Content uploaded by John F. Veiga
Author content
All content in this area was uploaded by John F. Veiga on Jul 04, 2014
Content may be subject to copyright.
http://jom.sagepub.com
Journal of Management
DOI: 10.1177/0149206304271768
2005; 31; 301 Journal of Management
Timothy D. Golden and John F. Veiga The Impact of Extent of Telecommuting on Job Satisfaction: Resolving Inconsistent Findings
http://jom.sagepub.com/cgi/content/abstract/31/2/301
The online version of this article can be found at:
Published by:
http://www.sagepublications.com
On behalf of:
Southern Management Association
can be found at:Journal of Management Additional services and information for
http://jom.sagepub.com/cgi/alerts Email Alerts:
http://jom.sagepub.com/subscriptions Subscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://jom.sagepub.com/cgi/content/refs/31/2/301
SAGE Journals Online and HighWire Press platforms):
(this article cites 28 articles hosted on the Citations
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
10.1177/0149206304271768ARTICLEJournal of Management / April 2005Golden, Veiga / Telecommuting
The Impact of Extent of
Telecommuting on Job Satisfaction:
Resolving Inconsistent Findings
Timothy D. Golden*
Lally School of Management and Technology, Rensselaer Polytechnic Institute,
110 8th Street, Troy, NY 12180-3590
John F. Veiga
Department of Management, University of Connecticut, BUS, 2100 Hillside Road,
Unit 1041, Storrs, CT 06269-1041
Although popular management wisdom has suggested that telecommuting enhances job satisfac-
tion, research has found both positive and negative relationships. In this study, the authors attempt
to resolve these inconsistent findings by hypothesizing a curvilinear, inverted U-shaped relation-
ship between the extent of telecommuting and job satisfaction. Using hierarchical regression
analysis on a sample of 321 professional-level employees, their findings suggest a curvilinear link
between extent of telecommuting and job satisfaction, with satisfaction appearing to plateau at
more extensive levels of telecommuting. In addition, task interdependence and job discretion mod-
erated this link, suggesting that some job attributes play an important, contingent role.
Keywords: telecommuting; job satisfaction; telework; virtual work
Despite the fact that upward of 20 million individuals in the United States are engaged in
telecommuting (Bureau of Transportation Statistics, 2003), management researchers have
been slow to investigate this increasingly popular work arrangement. Telecommuters spend
some portion of their time away from the conventional workplace, working from home, and
communicate by way of computer-based technology (Nilles, 1994). Out of the many benefits
* Corresponding author. Tel.: 518 276-2669; fax: 518 276-2665.
E-mail address: GoldenT@rpi.edu
Journal of Management, Vol. 31 No. 2, April 2005 301-318
DOI: 10.1177/0149206304271768
© 2005 Southern Management Association. All rights reserved.
301
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
claimed in the popular press, the most often cited is increased job satisfaction (Pinsonneault &
Boisvert, 2001). Indeed, the literature is filled with vivid characterizations, anecdotes, and
intriguing prescriptions for telecommuters (e.g., Apgar, 1998; Piskurich, 1998), but with few
exceptions (e.g. Cascio, 2000; Igbaria & Guimaraes, 1999), the evidence to support this claim
is equivocal.
One stream of research has found that, irrespective of the extent to which individuals
telecommute, they are more satisfied (McCloskey & Igbaria, 1998; Olson, 1989). Such find-
ings suggest that telecommuters, by virtue of their separation from the employer’s work loca-
tion, experience an increased ability to adjust work activities to meet their own needs and
desires and can more easily meet demands of both work and family responsibilities (Duxbury,
Higgins, & Mills, 1992; Riley & McCloskey, 1997). Conversely, another stream finds that
extensive levels of telecommuting may be dysfunctional and result in lower satisfaction
(Chapman, Sheehy, Heywood, Dooley, & Collins, 1995; Cooper & Kurland, 2002; Huws,
Korte, & Robinson, 1990). These findings stem from the contention that telecommuting
decreases face-to-face social interactions and strains relationships with coworkers and manag-
ers, and coupled with increased feelings of isolation, negatively affects job satisfaction.
To reconcile these inconsistent findings, we theorize a curvilinear relationship between the
extent of telecommuting and job satisfaction. Then, to more fully understand the differences
embedded in a telecommuter’s work activities, we examine the extent to which this relation-
ship is moderated by task interdependence, job discretion, and work-scheduling latitude. Our
primary focus is to make comparisons between experienced telecommuters on the basisof the
extent to which they telecommute. We focus on this because telecommuters spend varying
combinations of in-office face time and out-of-office telecommuting time (Apgar, 1998;
Feldman & Gainey, 1997), and because previous studies have primarily focused on more
course-grained comparisons between telecommuters and nontelecommuters (e.g., Crossman
& Burton, 1993; Fritz, Narasimhan, & Rhee, 1998; Igbaria & Guimaraes, 1999).
