Personal Use of Work Computers:

Distraction versus Destruction

PAUL M. MASTRANGELO, Ph.D.,

1

WENDI EVERTON, Ph.D.,

2

and JEFFERY A. JOLTON, Ph.D.

3

ABSTRACT

To explore definitions, frequencies, and motivation for personal use of work computers,
we analyzed 329 employees’ responses to an online survey, which asked participants to self-
report frequencies for 41 computer behaviors at work. This sample (65% female, 74% Euro-
pean ethnicity, mean age of 36 years) was formed by soliciting participants through Internet
Usenet groups, emails, and listservs. Results support a distinction between computer use that
is counterproductive and that which is merely not productive. Nonproductive Computer Use
occurred more when employees were younger (r =

0.31, p < 0.01), had Internet access at

work longer (r =

0.16, p < 0.01), and had faster Internet connections at work than at home (r =

0.14, p < 0.01). Counterproductive Computer Use occurred more when Internet access was
newer (r =

0.16, p < 0.01) and employees knew others who had been warned about misuse (r =

0.11, p < 0.05). While most employees who engaged in computer counterproductivity also
engaged in computer nonproductivity, the inverse was uncommon, suggesting the need to
distinguish between the two when establishing computer policies and Internet accessibility.

730

C

YBER

P

SYCHOLOGY

& B

EHAVIOR

Volume 9, Number 6, 2006
© Mary Ann Liebert, Inc.

INTRODUCTION

P

ERSONAL COMPUTERS

have quickly become the

mainstay in the American workplace. The

U.S. Federal Reserve estimated that the use of in-
formation technology and production of computer
products contribute $50 billion annually to the
country’s work output.

1

One reason for the rapid

growth of computer use at work is their ability to
connect through the Internet. A recent study found
that 88% of organizations implemented Internet ac-
cess between 1995 and 2000,

2

and they increasingly

use the Internet to communicate with employees,
applicants, and customers using, email, instant
messaging, and other interfaces. The ease of com-
munication and exchange of work documents has
lead to an increase in work done away from the
workplace, also known as telecommuting, which

is associated with increased productivity and in-
creased morale.

3

Computers have undoubtedly changed the way

people work, but they have also changed the way
people avoid and sabotage work. The Federal Bu-
reau of Investigation (FBI) and the Computer Secu-
rity Institute’s 2001 Computer Crime and Security
Survey found that 91% of participating computer
security practitioners detected employee abuse of
Internet access privileges at their workplace, in-
cluding inappropriate use of email systems and
downloading pornography.

4

Among people who

view online pornography or engage in sexually
charged online chat sessions, 20% of men and 12%
of women report using their work computers for a
portion of their sexual activity.

5

A representative

from an Internet surveillance company recently es-
timated that “recreational web surfing” cost U.S.

1

Genesee Survey Services, Rochester, New York.

2

Department of Psychology, Eastern Connecticut State University, Willimantic, Connecticut.

3

Kenexa, Lincoln, Nebraska.

14356c03.pgs 11/28/06 10:22 AM Page 730

businesses $5.3 billion in 1999.

6

Indeed, 14% of all

web connections are initiated at work.

7

Employers

are also seeing an increase in “cyber moonlight-
ing,” where employees use work hours to earn
money by searching the web or winning sweep-
stakes.

8

Gambling, stock trading, and music swap-

ping represent other distractions from work, but
employers must also contend with their workers at-
tempting to access private materials and spreading
malicious viruses. All of these employee behaviors
can be classified as computer deviance, which we
define as the personal use of a work computer that
violates formal or informal organizational norms.
Although the term deviance may be applicable in
the sense that employees deviate from an em-
ployer’s expectation for appropriate work behavior,
some forms of personal use may indirectly maxi-
mize performance and be sanctioned by employers
(as a coffee break is).

Employers’ reactions to personal use of work

computers are indeed mixed, as some companies do
“recognize and tolerate” a certain degree of personal
use.

9

Many employers have developed computer

use policies with various degrees of restriction, re-
view, and/or disciplinary action. A recent American
Management Association

10

survey suggests that 36%

of employers review files on work computers, 47%
review email, and 63% monitor Internet connec-
tions. Over 50% of these companies had disciplined
or dismissed at least one employee for email or In-
ternet misuse, and 38% used software that prevents
employees from accessing Internet sites deemed in-
appropriate or unrelated to the job. Each of these
managerial responses has a psychological and finan-
cial cost for the employer.

An employer’s use of monitoring devices poses a

threat to employee privacy, which can make an orga-
nization less attractive to employees.

11

Certainly the

expense of buying and staffing monitoring equip-
ment directly affects profitability. Furthermore, it
is uncertain whether terminating an employee for
“deviant use” of work computers improves organi-
zational performance because little is known about
employees who engage in these activities. One can
make the case that productive employees use a com-
puter for activities that may seem unrelated to work
when in fact these behaviors enhance computer
skills, increase business contacts, maintain work-life
balance, and alleviate work stress. The motivation of
employees who engage in these behaviors and the
environmental antecedents of these behaviors are
unclear at this time.

The purpose of this study was to (a) operationally

define specific instances of personal computer use at
work, (b) estimate base rates for these specific behav-

iors, (c) explore underlying dimensions of these be-
haviors, and (d) begin testing a model of employee
motivation to engage in these behaviors. We devel-
oped a 41-item measure of deviant computer use and
invited communities of Internet users to self-report,
via the World Wide Web, how frequently they en-
gaged in these activities. These items were factor
analyzed, and scores were assessed according to 29
additional items that were included in the survey.

