as
effectiveness components.
The concept of the personal work station has been
enhanced greatly through the introduction of what has
been termed "desk management" software. This software
provides easier access to multiple functions by
integrating and simplifying the variety of individual
tools one might desire to use. In addition, it provides
new functions, such as reminder files, electronic
calculators (on the screen), personal schedulers, and
note pads. Arthur Ross, CEO of HAL Research
Laboratories, has stated, "My desk is 50 percent clearer
because I keep all my notes on SideKick desk manager. It
is the handiest piece of software I've used in the six
80


years I've been involved with computers" (Shipley 1985,
p. 105).
It has been generally believed that computers would
increase the efficiency of operations by reducing staff,
reducing costs, and increasing the speed of performing
various operations. Early studies indicated, however,
that the benefits of automation had been fewer than
anticipated and the costs had been far greater (Kraemer,
Dutton and Northrop 1981). Some of the current
literature also indicates that corporate investments in
personal computer technology have not paid off in
increased productivity. It is popularly believed that
the use of personal computers increases productivity, but
there are no concrete studies to support such gains,
agrees Stephen Banker (1987). For example, the growth in
office productivity in the 1980s is far below the average
for America's post-World War II economy, according to
Daniel Lyons (1989) in an article in PC Week. Bozman
(1993) states that productivity in the white collar
sector did not increase despite substantial investments
in technology during the past 15 years. "We poured
cement over the service industry with a lot of data
processing," she states. But Bozman remains optimistic
81


about office productivity. "We're going to see a huge
jump in productivity by 2001," she predicts. "We're
getting smart and starting to automate business processes
instead of just tasks" (Bozman and Booker 1993, p. 28).
Since the turn of this century, offices have
deployed more sophisticated technologies but have not
changed the nature of the work that office personnel
perform, said John Mahnke (1990). Mahnke suggests that
companies may be required to turn to behavioral
scientists to study the processes of work and find ways
to help people function more effectively when doing
intellectual work. He noted that the Japanese, after
notable success in rethinking the factory, are now
involved in transforming the office into an environment
for the "knowledge worker" (Drucker 1991).
Mahnke goes on to say that management in the United
States mistakenly thinks offices are analogous to
information factories and can be improved by factory
methods, such as work flow analysis, work
rationalization, massive new technology investments and
quality circles. Such approaches may yield incremental
improvements, he states, but not breakthroughs. To
transform, companies need to prune, simplify and improve
82


business practices before automating them. Once
management and operations have been revised, automation
may sometimes not even be necessary (Mahnke 1990).
David Kirkpatrick (1992) states that managers have
been frustrated for years by their inability to prove
that office computing increases productivity. His
primary evidence: while service companies spent about
$800 billion on information technology in the past
decade, service productivity growth over that period has
been a measly 0.7%. He suggests that part of the problem
is that service companies have not made the staff
reductions that the new hardware should have made
possible. Kirkpatrick makes a similar point about
manufacturing. He points to a study by economists Gary
Loveman at the Harvard business school that reported
five-year cost and productivity data for 60 large
manufacturing firms and found no evidence that
information technology improved productivity (Kirkpatrick
1992).
However, in an almost contradictory note,
Kirkpatrick presents other evidence that there is, in
fact, improvement in productivity. For example, after
years of complaints that investment in desktop computing
83


does not pay off in productivity, Boeing Aerospace
Company found a radical new way of harnessing creative
energy. They began having electronic meetings and
invested in personal computer software that, they claim,
has reduced the time needed to complete a wide range of
team projects by an average of 91%. And there is more:
a study of business personal computer use reported in
1984 found that of businesses with sales of less than $5
million, 85% estimated their productivity to be at least
1.5 times greater than when they first began using
personal computers (Fleming 1984).
One can also argue that the computer can bring
intangibles that, while not contributing to the bottom
line, still end up as pluses for a company. There is no
doubt that any company processing a large number of
orders benefits somewhat from computerization, if only
through cutting down on errors. Increased worker
satisfaction is also frequently cited as an intangible,
on the assumption that happy workers are productive
workers (Trice and Byers 1992; Schein 1992).
84