Theory and Hypotheses
Telecommuting and Job Satisfaction
Most recently, Bailey and Kurland (2002) concluded, after reviewing the telecommuting
literature, that the empirical evidence linking telecommuting to job satisfaction is unclear, and
Cooper and Kurland (2002) called for future research to examine the impact of varying levels
of telecommuting. On one hand, researchers have argued that telecommuting has a positive
impact on job satisfaction (e.g., Belanger, 1999; Dubrin, 1991; Norman, Collins, Conner,
Martin, & Rance, 1995), primarily because individuals have a greater ability to satisfy their
own needs by adjusting work tasks (Baltes, Briggs, Huff, Wright, & Neuman, 1999; Pierce &
Newstrom, 1980, 1983). Telecommuters report decreased stress (Guimaraes & Dallow, 1999)
and more control over interactions with others due to the absence of unplanned interruptions
from managers and coworkers (Dubrin, 1991). Research also suggests that telecommuters are
better able to meet nonwork, family-related responsibilities (Duxbury et al., 1992; Riley &
McClosky, 1997), so that the likelihood of work-family conflict is reduced and job satisfaction
302 Journal of Management / April 2005
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
is enhanced. From this, we would infer that the more extensively individuals telecommute, the
more they will be able to satisfy both personal and organizational needs, thereby enhancing
job satisfaction.
On the other hand, research also suggests that these positive benefits may be offset by dete-
riorating work relationships and feelings of isolation (Cooper & Kurland, 2002; Ruppel &
Harrington, 1995). This research suggests that because telecommuters are separated from oth-
ers and away from the office environment, the negative impact of increased isolation and
decreased social interactions on relationships with supervisors and coworkers is likely to neg-
atively affect job satisfaction (Pool, 1990; Yap & Tng, 1990). When employees telecommute
extensively, there is little opportunity for them to engage in informal, face-to-face interactions
in the workplace or for an honest exchange of feelings not likely done via e-mail or phone,
such as during informal mentoring or proactive job-related feedback (Cooper & Kurland,
2002). Despite the availability of media-rich technologies such as video, it is generally
believed that such devices have social, as well as technological, constraints (Rice & Gattiker,
2001; Straus & McGrath, 1994) and that “no technology can duplicate the experience of work-
ing onsite together” (Shapiro, Furst, Spreitzer, & Von Glinow, 2002: 460). Hence, absent suf-
ficient face-to-face interactions, we would infer that the more extensively an individual
telecommutes, the greater the risk of experiencing frustration and isolation, and the greater the
toll on job satisfaction.
To reconcile these inconsistent findings, we reason that the link between extent of
telecommuting and job satisfaction might be curvilinear in the shape of an inverted U. That is,
at relatively low levels of telecommuting, in which a greater proportion of time is spent in the
office, telecommuters are still able to manage important face-to-face work relationships
(Dubrin, 1991; Feldman & Gainey, 1997), to minimize any feelings of isolation, and to begin
to satisfy both individual and organizational needs that enhance job satisfaction. However, at
relatively high levels of telecommuting, a significant loss of face time and more social isola-
tion are likely to increasingly offset the individual’s ability to satisfy both individual and orga-
nizational needs, negatively affecting job satisfaction. Hence, we propose the following:
Hypothesis 1: The relationship between extent of telecommuting and job satisfaction is curvilinear in
the shape of an inverted U.
The Nature of the Telecommuter’s Job
Although the literature suggests that telecommuters may enjoy greater job flexibility than
nontelecommuters (Dubrin, 1991; Guimaraes & Dallow, 1999), whether individuals can fully
benefit from telecommuting is likely to be influenced by the way in which they must perform
their work activities. Hence, we posit that the nature of a telecommuter’s job, or job context,
plays a contingent role. Specifically, we expect that the telecommuting–job satisfaction rela-
tionship will be moderated by differences in three salient aspects of the telecommuter’s job,
including task interdependence, job discretion, and work-scheduling latitude.
Task interdependence. Task interdependence refers to the degree to which organizational
members must rely on one another to perform their tasks effectively given the nature of their
Golden, Veiga / Telecommuting 303
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
jobs (Kiggundu, 1981, 1983). Individuals whose jobs entail higher levels of task interdepen-
dence have to make repeated adjustments and modifications based on information and reac-
tions from others (Thompson, 1967). For telecommuters with high task interdependence, it is
the back-and-forth nature of their jobs and the corresponding need to communicate more fre-
quently that could be potentially more frustrating (Duxbury & Neufeld, 1999). Specifically,
compared with those with low interdependence, those with high interdependence have a
greater need to communicate with others; and the more they telecommute, the greater their
reliance on telephone and e-mail to replace face-to-face interactions with coworkers. Given
that e-mail and telephone have more limited interactivity and decreased capacity for quickly
transferring messages than face-to-face interactions (Daft & Lengel, 1986), we expect that a
greater reliance on them is likely to be more frustrating for telecommuters with highly interde-
pendent jobs, especially when transferring tacit knowledge that is more “sticky” and more dif-
ficult to transfer in the absence of face-to-face interactions (Von Hipple, 1994).