ABCD model of work computer deviance

When employees are being paid under the expec-

tation of being productive, but are not behaving in a
productive manner, they deviate from work norms.
Such norms can also be violated when employees
who are not on duty use an employer’s resources,
materials, or facilities for nonwork purposes, result-
ing in organizational inefficiency and legal liability.
Thus, organizational deviance can be defined as un-
sanctioned nonproductive use of an employer’s
time or property. Although models of deviant em-
ployee behavior exist, they fail to consider the
unique aspects of computer misuse, because with
personal computers employees may engage in these
behaviors directly from their desks, unnoticed by
coworkers just a few feet away.

To generate hypotheses that are specific to orga-

nizational deviance via work computers, we cre-
ated the ABCD model of work computer deviance.
The model examines Access to computers/Inter-
net, Breaks from Work, Organizational Climate,
and Individual Differences in the context of envi-
ronmental and personal variables.

Access to computers/Internet. While computer and

Internet access at work is prevalent, there is still vari-
ability among employees’ access both at work and
away from work. Some employees may only have
access to computers or the Internet while at work,
and even then the ease of accessing a computer likely
varies according to job type. Those users with Inter-
net access at home may be enticed by computer dis-
tractions at work because of better, faster, or more
private computer facilities available. Finally, percep-
tions of accessibility may be reduced if the employee
is aware of a computer use policy and a monitoring
system in place at work. Hypotheses follow:

1a. Participants who only have Internet access at

work will report more deviant computer use
than participants who have access outside of
work

1b. Participants who more recently gained access

to the Internet at work will report more deviant

PERSONAL USE OF WORK COMPUTERS: DISTRACTION VERSUS DESTRUCTION

731

14356c03.pgs 11/28/06 10:22 AM Page 731

computer use than participants who have had
access at work for a longer period of time

1c. Participants with faster Internet connections at

work than at home will report more deviant
computer use than participants with equal or
slower connections at work

1d. Participants with computers assigned to them

will report more deviant computer use than
will participants who share a computer with
coworkers

1e. Participants who are not aware of a computer

use policy where they work will report more
deviant computer use than will participants
who are aware of a policy

1f. Participants who are not aware of monitoring

where they work will report more deviant
computer use than will participants who are
aware of monitoring

Breaks from work.

Work breaks are based on the

assumption that employees will perform better
after resting their minds and bodies, but unautho-
rized breaks or extensions obviously take away
from time spent at work. Engaging in counterpro-
ductive breaks, also known as time theft,

12

is made

easier when employees have access to the Internet
through the same computer that is used to accom-
plish work activities. In such a situation, employees
may use their work computers as either an ongoing
coping strategy to escape stress or to accomplish
personal tasks via computer that would have other-
wise required taking more time away from work
(such as banking or shopping).

13,14

To maintain brevity in this survey, we did not di-

rectly measure stress. Instead, we asked about the
number of hours per week that participants spent
at work and the number of concurrent jobs held in
the past 6 months. These questions would roughly
measure stress in a participant’s work-life balance.
As the number of hours worked or concurrent jobs
increased, so should the desire to take a break from
work activities, either to relax or do personal tasks.
Hypotheses follow:

2a. Participants working more hours per week will

engage in more deviant computer use than will
participants working relatively fewer hours
per week.

2b. Participants holding more concurrent jobs will

engage in more deviant computer use than will
participants holding fewer concurrent jobs.

Organizational climate.

Employees’ perceptions

of their workplace are shaped through both formal

and informal socialization processes that include
orientation, training, and presentation of human
resource policies.

15

Through modeling, an em-

ployee’s deviant computer use would likely be in-
fluenced by observing coworkers’ behaviors and
the consequences received.

16

Hypotheses follow:

3a. Participants who have less organizational

tenure will report more deviant computer use
than will participants who have more tenure
(controlling for age)

3b. Participants who are not aware of coworkers

being warned for misuse will report more de-
viant computer use than will participants who
are aware of coworkers being warned

As employees perceive aspects of their work en-

vironments, they also evaluate their jobs. These
evaluations become evident in job satisfaction, job
involvement, turnover intentions, and with-
drawal behaviors (i.e., tardiness, absenteeism

17

).

Because computers provide another avenue of
psychologically removing oneself from a job, em-
ployees may engage increasingly in deviant com-
puter use as they withdraw from their current
jobs. We would expect to see more deviant com-
puter use among relatively dissatisfied employees
who are looking to leave their jobs. Hypotheses
follow:

3c. Participants who report lower levels of job sat-

isfaction will report more deviant computer
use than will participants who are aware of co-
workers being warned

Individual differences. Internet access may have

different effects on people according to age, sex, and
minority status. Younger employees, who have
grown up with computers, may see more potential
uses from Internet access than would older employ-
ees. Likewise, males and white Americans tend to
have more Internet experience than females and mi-
nority members, although this disparity has recently
shown signs of disappearing.

18

Hypotheses follow:

4a. Younger participants will report more deviant

computer use than will older participants

4b. Male participants will report more deviant

computer use than will female participants

4c. Caucasians will report more deviant computer

use than will non-Caucasians

Research on deviant employee behavior has

mainly focused on individual differences in per-

732

MASTRANGELO ET AL.

14356c03.pgs 11/28/06 10:22 AM Page 732

sonality, which is considered by many to consist of
Extroversion, Agreeableness, Conscientiousness,
Emotional Stability, and Openness to Experi-
ence.

19,20

Highly conscientious employees are usu-

ally considered dependable, careful, responsible,

21

and less likely to engage in theft, illegal activities,
and drug abuse.

22

Deviant computer use, as another

form of counterproductivity, should also be related
to measures of conscientiousness, although other
more specific personality constructs may play a role
as well. Impulsive, sensation-seeking employees
may find the temptation of computer and Internet
availability to be overwhelming. In fact, some clini-
cal psychologists suggest that people can become
addicted to Internet use.