Productivity in Office Work
Does an employee's overall satisfaction with a
personal computer environment increase his/her
productivity? What role do variety, challenge, and job
satisfaction play in improving productivity? To answer
these questions, at least two different approaches are
possible. Ideally, one would attempt to evaluate the
productivity of a Federal employee based directly upon
the amount of work completed and its value per unit of
time. In theory, productivity should be a matter of
economics: the salary paid the employee is subtracted
from the organizational benefit to establish a net value.
In practice, however, such a determination is much more
difficult due to an inability to quantify the
organizational benefit of white-collar employee
productivity and the intangible nature of work in the
Federal government.
A second approach could attempt to utilize a
surrogate measure of employee productivity. Under
certain conditions, "system utilization," defined as the
percentage of time that an employee actively employs a
personal computer for work-related activities, can be
used as a surrogate measure of worker productivity. If
85


employees can be convinced that computer utilization
contributes to their productivity, the system will be
used more often, even in situations where it is not a
requirement of their job (i.e., finding new, innovative
ways not related to daily routine work). Employee
interaction with the personal computer would be a
manifestation of their beliefs regarding its utility and
effectiveness. This study reviewed computer utilization,
including output (frequency and ease of tasks) as well as
overall job satisfaction in an attempt to measure overall
employee productivity. What follows is a general
overview of earlier research relevant to these types of
measurements, together with an assessment of the
limitations of this previous research.
Baroudi, Olson, and Ives (1986) studied the impact
of user involvement in information system development on
two variables, system utilization and overall user
satisfaction with the system. Their research examined
the direction of causality between overall satisfaction
and system usage. In the past, the assumption was that
user involvement in the design of information systems
would result in increased system usage and a favorable
perception of the system. Their study surveyed a
86


homogeneous sample of 200 mid-level production managers
who were assigned similar responsibilities in many
different organizations.
Using path analysis, Baroudi, Olson, and Ives
determined that user involvement directly affects a
user's overall satisfaction with the system. In
addition, it has direct and indirect effects on system
utilization through overall user satisfaction with the
system.
While the study identified a positive relationship
between user satisfaction and system utilization, its
ability to generalize beyond the respondent population is
limited because the sample was restricted to mid-level
production managers in manufacturing industries.
In addition, the researchers utilized a perceptual
rather than an objective measure of system utilization,
and the instrument was not rigorously validated in the
study. Techniques providing a more objective measure of
computer system utilization have been considered by other
researchers.
Lee (1986) conducted an investigation of computer-
usage patterns by employees in manufacturing and service
industries. He developed a framework to categorize
87


different types of computer users by background,
training, formal education, use of the personal computer
and organizational position. His major concern was to
answer the following questions: What type of individual
uses a personal computer, how much does (s)he use it and
what problems does (s)he face? Usage was measured by two
variables, the number of hours spent per week using the
personal computer and the number of different software
applications used. The various respondents were selected
by management in 12 different firms to take part in the
survey based on their use of and expertise with the
personal computer. The data were collected from 311
respondents over an eight-month period.
Lee found that the number of training courses
completed was strongly related to the use of the personal
computer in manufacturing and service organizations.
Education level completed was not significantly related
to either computer utilization or the number of different
applications used. It appeared that the number of
training courses, both formal and informal, was more
relevant than the formal degree attained.
In addition, Lee found that users in non-technical
positions used a personal computer significantly more
88