Hence, at less extensive levels of telecommuting, we expect that the increasing job satisfac-
tion experienced by telecommuters will be only somewhat dampened for those with high inter-
dependence compared with those with low interdependence. At more extensive levels of
telecommuting, though, when job satisfaction begins to decline, we expect the decline will be
at an increasingly faster rate for those with highly interdependent jobs due to the mounting
frustration associated with misunderstandings and an inadequate ability to effectively commu-
nicate complex ideas through e-mail and telephone (Rice & Gattiker, 2001; Straus & McGrath,
1994). We therefore expect the differences in job satisfaction between those with high as com-
pared with low interdependence to be most pronounced at extensive levels of telecommuting.
Stated formally,
Hypothesis 2: Task interdependence moderates the curvilinear relationship between extent of
telecommuting and job satisfaction. Specifically, for telecommuters with high interdependence,
the rise in job satisfaction associated with less extensive levels of telecommuting will be slower,
the decline in job satisfaction associated with more extensive telecommuting will be faster and
more pronounced, and the greatest difference in satisfaction will occur at extensive levels of
telecommuting.
Job discretion. Job discretion is the extent to which individuals have control over how an
assigned task is to be implemented (Langfred, 2000). Individuals with lower levels of discre-
tion have less autonomy in how they perform their jobs (Pierce, Newstrom, Dunham, & Bar-
ber, 1989) and are more likely to have to seek out information from a variety of sources (Nor-
man et al., 1995), spending more time and effort seeking clarification, direction, and approval.
Consequently, as suggested earlier, owing to the additional frustrations associated with an
increased reliance on e-mail and telephone (e.g., Daft & Lengel, 1986; Rice & Gattiker, 2001),
and greater absence of face-to-face interactions necessary for clarifying tacit concepts (Von
Hipple, 1994), telecommuters with less discretion are apt to experience rising levels of
frustration the more extensively they telecommute.
Specifically, when telecommuting less extensively, individuals with little discretion are not
as likely to experience much additional frustration, whereas at more extensive levels, mount-
304 Journal of Management / April 2005
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
ing frustration is apt to increasingly dampen associated levels of job satisfaction. Taken
together, at less extensive levels of telecommuting, we expect that the increasing job satisfac-
tion experienced by telecommuters will be somewhat less for those with low discretion com-
pared with those with high discretion. At more extensive levels of telecommuting, we expect
that the associated decline in job satisfaction will be at an increasingly faster rate for those with
low discretion because they are more likely to become increasingly hampered and frustrated
by the continual need to seek clarification and approval with less face-to-face interaction (Rice
& Gattiker, 2001; Straus & McGrath, 1994).
Hypothesis 3: Job discretion moderates the curvilinear relationship between extent of telecommuting
and job satisfaction. Specifically, for telecommuters with low discretion, the rise in job satisfac-
tion associated with less extensive levels of telecommuting will be slower, the decline in satisfac-
tion associated with more extensive telecommuting will be faster and more pronounced, and the
greatest difference in satisfaction will occur at extensive levels of telecommuting.
Work-scheduling latitude. Work-scheduling latitude refers to the ability to adjust the sched-
uling of work tasks to meet one’s needs and desires (Baltes et al., 1999). Whereas individuals
with greater scheduling latitude have more freedom in determining when their work must be
done, those with less latitude are bound by predetermined hours and are often required to be
accessible during specific hours of the day (Baltes et al., 1999). Telecommuters with greater
scheduling latitude are better able to conduct their work when it suits them (Hartman, Stoner,
& Arora, 1992) and can shift work activities to more preferable time periods, thereby mini-
mizing or precluding work’s intrusion on the family domain (Duxbury et al., 1992; Riley &
McCloskey, 1997). In addition, a significant amount of literature on employee work flexibility
suggests that to the extent individuals have a greater ability to adjust the timing of work
demands to meet their personal and family needs, the more satisfied they are with their work
lives (Baltes et al., 1999).
In contrast, those with less latitude are less able to adjust the timing of work demands and
are likely to derive comparatively less satisfaction from telecommuting because the frustration
associated with accumulating pressures to be responsive to family demands are nearly always
perceived as salient and in need of attention (Igbaria & Guimaraes, 1999). Hence, at lessexten-
sive levels of telecommuting, we expect that the increasing job satisfaction experienced by
telecommuters will be somewhat dampened for those with less latitude compared with those
with more. At more extensive levels of telecommuting, we expect that the associated decline in
job satisfaction will be at a faster rate for those with less latitude because they are increasingly
frustrated by being unable to mitigate scheduling-based pressures associated with family
demands.
Hypothesis 4: Scheduling latitude moderates the curvilinear relationship between extent of
telecommuting and job satisfaction. Specifically, for telecommuters with less latitude, the rise in
job satisfaction associated with less extensive levels of telecommuting will be slower, the decline
in satisfaction associated with more extensive telecommuting will be faster and more pronounced,
and the greatest difference in satisfaction will occur at extensive levels of telecommuting.