23,24

We were unable to find

any research that addressed the potential for work
computers to be used in a compulsive manner, but
took the opportunity to include a measure of com-
pulsive computer use. Hypotheses follow:

4d. Participants who score relatively higher on a

measure of compulsive computer use will report
more deviant computer use than will those with
lower scores

METHODS

Participants

Valid data were obtained from 329 participants,

all of whom completed the survey via the World
Wide Web. This sample was formed by soliciting
participants through Internet Usenet groups and
listservs that were related to job postings, emails,
and university professors from three different insti-
tutions in the United States. Approximately 65% of
the sample were female, 74% were Caucasian, and
14% were African-American. Participants averaged
36 years of age (ranging from 19 to 66 years of age).
Although this web site was accessible to anyone on
the Internet, the study was most highly publicized
through faculty, staff, and students of a mid-At-
lantic university. Not surprisingly, 37% of partici-
pants reported that their primary employer fit the
category “educational institutions/government
/membership organization and association.” An
additional 23% were employed in the service in-
dustry and 12% were employed in health care. The
median salary range was $35,000–45,000. Relevant
to the purpose of this study, 91% reported having
work that required computer use, 81% reported
having an assigned computer at work, and 95% re-
ported having access to the Internet at work.

Measures

Work computer deviance was defined through a

set of 41 items about computer use at work during
the previous 6 months (Table 1). All items used a re-
sponse format with the following anchors: not in
the past 6 months, rarely, 1–5 times per month, 1–5
times per week, 1–5 times per day, 1–5 times per
hour, and almost constantly. (Because of participant
feedback during the study, we added an additional
response option [never did this], which we equated
with “not in the past 6 months” for this manu-
script.) Items were developed by the authors in
conjunction with six graduate students from an in-
dustrial/organizational psychology program.

In addition, the survey included five items re-

garding access and speed of the participants’ work
computers, four items about their employers’ poli-
cies for computer misuse, seven items on the nature
of their employment situations, three demographic
items, and ten items on signs of compulsive com-
puter use (which were based on previous measures
of Internet addiction

24

and the DSM-IV-TR criteria

for pathological gambling

25

). This scale is typified

by items such as “I have tried to hide from others
how much time I am actually on the computer” and
“I have routinely cut short on sleep to spend more
time on the computer” (Cronbach’s alpha was 0.88).

Procedure

Participants found the web-based survey either by

receiving an email that included a link to the site or by
someone forwarding the email. The web page ex-
plained that the study was about how people use
computers at work. Participants could remain anony-
mous unless they were a student choosing to seek
course credit in exchange for participation. After par-
ticipants clicked the “submit” button, they viewed a
page that provided a chance to email comments di-
rectly to the first author. This page also included links
to other relevant surveys and web sites as a means of
encouraging participants to learn more about work-
place issues and organizational research.

Statistical analysis

After examining the base rates of deviant com-

puter use, we used a factor analysis to examine
the underlying dimensions among these behaviors.
Ultimately we created two separate dependent
variables (factor scores for counterproductive com-
puter use and nonproductive computer use) to test
our hypotheses via correlations and partial correla-
tions. Details of these analyses follow.

PERSONAL USE OF WORK COMPUTERS: DISTRACTION VERSUS DESTRUCTION

733

14356c03.pgs 11/28/06 10:22 AM Page 733

T

ABLE

1.

R

ESPONSE

P

ERCENTAGES FOR

D

EVIANT

C

OMPUTER

U

SE

I

TEMS

R

ANKED BY

M

OST

P

REVALENT

Deviant computer use items: have you done these at work?

Response options

a

1

2

3

4

5

6

7

Used email/chat to send personal messages to person

12

9

15

23

33

3

6

Used email/chat to forward jokes/humorous material

20

20

17

26

13

1

2

Browsed website(s) WITHOUT any specific purpose (Surfing)

33

22

12

18

12

2

2

Downloaded other computer documents (NOT job related)

38

26

22

11

2

0

1

Went shopping ON-line while at work

42

30

21

4

1

0

0

Downloaded pictures WITHOUT nudity or sexual content

46

26

20

6

2

0

0

Chatted one-on-one with someone (NOT job related)

60

11

8

8

8

2

2

Played computer games against your computer while at work

61

20

7

7

3

1

0

Downloaded computer programs/applications (NOT job related)

63

23

10

4

0

0

1

Browsed Usenet groups for reasons that were not job-related

65

10

11

8

4

1

0

Used the Internet to conduct personal bank transaction

70

9

14

7

0

0

0

Used the Internet while at work to research stocks

72

14

6

4

2

0

1

Chatted (instant message, IRC, etc.) in a multi-user forum

74

8

5

6

4

1

2

(NOT job related) while at work

Downloaded music files onto computer (Napster, MP3, etc.)

78

10

7

3

2

0

0

Used the Internet while at work to visit money-making sites

82

11

5

1

0

1

0

(viewing advertisements, etc.)

Used the Internet while at work to visit sweepstakes sites that

82

12

3

1

1

1

0

award prizes (iwon.com, etc.)