than those in technical positions. The mean was 10.15
hours per week for non-technical users and 8.39 hours per
week for technical users. Lee's study provided empirical
data on computer usage within both the manufacturing and
service sectors.
Igbaria, Pavri, and Huff (1989) studied the
relationship between system utilization and alternate
approaches to training including self-training, in-house
training, and vendor training. The researchers examined
the relationship between a computer user and computer
system utilization in a survey of 766 managers in 54
different companies who had access to a personal computer
in the daily execution of their job. They employed five
measures of personal computer usage: the amount of time
used in decision making, daily use in hours per day,
frequency of use, the number of software applications
used, and the level of expertise or sophistication
exhibited by the computer user.
Igbaria et al. found that the degree of personal
computer usage for all five variables was positively and
significantly related to Self-training, in-house training
and vendor training. This study contradicts the
89


relationship between formal education and system
utilization that was identified by Lee.
In addition, this study identifies a strong,
significant and positive relationship between each
category of training and the level of user
sophistication. However, Igbaria et al. did not measure
whether employee training, differentiated by the level of
user sophistication, increased computer use.
Igbaria, Pavri and Huff restricted their survey to
managers. Lee, however, had previously established that
managers comprise only a small percentage of the PC-using
population (Lee 1986, p. 314). Other studies have
concluded that because managers are time-constrained,
their level of computer proficiency is relatively low
compared to other classes of computer users (Djurdjevic
1986). As a result, limiting their study to managers
weakens this study significantly and makes it difficult
to generalize beyond the respondent population.
Mawhinney and Lederer (1990) evaluated the impact of
individual, organizational, computer system and process
variables on personal computer system utilization.
Organizational variables included the user's level in the
organization, span of control, type of position and
90


satisfaction with the computer system. Computer system
variables included satisfaction with the ease of
learning, the ease of use, the system accessibility, the
system response time, and the overall system accuracy.
Process variables included participation in the
acquisition of the personal computer, training, and
technical support.
A survey was conducted consisting of 373 managers
enrolled in an executive MBA program at a northeastern
university and 475 members of a professional organization
that deals with "productivity and inventory management."
One hundred and five completed surveys were returned; 66
from managers enrolled in the executive MBA program and
39 from members of the professional organization. The
overall response rate was 12 percent when non-users and
non-managers were excluded from the survey.
Several variables, span of control, satisfaction,
and training, were significant within one sample group
and not the other. Two relationships were significant
and consistently supported across both groups of
respondents. Computer utilization was found to be
negatively related to the span of control, and positively
91


correlated with overall system satisfaction for members
in only one of the two respondent groups.
Like the preceding studies, Mawhinney and Lederer
included only managers in their survey population. As
previously stated, this limits the results of their study
as a guide to understanding of the impact of personal
computer technology on all types of users.
The Mawhinney and Lederer response rate of 12
percent demonstrates a very high degree of sampling
mortality. Excessively high levels of experimental
mortality suggest that the conclusions of the survey may
be easily biased. In this case, the length and
complexity of the questionnaire may have intimidated and
overwhelmed novice computer users who thereby opted not
to complete the survey. As a result, their survey is
self-selecting. Presumably, users who were comfortable
with personal computer technology chose to complete the
survey, while, those with less experience did not
complete it.
Finally, their selection of two heterogeneous
groups, without sufficient experimental controls,
precludes their ability to explain the experimental
differences reported between the two groups. Although
92


the demographics of the respondents were sufficiently
controlled in their questionnaire, selecting all members
of an MBA program and a "production and inventory
management professional organization" does not provide
insight into how to compare the results of the study with
the population in general. In addition, the selection of
these groups was biased towards older employees, higher
educational levels, and greater levels of computer user
sophistication.
Thompson, Higgins and Howell (1991) evaluated the
relationship of organizational factors, including
supervisor encouragement, peer pressure, and cultural
factors such as organizational norms, on the utilization
of personal computers. A survey was conducted consisting
435 managers and professional workers within nine
divisions of a large multi-national firm. The
respondents were selected by their management to receive
the surveys based on their known expertise with personal
computers. A total of 278 questionnaires were returned.
Since their survey was designed solely for computer
users, thirty-six respondents reported that they did not
have access to a personal computer, reducing the overall
response rate to 49 percent. The data were collected
93