Golden, Veiga / Telecommuting 305
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
Method
Sample and Procedures
To control for organizational differences in company telecommuting practices, we sought
out a large firm that has a large and active telecommuting program. The high-tech firm that
agreed to participate in our study employs more than 40,000 people and actively encourages
its professional employees to telecommute. A random sample of 1,000 of the firm’s profes-
sional-level telecommuters was contacted via e-mail by a senior executive and encouraged to
participate. The firm’s human resources department, which maintained a record of
telecommuters throughout the company and calculated that 6,700 of its full-time professional
employees telecommute, constructed this sample. Following recommendations by Simsek
and Veiga (2001), we used an anonymous and confidential Web-based survey that resided on
the senior author’s server. We received usable responses from 321 respondents, representing a
32% response rate. Respondents spent on average 23% of their workweek telecommuting and
held a variety of full-time positions, including systems analysis (22%), marketing (8%), pro-
gramming (34%), engineering (17%), accounting (6%), sales (9%), and other (3%). Respon-
dents were all college graduates, 81% were male, and they averaged 36 years of age.
Measures
Job satisfaction. Job satisfaction was assessed using the three-item scale of Overall Job
Satisfaction contained in the Michigan Organizational Assessment Questionnaire (Cammann,
Fichman, Jenkins, & Klesh, 1979). This scale assesses the organizational member’s overall
affective responses to their jobs on a 7-point scale ranging from 1 (strongly disagree)to7
(strongly agree). The scale is widely used in a variety of research settings and shown to be
internally consistent with reliability ranging from .77 to .87 (Cammann et al., 1979; Jex &
Gudanowski, 1992). Following Cammann and colleague’s procedures, the three items were
averaged to yield an overall job satisfaction score (α=.85).
Extent of telecommuting. Extent of telecommuting was assessed by asking respondents to
indicate the average number of hours per week they consistently spent away from their office
working as a telecommuter. As verification of the reliability of this index, respondents were
also asked to indicate the proportion of an average workweek that they spent telecommuting.
Analysis revealed no discernible difference between these two measures (ρ= .95) or in our
results, so that for the sake of clarity, we report only hours per week.
Task interdependence. Task interdependence was assessed using five items from the mea-
sure developed by Pearce and Gregersen (1991). This measure of employee-perceived task
interdependence contains items based on the theoretical domain for reciprocal interdepen-
dence specified by Thompson (1967). This measure facilitates analysis at the individual
employee level, in contrast to earlier measures that reflect group and managerial perspectives
(e.g., Van de Ven, Delbecq, & Koenig, 1976). Similar analyses of employee-reported interde-
306 Journal of Management / April 2005
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
pendence have also been used by other researchers (Jehn, 1995), and this measure by Pearce
and Gregersen (1991) has previously been found to have acceptable internal reliability (α=
.76). Items were assessed on a 5-point scale ranging from 1 (strongly disagree)to5(strongly
agree). The item responses were averaged to create the final score (α= .83).
Job discretion. Job discretion was assessed using a four-item autonomy measure developed
by Langfred (2000), which asks respondents to assess the amount of discretion and control
they have in the implementation of assigned work tasks on a scale ranging from 1 (very little)
to 5 (very much). Despite reported reliability alphas ranging from .71 to .90, our initial analy-
sis revealed an unacceptable alpha; however, by dropping one of the items (“number of written
rules and procedures pertaining to job”), we improved the measure’s alpha to .74. Therefore,
the final score was calculated by averaging the responses for three items.
Work-scheduling latitude. Latitude was assessed using a five-item measure developed by
Pierce and Newstrom (1983). This measure asks respondents to think about their opportunity
to exercise choice in defining the hours and patterns of hours that they work. The five items are
anchored on a 5-point scale ranging from 1 (very little)to5(very much). The item responses
were averaged to create the final score (α= .91).
Control variables. In all analyses, we controlled for gender, age, functional specialization,
and telecommuting tenure. We controlled for gender because men and women may experience
differing levels of work-family conflict (Parasuraman & Greenhaus, 2002), and
telecommuting is often viewed as a means to help balance the demands of work and family life
(Bailey & Kurland, 2002). Similarly, age may play a role in the experiences of telecommuters,
thereby affecting job satisfaction (Belanger, 1999). Although participants in our sample were
occupationally similar, we nonetheless controlled for functional specialization in order to pre-
clude any possibility of unintended effects. Functional specialization was dummy coded as
follows: 1 = system analysis, 2 = marketing, 3 = programming, 4 = engineering, 5 = account-
ing, 6 = sales, 7 = other. With respect to telecommuting tenure, Ramsower (1983) reported that
job satisfaction was negatively related to tenure and that it declined in a 6-month period.
Despite the fact that Ramsower’s study only followed a small number of clerical workers (N=
16) rather than professional employees, we asked respondents to report in years how long they
had been telecommuting (responses ranged from 1 to 7 years) and controlled for tenure in
order to preclude such “honeymoon” effects.
Results
Table 1 presents means, standard deviations, correlations, and internal consistencies for
each of the measures. Five correlations were significant, and with the exception of our qua-
dratic term, correlation coefficients ranged from .00 to .45, with only one correlation coeffi-
cient above .40; the correlation between job discretion and work scheduling latitude was .45.