Built website(s) while at work that were not job-related

83

10

5

2

1

0

0

Attempted to access sites/material at work without permission

84

10

2

0

1

0

2

Bid on auctions On-line (ebay, etc.) while at work

87

10

2

1

0

0

0

Chatted in a sexual manner with someone while at work

89

6

2

2

1

0

0

Browsed website(s) for nude or sexually explicit pictures

89

6

2

2

1

0

0

Used the Internet to telephone person (not job related)

90

6

2

2

0

1

0

Used the Internet to research odds on sporting events

91

6

2

1

1

0

0

Bought, sold, or traded stocks ON-line while at work

91

6

2

0

0

0

1

Used the Internet to research or download ways to cheat at

91

6

2

1

0

0

0

computer games

Downloaded pictures containing nudity or sexual content

92

5

1

2

0

0

0

Played computer games vs. other person(s) on the Internet

92

5

2

1

0

0

0

Created CDs while at work using downloaded music

92

5

2

1

0

0

0

Used email/chat while at work to ask a coworker out for a date,

92

4

2

1

0

0

1

romantic or sexual meeting

Forwarded someone’s email containing private comments to make

92

6

1

0

0

0

0

that person look bad

Used email/chat while at work to send sexually explicit/nude

92

4

2

1

0

0

1

pictures to coworker

Browsed Usenet groups for nude or sexually explicit pictures

94

4

0

2

0

0

0

Used the Internet to share “insider” stock information

95

4

0

0

0

0

0

Chatted while misrepresenting yourself to others

96

3

0

0

1

0

0

(gave false age, sex, etc.) while at work

Used email to make confidential work information public

96

4

0

0

0

0

0

knowledge either within or outside the workplace

Used the Internet while at work to place bets on sporting events

96

3

0

0

0

0

0

Used the Internet while at work to gamble at an online casino

97

3

0

0

0

0

0

Used the Internet to arrange theft/sabotage with other employees

97

2

0

0

0

0

0

Used the Internet while at work to buy/sell/traffic illegal drugs

97

2

0

0

0

0

0

Intentionally spread computer viruses to workplace computer(s)

98

2

0

0

0

0

0

Used email/chat while at work to threaten a coworker

98

2

0

0

0

0

0

a

Anchors for response options are as follows: (1) Never did this [or] Not in past 6 months, (2) Rarely, (3) 1–5

times per month, (4) 1–5 times per week, (5) 1–5 times per day, (6) 1–5 times per hour, (7) Almost constantly.

Percentages may not add to 100 because of rounding. N equaled 321–327 for these items.

14356c03.pgs 11/28/06 10:22 AM Page 734

RESULTS

Base rates of deviant computer use

Frequencies of participants’ responses to the 41

items measuring deviant computer use are pre-
sented in Table 1, with most prevalent behaviors
listed first. In the previous 6 months, 88% used
email at work for personal messages, and 80% for-
warded jokes. Roughly 60%–65% of this sample
browsed the web without a specific purpose or
shopped online. At the other end of the spectrum,
about 10% engaged in some form of sexual activity
through the Internet at work, and 5% or fewer used
their work computer to make confidential informa-
tion public or gamble.

Underlying dimensions of deviant computer use

To explore the underlying dimensions of deviant

computer use, we calculated a factor analysis. The
item with the lowest base rate (used email/chat
while at work to threaten a coworker) needed to be
removed from the analysis, but we were then able
to calculate the Principal Axis Factor Analysis
(KMO = 0.90) using data from all participants with
mean substitution for missing item responses. Al-
though some final communalities were quite low
(playing games against computer, building web-
sites, researching stocks, personal bank transac-
tions), we retained all items in the interest of
exploring as many behaviors as the procedure
would allow.

Analysis of initial eigenvalues suggested a two-

factor solution, which explained 45% of the
variance among the 40 items before rotation.
These two factors were rotated using the Oblimin
procedure with Kaiser Normalization (Table 2).
We labeled the first factor “Counterproductive
Computer Use” because it correlated most
strongly with behaviors that conflict with organi-
zations’ goals. These behaviors place the em-
ployer at risk legally (e.g., engaging in illegal
activity) and financially (e.g., losing proprietary
information). We labeled the second factor “Non-
productive Computer Use” because it correlated
with behaviors that are not destructive, yet not di-
rectly productive. These behaviors are mainly
means of socially connecting (e.g., emailing, chat-
ting) and accomplishing personal tasks (e.g.,
shopping and banking online). These factor ana-
lytical results combined with the frequency tabu-
lations suggest that deviant computer use is
usually a means of distraction, but for some it is a
means of destruction.

Test of hypotheses

To reduce the risk of type I error, we used the fac-

tor analysis to produce two nonorthogonal factor
scores, which became the dependent variables for
our hypotheses regarding deviant computer use.
Table 3 displays Pearson Product-Moment Correla-
tions among the factor scores and the observed
variables.

Our first set of hypotheses assumed that having

faster, more personal, and less restricted access at
work would positively correlate with deviant com-
puter use. However, results suggested a more com-
plex reality. Only two of these seven Access
hypotheses (1b and 1c) received support, and the
type of deviant computer use (counterproductive
versus nonproductive) seems to mediate the rela-
tionship with access. Compared to participants
who have had Internet access at work longer, par-
ticipants who recently gained access reported a
lower frequency of nonproductive computer use (r
= 0.16, p < 0.01), but a higher frequency of counter-
productive use (r =

0.16, p < 0.01). This pattern

suggests that the novelty of Internet access at work
is linked to higher rates of “indecent” behavior,
such as viewing pornography. However, employ-
ees who have more experience with Internet access
at work favor more acceptable (albeit nonproduc-
tive) behavior, such as emailing and file down-
loads. Indeed, having a faster Internet connection
at work was positively correlated with nonproduc-
tive computer use (r = 0.14, p < 0.01), but not coun-
terproductive computer use (r =

0.03, ns).

The other hypotheses related to computer access

were not supported, and in one instance the corre-
lation directly contradicted our hypothesis. Specifi-
cally, participants who did not have a computer
assigned to them at work (19% of our sample) were
more likely to engage in counterproductive com-
puter use than were participants with an assigned
computer (r =

0.16, p < 0.01), and this correlation

remained significant even after statistically control-
ling for being aware of a policy, hours worked, job
satisfaction, and ethnicity—all the variables that
were significantly correlated with having an as-
signed computer. We suspect that participants
without an assigned work computer differ in an
unexpected, yet meaningful way from those who
do have their own personal work access, but the
data we collected cannot account for this differ-
ence.