Although multicollinearity did not appear to be an issue, we also calculated correlations
between the quadratic telecommuting term and our three moderator variables (task interde-
Golden, Veiga / Telecommuting 307
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
308
Table 1
Descriptive Statistics and Correlations
Variable MSD123 45 67 89
1. Job satisfaction 5.96 1.12 (.85)
2. Age 36.33 7.5 .06
3. Gender 1.16 0.37 –.09 .04
4. Functional specialization 3.14 1.61 –.13* .03 .02
5. Telecommuting tenure 4.72 2.15 –.03 –.06 –.04 .00
6. Extent of telecommuting 11.72 8.74 .16** .05 –.03 –.11 .06
7. Task interdependence 4.17 0.65 .04 .02 –.02 –.06 –.02 .00 (.83)
8. Job discretion 3.34 0.96 –.08 .11 –.16** .01 –.02 –.03 –.05 (.74)
9. Work-scheduling latitude 3.85 0.89 –.03 .05 –.15** .01 .05 –.05 –.03 .45** (.91)
Note: N = 321. Numbers in parentheses are internal reliabilities, or alphas, for each measure.
*p< .05
**p< .01
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
pendence r= .02, n.s.; job discretion r= –.04, n.s.; work-scheduling latitude r= –.05, n.s.) in
order to ensure that the power of moderated regression was not undermined (Cortina, 1993).
In addition, all variables were standardized prior to conducting regression analyses.
Before proceeding to our primary analysis, and given the concerns of common method
variance, we followed the recommendations of Podsakoff, MacKenzie, Lee, and Podsakoff
(2003) for situations in which the predictor and criterion variables could not be obtained from
two different sources and the source of method bias is unspecified. In this context, as pre-
scribed by Podsakoff and Organ (1986), we conducted Harman’s single-factor test by per-
forming an exploratory factor analysis on our study’s variables and examined the unrotated
factor solution. As shown in the appendix, examination of the factor analysis revealed factors
that loaded cleanly and that paralleled ours. As can be seen, no single factor accounted for the
majority of covariance among the measures, that is, the first factor accounted for 25.9% of the
variance, whereas the other factors accounted for an additional 45.8%. Because our factors
separated cleanly and the first factor did not account for the majority of the variance, there was
no general factor. This suggests that common method variance is not likely to be a problem in
our study (Podsakoff & Organ, 1986). Together with the low to moderate correlations between
the scales, these results indicate that our model represents concepts that are not only
theoretically but also empirically distinguishable.
Hierarchical regression analysis was used to test our model because it can be used to test
curvilinear relationships directly and because it is able to test for moderating relationships. We
tested for curvilinearity by computing a squared telecommuting term and entering it in the
third step after controls and the linear terms were entered. In the regression model, a
curvilinear relationship is evident if the addition of this quadratic term results in significant
incremental variance after the linear effect has been taken into account (Cohen, Cohen, West,
& Aiken, 2003). Hypothesis 1 predicted that the relationship between the extent of
telecommuting and job satisfaction would be curvilinear in the shape of an inverted U. As
shown in Table 2 (Model 3), the squared term coefficient and the increase in model fit were
significant (β= –.31, p< .001; ∆R2= .06, p< .001). Moreover, the negative beta weight for the
quadratic term indicates that the relationship follows an inverted U shape, as predicted. As
shown in Figure 1, graphing this curvilinear relationship suggests that the curve initially
slopes upward, then turns back, and then becomes slightly negative at more extensive levels of
telecommuting. Thus, Hypothesis 1 was supported.
Testing for Moderating Effects
To assess moderation of the curvilinear relationship, we created interaction terms com-
posed of the quadratic telecommuting term and each of the moderators. Following procedures
outlined by Baron and Kenny (1986) and Aiken and West (1991), these terms were entered last
into the regression analysis after the direct effects and linear interaction terms. Moderator
terms were entered together as a block into the regression analysis in order to account for their
simultaneous effects in which perhaps one might cancel another out; therefore, our analysis
represents a more conservative and realistic approach (Kohler & Mathieu, 1993). Consistent
with the approach taken by other researchers (e.g., Xie & Johns, 1995), evidence of modera-
Golden, Veiga / Telecommuting 309
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
tion is found when there is significant incremental variance explained by the quadratic-by-lin-
ear interaction terms beyond that explained by the linear interaction terms.
Hypothesis 2 predicted that task interdependence would moderate the curvilinear relation-
ship between extent of telecommuting and job satisfaction. In Table 2 (Model 6), the signifi-
cance of the cross product term shows the interaction of task interdependence and extent of
telecommuting2on job satisfaction to be significant (β= –.15, p< .05); the overall variance
explained by the block of moderators was also significant (∆R2= .04, p< .001). Hence, the
moderation effect was supported. To facilitate interpretation of the quadratic-by-linear inter-
action effect, the individual interaction was graphed following procedures prescribed by
Cohen and colleagues (2003) and Aiken and West (1991), in which job satisfaction for those
with high and low task interdependence were graphed across +1/–1 standard deviations of
extent of telecommuting. Graphing the quadratic-by-linear interaction effect in this way dem-
onstrates how the shape or form of the curvilinear relationship varies as a function of the level
of task interdependence (Aiken & West, 1991). As shown in Figure 2, the direction of the
moderation is in the direction expected; those telecommuters with high task interdependence
experienced a somewhat slower rise in job satisfaction compared with those with low
interdependence, with the impact most pronounced at extensive levels of telecommuting.