Our second set of hypotheses assumed that par-

ticipants with stressful work-life situations would
engage in deviant computer use at work. Although
the number of hours worked per week did not sig-

PERSONAL USE OF WORK COMPUTERS: DISTRACTION VERSUS DESTRUCTION

735

14356c03.pgs 11/28/06 10:22 AM Page 735

T

ABLE

2.

P

RINCIPAL

A

XIS

F

ACTOR

A

NALYSIS OF

D

EVIANT

C

OMPUTER

U

SE

I

TEMS

U

SING

O

BLIMIN

R

OTATION

Factor

Counterproductive

Nonproductive

Used the Internet while at work to place bets on sporting events

0.93

0.20

Intentionally spread computer viruses to infect workplace computer(s)

0.93

0.14

Used the Internet while at work to buy, sell, or otherwise traffic

0.91

0.15

illegal drugs AT WORK

Used the Internet while at work to gamble at an online casino

0.90

0.19

Used email to make confidential work information public knowledge

0.85

0.16

either within or outside the workplace

Used the Internet while at work to share “insider” stock information

0.77

0.16

Browsed Usenet groups for nude or sexually explicit pictures at work

0.73

0.29

Forwarded someone’s email containing private comments to make

0.72

0.19

that person look bad

Downloaded pictures containing nudity or sexual content while at work

0.71

0.34

Used the Internet while at work to arrange theft or sabotage with

0.69

0.25

other employees

Created CDs while at work using downloaded music

0.66

0.41

Used the Internet while at work to research odds on sporting events

0.66

0.29

Used email/chat while at work to ask a coworker out for a date,

0.65

0.35

romantic or sexual meeting

Used the Internet while at work to research or download ways to

0.64

0.25

cheat computer games

Used email/chat while at work to send sexually explicit/nude

0.63

0.30

pictures to coworker

Chatted while misrepresenting yourself to others (gave false age,

0.63

0.27

sex, etc.) while at work

Bid on auctions On-line (ebay, etc.) while at work

0.63

0.44

Browsed website(s) while at work for nude or sexually explicit pictures

0.62

0.25

Bought, sold, or traded stocks ON-line while at work

0.61

0.31

Chatted in a sexual manner with someone while at work

0.58

0.35

Used the Internet while at work to telephone person (not job related)

0.53

0.36

Played computer games vs. other person(s) on the Internet while at work

0.52

0.31

Used the Internet while at work to visit money-making sites

0.51

0.46

Used the Internet while at work to visit sweepstakes sites that award

0.44

0.35

prizes (iwon.com, etc.)

Attempted to access sites/material at work without permission

0.41

0.31

Downloaded computer programs/applications (NOT job related) at work

0.32

0.72

Browsed website(s) while at work WITHOUT any specific purpose

0.25

0.68

(Surfing)

Downloaded other computer documents (NOT job related) while at work

0.26

0.66

Used email/chat while at work to send personal messages to person

0.20

0.62

Downloaded pictures WITHOUT nudity or sexual content while at work

0.14

0.57

Chatted one-on-one with someone (NOT job related) while at work

0.24

0.56

Downloaded music files onto computer (MP3, etc.) while at work

0.39

0.56

Went shopping ON-line while at work

0.32

0.53

Used email/chat while at work to forward jokes/humorous material

0.21

0.51

to someone

Chatted (instant message, IRC, etc.) in a multi-user forum (NOT job

0.11

0.49

related) while at work

Built website(s) while at work that were not job-related

0.13

0.41

Played computer games against your computer while at work

0.24

0.39

Used the Internet while at work to conduct personal bank transaction

0.10

0.37

Used the Internet while at work to research stocks

0.36

0.36

Browsed Usenet groups for reasons that were not job-related while at work

0.29

0.32

14356c03.pgs 11/28/06 10:22 AM Page 736

PERSONAL USE OF WORK COMPUTERS: DISTRACTION VERSUS DESTRUCTION

737

T

ABLE

3.

C

ORRELA

TIONS

AMONG

O

R

THOGONAL

D

EVIANT

C

OMPUTER

U

SE

F

ACT

OR

S

CORES

AND

O

BSER

VED

V

ARIABLES

12

3

4

5

6

78

9

1

0

11

1

2

1

3

1

4

1

5

1

6

1

7

1.

Counterpr

oductive Use

2.

Nonpr

oductive Use

0.39**

3.

Only W

ork

Access

0.03

0.02

4.

T

ime Since 1st

Access

0.16**

0.16**

0.12*

5.

Faster Internet at W

ork

0.03

0.14**

0.37**

0.22**

6.

Assigned Computer

0.16**

0.06

0.07

0.30**

0.14*

7.

A

war

e of Policy

0.06

0.00

0.03

0.10

0.15**

0.14**

8.

A

war

e of Monitoring

0.09

0.02

0.04

0.01

0.08

0.02

0.48**

9.

Hours W

orked

0.03

0.01

0.04

0.19**

0.07

0.31**

0.10

0.1

1*

10.

Concurr

ent Jobs

0.17**

0.18**

0.01

0.01

0.06

0.02

0.02

0.05

0.06

11

.

O

rganization T

enur

e

0.02

0.15**

0.12*

0.27**

0.07

0.05

0.15**

0.15**

0.06

0.05

12.

Knows Other W

arned

0.1

1*

0.10

0.04

0.10

0.01

0.03

0.34**

0.47**

0.14*

0.03

0.13*

13.

Job Satisfaction

0.02

0.02

0.05

0.08

0.10

0.19**

0.04

0.14*

0.10

0.00

0.01

0.09

14.

Age

0.03

0.31**

0.07

0.23**

0.03

0.08

0.14**

0.01

0.07

0.00

0.57**

0.01

0.12*

15.

Sex

0.14*

0.24**

0.08

0.15**

0.04

0.08

0.01

0.02

0.10

0.05

0.04

0.06

0.01

0.07

16.