Hypothesis 3 predicted that job discretion would moderate the curvilinear relationship
between extent of telecommuting and job satisfaction. As shown in Table 2 (Model 6), support
for job discretion as a moderator was found (β= .27, p< .001) and, as reported above, the over-
all variance explained by the block of moderators was significant (∆R2= .04, p< .001). To
310 Journal of Management / April 2005
(Mean + 1 s.d.)
(
Mean
)
(
Mean – 1 s.d.
)
·
4.8
5.2
5.6
6
6.4
11.7 20.43.0
15.1
Job
Satisfaction
Extent of Telecommuting (hours)
Figure 1
Extent of Telecommuting and Job Satisfaction
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
311
Table 2
Hierarchical Regression Analysis—Dependent Variable: Job Satisfaction
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
Step 1: Control variables
Telecommuting tenure –.01 –.02 –.03 –.03 –.02 –.01
Age .09 .07 .05 .06 .07 .03
Gender –.08 –.08 –.10 –.12* –.12* –.10
Functional specialization .15* .12 .11 .10 .11 .12
Step 2
Extent of telecommuting .13* .33*** .33*** .33*** .31***
Step 3 (Hypothesis 1)
Extent of Telecommuting2–.31*** –.31*** –.31*** –.37***
Step 4
Task interdependence .02 .06 –.04
Job discretion –.12 –.05 .17
Work-scheduling latitude .04 .01 –.03
Step 5
Telecommuting ×Task Interdependence .09 –.01
Telecommuting ×Job Discretion .16* .30***
Telecommuting ×Work-Scheduling Latitude –.08 –.12
Step 6 (Hypotheses 2, 3, 4)
Telecommuting2×Task Interdependence –.15*
Telecommuting2×Job Discretion .27***
Telecommuting2×Work-Scheduling Latitude .02
Change in R2.04* .02* .06*** .01 .02 .04***
R2.04 .05 .11 .12 .14 .17
Adjusted R2.02 .04 .09 .09 .10 .13
F2.81* 3.33* 5.96*** 4.36*** 3.82*** 4.00***
N= 321
*p< .05
***p< .001
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
facilitate interpretation of the quadratic-by-linear interaction effect, we again graphed the
individual interaction for those with high and low discretion. As shown in Figure 3, the direc-
tion of the moderation is as expected; those telecommuters with high discretion experienced
enhanced levels of job satisfaction compared with those with low discretion. However,
although we had expected the impact to be most pronounced at extensive levels of
telecommuting, it appears that the influence is most pronounced at fairly low levels of
telecommuting.
Hypothesis 4 predicted that work-scheduling latitude would moderate the relationship
between the extent of telecommuting and job satisfaction. However, as shown in Table 2
(Model 6), the interaction term was not statistically significant (β= .02, n.s.), and this hypothe-
sis was not supported.
Discussion
In this study, we resolve previously inconsistent findings by proposing and finding a
curvilinear inverted U-shaped relationship between the extent of telecommuting and job satis-
faction. Our findings cast doubt on conventional wisdom, namely, that the more employees
telecommute, the more satisfied they are irrespective of the nature of their jobs (Pinsonneault
& Boisvert, 2001). As shown in Figure 1, job satisfaction initially increases as the extent of
telecommuting rises; however, at higher levels of telecommuting it starts to level off and
312 Journal of Management / April 2005
(
Mean
)
(
Mean - 1 s.d.
)
(
Mean + 1 s.d.
)
4.8
5.2
5.6
6
6.4
6.8
3.0 11.7 20.4
Low Interde
p
endence
(Mean + 1 s.d.)
Job
Satisfaction
High Interdependence
Extent of Telecommuting (hours)
Figure 2
Task Interdependence Moderation Effect
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
decrease slightly, and essentially plateaus. Although this suggests that the impact of
telecommuting on job satisfaction is more complex than previously thought, it also suggests
that gains in job satisfaction at lower levels of telecommuting are relatively substantial
compared with those at higher levels.
In finding this curvilinear relationship, our results suggest that there may be a crucial
threshold in the amount of time an individual can telecommute beyond which the benefits of
additional gains in job satisfaction are not accrued. (Although not necessarily indicative of the
general population of telecommuters, in our sample, the threshold at which job satisfaction
plateaus is 15.1 hours per week). Even though the flexibility and freedom afforded by
telecommuting may initially increase satisfaction up to a point (Dubrin, 1991; Feldman &
Gainey, 1997), our results suggest its use must be curtailed if employees are to optimize their
work outcomes. This implies that merely replacing face-to-face interactions with those carried
out via technology may not be entirely sufficient, even for technology-based professionals
such as those we studied. The limited interactivity of the electronic media (Daft & Lengel,
1984, 1986) coupled with increased social isolation (Ruppel & Harrington, 1995) may take a
toll on the benefits of telecommuting. Although this cross-sectional interpretation warrants
further research, it does provide a plausible resolution to the contradictory claims voiced by
prior researchers and practitioners.