Ethnicity

0.14*

0.04

0.02

0.15**

0.00

0.15**

0.10

0.05

0.07

0.01

0.07

0.02

0.04

0.07

0.18**

17.

Compulsive Use

0.22**

0.35**

0.15**

0.06

0.02

0.04

0.04

0.05

0.01

0.20**

0.12*

0.09

0.00

0.07

0.31**

0.04

N for each variable

329

329

327

327

327

327

328

328

325

324

318

328

322

324

324

324

309

*Corr

elation is significant at the 0.05 level (two-tailed).

**Corr

elation is significant at the 0.01 level (two-tailed).

14356c03.pgs 11/28/06 10:22 AM Page 737

nificantly correlate with deviant computer use, the
number of concurrent jobs held in the previous six
months correlated significantly with both counter-
productive (r = 0.17, p < 0.01) and nonproductive
computer use (r = 0.18, p < 0.01). Both correlations
remained significant even after removing variance
from participants’ age, sex, ethnicity, and job satis-
faction.

The third set of hypotheses, which pertained to

learning one’s organizational climate and having a
positive evaluation of one’s job, received no support.
Although organization tenure correlated signifi-
cantly and negatively with nonproductive computer
use (r =

0.15, p < 0.01), the relationship disappeared

after controlling for age. Knowing a coworker who
had been warned for deviant computer use was sig-
nificantly associated with more counterproductive
computer use, contradicting hypothesis 3b. Further
analysis revealed that counterproductive computer
use actually increased among participants who both
had been warned and who knew a coworker who
had been warned, F(1,23.30) = 27.84, p < 0.01 (partial
eta-squared = 0.08). These results suggest that coun-
terproductive computer use is influenced more by
the association with coworkers who also engage in
deviant computer use than by observations of their
coworkers being warned.

The last set of hypotheses, pertaining to individ-

ual differences, received the most support. Younger
participants reported significantly more nonpro-
ductive computer use (r =

0.31, p < 0.01) than did

older participants, but there was no significant rela-
tionship for counterproductive computer use (r =

0.03, ns.). Male participants reported more non-
productive computer use (r = 0.14, p < 0.05) and
more counterproductive computer use Behavior (r =
0.24, p < 0.01) than did females. Non-Caucasian par-
ticipants reported more counterproductive com-
puter use than did Caucasians (r =

0.14, p < 0.05),

but ethnicity was not related to nonproductive com-
puter use (r =

0.04, ns). Also, participants who

scored higher on Compulsive Computer Use re-
ported more counterproductive computer use (r =

0.22, p < 0.01) and more nonproductive computer
use (r =

0.35, p < 0.01) than those with low com-

pulsivity. Interestingly, the sex differences in de-
viant computer use are reduced after controlling for
Compulsive Computer Use: for counterproductive
computer use, pr =

0.08, ns., and for nonproduc-

tive computer use, pr =

0.15, p < 0.05. Controlling

for Compulsive Computer use altered no other
finding for individual differences.

Although not specifically hypothesized, it is in-

teresting that employees reported higher levels of

job satisfaction when they had a work computer as-
signed to them (r =

0.19, p < 0.01) and when they

were not aware of any computer monitoring at
work (r =

0.14, p < 0.05). These correlations re-

mained significant even after removing the effects
of age, sex, ethnicity, salary, organizational tenure,
and the number of hours worked per week (pr =

0.16, p < 0.01 for assigned computer and pr =

0.12, p < 0.05 for aware of monitoring). These re-
sults suggest that employees have come to value
their (presumably) private access to the Internet
from work.

DISCUSSION

Deviant computer use at work can be separated

into two distinct, but related dimensions: nonpro-
ductive and counterproductive activities. Non-
productive computer use is far more prevalent
and consists of time spent socially connecting
(i.e., personal use of email, chat sessions) or en-
gaging in activities unrelated to work (i.e., down-
loading files, shopping online). This use is more
common among younger employees, males, and
those who use computers compulsively. To a
lesser extent, nonproductive computer use occurs
among employees who have had an Internet con-
nection at work longer, whose Internet connec-
tion is faster at work than at home, and who
concurrently work more jobs. Thus, the profile of
the nonproductive computer user is someone
who is quick to take advantage of the employer’s
broadband connection because of his or her com-
fort with using computers for entertainment and
“personal productivity.”

Although one could argue that these behaviors

ultimately do support an employee’s productivity
(as a means of rest, providing information, net-
working), some employees may be unable to know
when such behavior has become detrimental to
necessary performance. Unlike the “Internet addic-
tion” that is described by some clinical psycholo-
gists as being focused on sexual activity and
gambling,

23,24

compulsive use of computers at work

includes taking too much time away from job tasks
to socialize, play games, or attend to personal
tasks. These employees may be easily distractible,
or their chosen computer distractions may be par-
ticularly enticing. Just as a slot machine offers a
variable reinforcement schedule, a computer can
provide pleasurable outcomes that reward down-
loading large files or having virtual conversations.
Furthermore, the degree of pleasure will likely

738

MASTRANGELO ET AL.

14356c03.pgs 11/28/06 10:22 AM Page 738

vary: some downloaded pictures may not meet
with expectations, and some online interactions
may not be positive. The data from the current
study makes it difficult to pinpoint exactly what
might contribute to “unchecked” nonproductive
behavior. Likewise, the variables we included in
the current study did not measure levels of bore-
dom or anxiety, both of which may precede some
forms of employee deviance.

14

For now we conclude

that if some forms of nonproductive computer use
can indirectly contribute to employees’ productivity,
then it likely happens among those with high work
motivation.