Our findings also suggest that the job satisfaction of telecommuters needs to be understood
not only in terms of the direct impact of extent of telecommuting but also in terms of the condi-
tional effects resulting from differences in work activities embedded in the telecommuter’s
job. When our three job-related moderators were analyzed simultaneously, both task interde-
Golden, Veiga / Telecommuting 313
(
Mean
)
(
Mean – 1 s.d.
)
4.8
5.2
5.6
6
6.4
6.8
3.0 11.7
Hi
g
h Discretion
Low Discretion
20.4
(Mean + 1 s.d.)
Job
Satisfaction
Extent of Telecommuting (hours)
Figure 3
Job Discretion Moderation Effect
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
pendence and job discretion moderated the curvilinear relationship between extent of
telecommuting and job satisfaction as hypothesized. As suggested by Figures 2 and 3, one
implication of this finding is that telecommuters whose jobs entail low levels of task interde-
pendence and/or high levels of job discretion tend to experience comparatively greater levels
of job satisfaction across all levels of telecommuting. Perhaps individuals with little discretion
who telecommute extensively have found ways to communicate more effectively and hence
are better able to cope with the frustrations of seeking clarification, approval, and guidance,
whereas those at less extensive levels have not. Perhaps this is simply a case in which those
with higher discretion experience a greater enhancement to their job satisfaction at less exten-
sive levels of telecommuting. Although either of these explanations may be plausible, we sug-
gest future researchers conduct additional tests of these and other job context factors to more
fully understand the dynamics of these relationships.
Contrary to expectations, work-scheduling latitude did not play a moderating role. Given
the conservative nature of our test in which moderators were entered as a block, the other mod-
erators may have played a more dominant part. On the other hand, this finding may be an arti-
fact of our sampling frame, in that scheduling latitude may be an inherent job condition that is
not reflected in job satisfaction for salaried professionals (Baltes et al., 1999). In other words,
the ability to adjust the scheduling of work tasks is an integral aspect of professional work,
particularly for those who choose to telecommute, so that it is unlikely to influence the level of
job satisfaction even at more extensive levels of telecommuting. It is also possible, given the
moderate correlation between work-scheduling latitude and job discretion (r= .45), that our
measure may not have adequately captured the scheduling latitude inherent in the profes-
sional-level telecommuters in our sample. Future research using different measures of these
variables could help shed insight into this possibility.
As this research stream progresses, we need to know more about how the extentof
telecommuting affects employee productivity. Although we recognize that job satisfaction
and performance are not highly correlated (Fisher, 2003), we would still expect to find
increased performance at relatively moderate levels of telecommuting tapering off signifi-
cantly at extensive levels paralleling our job satisfaction findings. Although our sample was
too diverse in the types of work they performed, making actual performance difficult to mea-
sure and compare, future research might want to examine, for example, a large sales organiza-
tion where performance comparisons are more easily made.
Although this research offers new insights, because the study was correlational and did not
involve the manipulation of variables, causality cannot be inferred. We cannot speak to
whether telecommuting improves job satisfaction per se as is often claimed (Pinsonneault &
Boisvert, 2001), only that the extent of telecommuting is related to the level of job satisfaction
among those who telecommute. In addition, the present research design cannot rule out the
possibility of reverse/reciprocal causality between extent of telecommuting and job satisfac-
tion. Specifically, low job satisfaction might cause extent of telecommuting; individuals might
telecommute more extensively in order to improve their job satisfaction. The telecommuting
literature, though, suggests that managers often fear giving up control over subordinates and
hence are reluctant to support telecommuting, making it hard to imagine that managers would
encourage unsatisfied subordinates to telecommute and run the risk of less control (Bailey &
314 Journal of Management / April 2005
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
Kurland, 2002). Nevertheless, longitudinal designs in which the complex interplay of these
variables is examined over time, or the use of experimental or quasi-experimental
manipulations, should prove particularly insightful.
Finally, in this study, we assessed ongoing job satisfaction among telecommuters who had
been telecommuting, on average, for more than 4 years. However, the quality of pre-
telecommuting satisfaction, as well as other antecedents, may well play an influential role.
Moreover, given that participants in this study were all high-technology-based professionals
from one firm, the generalizability of our findings may also be limited. Future research with
data from multiple organizations would permit an examination of whether firm differences in
contextual factors, such as organizational support, also play a role. By investigating tele-
commuting as a function of theses factors, we can further unravel the many complexities
inherent in this important and emerging form of work.