Unlike nonproductive computer use, there is lit-

tle argument about the threat that counterproduc-
tive computer use poses to organizational success.
These are socially undesirable behaviors such as
gambling at work, downloading pornography,
asking coworkers for dates, and violating confiden-
tiality. Counterproductive computer use is more
prevalent when Internet access at work is new, and
is somewhat more common among males than fe-
males, but not more common for younger employ-
ees. Although only five to ten percent of sampled
employees reported engaging in these behaviors,
this form of deviant computer use attracts media
attention and gives employers the greatest legal
and financial reasons to take personnel action

6

. Re-

sults from these data suggest that as Internet access
becomes more commonplace at work and at home,
the rate of indecent, deviant computer use may
decline. Yet, the “Anna Kournikova virus” and
“nakedwive virus” demonstrate how easily em-
ployees (perhaps men in particular) can be tempted
to open viral email attachments that purportedly
contain images of women.

26,27

Limitations

Before discussing the implications of this study,

we should point out that we did not use propor-
tionate sampling techniques. Instead, participants
in our sample were invited via affiliation with on-
line communities, such as a listserv, a web board, or
a university email list. Our results may be specific
to the over-sampling of educated employees in the
public sector, service industry, and health care pro-
fessions. Our results may also be biased by the use
of self-reported behavior. Although this methodol-
ogy is prevalent in studies of counterproductive
work behavior, employees may censor their re-
sponses through an attempt to self-enhance. It is
unclear what affect the use of an online survey has
on these self-report biases. Furthermore, we were

not able to find a comparable set of base rates from
seven or eight years ago, when the Internet first be-
came prominent in the United States. We are there-
fore unable to know how deviant computer use is
being affected by changes in computer technology,
availability of Internet access, and employers’ ef-
forts to reduce lost time.

Implications

Employers, managers, and computer security

practitioners can take solace that only 10% or less
of employees in our sample engaged in the most
antagonistic forms of deviant computer use, and
that the most common forms of deviant computer
use involved nothing more than email and chat
sessions. More disquieting, however, is that the
likelihood that an employee will engage in de-
viant computer use does not change depending on
employee awareness of their computer use being
monitored. One explanation is that employees do
not take computer use policies seriously, perhaps
because they are enforced intermittently or have
minimal punitive impact. With only 19% of or-
ganizations providing management training to
prevent Internet misuse

2

and limited information

technology (IT) resources for monitoring em-
ployee behavior, deviant computer use may be
difficult to detect. A second explanation for em-
ployee apathy toward monitoring is that counter-
productive computer behavior may be strongly
related to personality, which cannot be easily
modified by the environment. One recent study
supports this explanation, finding that employees
with lower scores on altruism and competence are
significantly more likely to use their work com-
puters for sexual purposes.

28

A third explanation

for indifference toward computer policies is that
employees may feel entitled to use work comput-
ers for personal purposes, believing that there is
nothing inherently wrong with the behavior de-
spite a restrictive policy. Unfortunately, data from
the current study do not provide any answers for
the ineffectiveness of computer use policies.

These data do suggest that employers with poli-

cies should use a more complex definition of deviant
computer use that separates computer use for social
tasks from use that is indecent. Unacceptable behav-
ior does occur through work computers, but at the
same time employers may want to consider how to
encourage computer use that benefits organizational
productivity. Given our current finding that an as-
signed computer with Internet access is associated
with higher levels of job satisfaction, blocking all

PERSONAL USE OF WORK COMPUTERS: DISTRACTION VERSUS DESTRUCTION

739

14356c03.pgs 11/28/06 10:22 AM Page 739

Internet access may have negative consequences.
Having employees in contact with spouses, children,
and friends may increase job satisfaction and relieve
role conflict caused by work-family issues. Further-
more, it is unclear if there are hidden benefits from
having employees interacting on a personal level
with people internal and/or external to the organi-
zation. If we are living in an information age, then
employees may need to foster sources of informa-
tion even if this “cyber-networking” occasionally
leads them far from work tasks. As definitions of
work activities evolve with changing technology,
perhaps strict adherence to these work activities is
outdated, unobtainable, or even unwarranted.

REFERENCES

1. McGee, J.K. (2000). It’s official: IT adds up. Infor-

mationweek. Available at: www.Informationweek.
com/782/productivity.htm. Accessed September 13,
2006.

2. Case, C.J. & Young, K.S. (2002). Employee Internet

management: Current business practices and out-
comes. CyberPsychology & Behavior 5:355–361.

3. Hill, E.J., Miller, B.C., Weiner, S.P., et al. (1998).

Influences of the virtual office on aspects of work
and work/life balance. Personnel Psychology 51:667–
683.

4. CSI. (2001). Financial losses due to Internet intru-

sions, trade secret theft and other cyber crimes soar.
Available at: www.gocsi. com/prelea_000321.htm.
Accessed September 13, 2006.

5. Cooper, A., Delmonico, D.L., & Burg, R. (2000).

Cybersex users, abusers, and compulsives: new find-
ings and implications. Sexual Addiction and Compul-
sivity 7:5–29.

6. Bronikowski, L. (2000). Esniff.com sniffs out cyber-

slacking. ColoradoBiz 27:46.

7. CNN. (2001). Report: 60 percent of U.S. on Web in

January. Available at: www.cnn.com/2001/TECH/
internet/02/15/internet.access.reut/index.html. Ac-
cessed September 13, 2006.

8. Marsan, C.D. (2000). Are workers cyber-moonlighting?

Available at: www.cnn.com/2000/TECH/computing/
03/08moonlight.idg/index.html. Accessed September
13, 2006.

9. Verton, D. (2000). Survey: employers OK with

e-surfing. Available at: www.cnn.com/2000/TECH/
computing/12/20/e-surfing.ok.idg/index/html. Ac-
cessed September 13, 2006.

10. American Management Association. (2001). Work-

place testing: monitoring and surveillance. Available
at: www.amanet.org/research/summ.htm. Accessed
September 13, 2006.

11. Stanton, J.M., & Lin, L.F. (2001). Electronic monitor-

ing, privacy, and organizational attraction: a field
experiment. Presented at the Sixteenth Annual Con-
ference of the Society for Industrial and Organiza-
tional Psychology, San Diego.

12. Sackett, P.R., & Wanek, J.E. (1996). New develop-

ments in the use of measures of honesty, integrity,
conscientiousness, dependability, trustworthiness,
and reliability for personnel selection. Personnel Psy-
chology 49:787–829.

13. Fox, S., & Spector, P.E. (1999). A model of work frus-

tration-aggression. Journal of Organizational Behavior
20:915–931.

14. Spector, P.E., & Fox, S. (2001). An emotion-centered

model of voluntary work behavior: some parallels
between counterproductive work behavior (CWB)
and organizational citizenship behavior (OCB). Per-
sonal communication.

15. Schneider, B., & Rentsch, J. (1988). Managing cli-

mates and cultures: a futures perspective. In: Hage, J.
(ed.), Futures of organizations. Lexington, MA: Lexing-
ton Books, pp. 181–200.

16. Schneider, B., Goldstein, H.W., & Smith, D.B. (1995).

The ASA framework: an update. Personnel Psychology
48:747–774.

17. Rosse, J.G. (1988). Relations among lateness, absence,

and turnover: is there a progression of withdrawal?
Human Relations 41:517–531.

18. Diederich, T. (1998). Survey: online population reflects

general population. Available at: www.computer
world.com/cwi/story/0,1199,NAV47_STO18758,00.ht
ml. Accessed September 13, 2006.

19. Barrick, M.R., & Mount, M.K. (1991). The big five

personality dimensions and job performance: a
meta-analysis. Personnel Psychology 44:1–26.

20. Goldberg, L.R. (1990). An alternative “description of

personality”: the Big-Five factor structure. Journal of
Personality and Social Psychology 59:1216–1229.

21. Murphy, K.R. (1993). Honesty in the workplace. Pacific

Grove, CA: Brooks/Cole.

22. Ones, D.S., Viswesvaran, C., & Schmidt, F.L. (1993).

Comprehensive meta-analysis of integrity test va-
lidities: findings and implications for personnel se-
lection and theories of job performance. Journal of
Applied Psychology 78:679–703.

23. Young, K.S. (1996). Internet addiction: the emergence

of a new clinical disorder. Presented at the 104

th

An-

nual Meeting of the American Psychological Associa-
tion, Toronto, Canada.

24. Morahan-Martin J., & Schumacher, P. (2000). Inci-

dence and correlates of pathological Internet use
among college students. Computers in Human Behav-
ior 16:13–19.

25. American Psychiatric Association. (2000). Diagnostic

and statistical manual of mental disorders, (4th ed., text
revision). Washington, DC: APA.

26. Jesdanun, A. (2001). New computer virus spreads

within hours. Associated Press 12 February.

740

MASTRANGELO ET AL.

14356c03.pgs 11/28/06 10:22 AM Page 740

27. Sullivan, B. (2001). New virus tempts Kournikova

fans. MSNBC. Available at: www.msnbe.com/news/
529834.asp. Accessed September 13, 2006.

28. Mastrangelo, P.M., Everton, W.J., Jolton, J.A., et al.

(2003). Misuse of work computers: not as sexy as you
think. Presented at the 111

th

Annual Convention of

the American Psychological Association, Toronto,
Canada.

Address reprint requests to:

Dr. Paul M. Mastrangelo

Genesee Survey Services

3136 Winton Rd. S

Rochester, NY 14623

E-mail: [e-mail]

PERSONAL USE OF WORK COMPUTERS: DISTRACTION VERSUS DESTRUCTION

741

14356c03.pgs 11/28/06 10:22 AM Page 741

This article has been cited by:

1. Troy J. Strader, J. Royce Fichtner, Suzanne R. Clayton, Lou Ann Simpson. 2013. The Impact of Context on Employee

Perceptions of Acceptable Non-Work Related Computing. International Journal of Technoethics 2:2, 30-44. [

CrossRef

]

2. Damien Page. 2012. Teacher misbehaviour: an analysis of disciplinary orders by the General Teaching Council for England.

British Educational Research Journal 1-20. [

CrossRef

]

3. Murugan Anandarajan , Claire A Simmers , Rob D'Ovidio . 2011. Exploring the Underlying Structure of Personal Web Usage

in the Workplace. Cyberpsychology, Behavior, and Social Networking 14:10, 577-583. [

Abstract

] [

Full Text HTML

] [

Full

Text PDF

] [

Full Text PDF with Links

]

4. Vivien K.G. Lim, Don J.Q. Chen. 2011. Cyberloafing at the workplace: gain or drain on work?. Behaviour & Information

Technology 1-11. [

CrossRef

]

5. Jessica Vitak, Julia Crouse, Robert LaRose. 2011. Personal Internet use at work: Understanding cyberslacking. Computers

in Human Behavior 27:5, 1751-1759. [

CrossRef

]

6. Leonard Reinecke . 2009. Games at Work: The Recreational Use of Computer Games During Working Hours.

CyberPsychology & Behavior 12:4, 461-465. [

Abstract

] [

Full Text PDF

] [

Full Text PDF with Links

]

7. R. Kelly Garrett, James N. Danziger. 2008. Disaffection or expected outcomes: Understanding personal Internet use during

work. Journal of Computer-Mediated Communication 13:4, 937-958. [

CrossRef

]

8. R. Kelly Garrett , James N. Danziger . 2008. On Cyberslacking: Workplace Status and Personal Internet Use at Work.

CyberPsychology & Behavior 11:3, 287-292. [

Abstract

] [

Full Text PDF

] [

Full Text PDF with Links

]