APPENDIX
Exploratory Factor Analysis of Survey Measures
Component
Measure 1 2 3 4 5
Telecommuting –.027 –.011 .168 –.007 .881
Telecommuting2–.027 .029 –.143 –.017 .898
Job Satisfaction1 –.022 .030 .912 –.026 –.062
Job Satisfaction2 –.020 .080 .868 –.069 –.018
Job Satisfaction3 –.024 –.052 .857 .014 .105
Task Interdependence1 –.033 .779 –.053 .061 .042
Task Interdependence2 .020 .850 .048 .096 –.003
Task Interdependence3 –.023 .763 .013 –.113 .037
Task Interdependence4 .063 .659 –.001 –.044 –.081
Task Interdependence5 –.060 .814 .069 –.071 .039
Job Discretion1 .243 –.099 –.062 .771 –.068
Job Discretion2 .264 .031 .012 .812 –.004
Job Discretion3 .374 –.007 –.042 .663 .052
Work-Scheduling Latitude1 .822 .031 –.150 .235 –.051
Work-Scheduling Latitude2 .854 .059 .011 .192 –.032
Work-Scheduling Latitude3 .782 –.182 .073 .197 –.026
Work-Scheduling Latitude4 .880 .032 –.043 .135 .030
Work-Scheduling Latitude5 .877 –.004 –.057 .206 –.010
Variance explained 25.9 17.1 13.2 9.0 6.2
Eigenvalue 4.6 3.1 2.4 1.6 1.3
Note: Numbers in italics indicate the highest factor scores for each item.
Golden, Veiga / Telecommuting 315
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
References
Aiken, L. S., & West, S. G. 1991. Multiple regression: Testing and interpreting interactions.Newbury Park, CA: Sage.
Apgar, M., IV. 1998. The alternative workplace: Changing where and how people work. Harvard Business Review,
76(3): 121-136.
Bailey, D. E., & Kurland, N. B. 2002. A review of telework research: Findings, new directions, and lessons for the
study of modern work. Journal of Organizational Behavior, 23: 383-400.
Baltes, B. B., Briggs, T. E., Huff, J. W., Wright, J. A., & Neuman, G. A. 1999. Flexible and compressed workweek
schedules: A meta-analysis of their effects on work-related criteria. Journal of Applied Psychology, 84(4): 496-
513.
Baron, R. M., & Kenny, D. A. 1986. The moderator-mediator variable distinction in social psychological research:
Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6): 1173-
1182.
Belanger, F. 1999. Workers’propensity to telecommute: An empirical study. Information and Management, 35: 139-
153.
Bureau of Transportation Statistics. 2003. Omnibus household survey results (August). Washington, DC: Government
Printing Office.
Cammann, C., Fichman, M., Jenkins, D., & Klesh, J. 1979. The Michigan Organizational Assessment Questionnaire.
Unpublished manuscript, University of Michigan, Ann Arbor.
Cascio, W. F. 2000. Managing a virtual workplace. Academy of Management Executive, 14(3): 81-90.
Chapman, A. J., Sheehy, N. P., Heywood, S., Dooley, B., & Collins, S. C. 1995. The organizational implications of
teleworking. In C. L. Cooper & I. T. Robertson (Eds.), International review of industrial and organizational psy-
chology, Vol. 10: 229-248. London: Wiley.
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. 2003. Applied multiple regression/correlation analysis for the behav-
ioral sciences (3rd ed.). Mahwah, NJ: Lawrence Erlbaum.
Cooper, C., & Kurland, N. B. 2002. Telecommuting, professional isolation and employee development in public and
private organizations. Journal of Organizational Behavior, 23: 511-532.
Cortina, J. 1993. Interaction, nonlinearity, and multicollinearity: Implications for multiple regression. Journal of Man-
agement, 18: 915-922.
Crossman, G., & Burton, P. F. 1993. Telework stereotypes: A case study. Journal of Information Science, 19: 349-362.
Daft, R. L., & Lengel, R. H. 1984. Information richness: A new approach to managerial behavior and organizational
design. Research in Organizational Behavior, 6: 191-233.
Daft, R. L., & Lengel, R. H. 1986. Organizational information requirements, media richness and structural design.
Management Science, 32: 554-571.
Dubrin, A. J. 1991. Comparison of the job satisfaction and productivity of telecommuters versus in-house employees:
A research note on work in progress. Psychological Reports, 68: 1223-1234.
Duxbury, L. E., Higgins, C. A., & Mills, S. 1992. After hours telecommuting and work-family conflict: A comparative
analysis. Information Systems Research, 3: 173-190.
Duxbury, L., & Neufeld, D. 1999. An empirical evaluation of the impacts of telecommuting on intra-organizational
communication. Journal of Engineering and Technology Management, 16: 1-28.
Feldman, D. C., & Gainey, T. W. 1997. Patterns of telecommuting and their consequences: Framing the research
agenda. Human Resource Management Review, 7(4): 369-388.
Fisher, C. 2003. Why do lay people believe that satisfaction and performance are correlated? Possible sources of a
commonsense theory. Journal of Organizational Behavior, 24(6): 753-777.
Fritz, M. B., Narasimhan, S., & Rhee, H. 1998. Communication and coordination in the virtual office. Journal of Man-
agement Information Systems, 14(4): 7-28.
Guimaraes, T., & Dallow, P. 1999. Empirically testing the benefits, problems, and success factors for telecommuting
programmes. European Journal of Information Systems, 8: 40-54.
Hartman, R. I., Stoner, C. R., & Arora, R. 1992.Developing successful organizational telecommuting arrangements:
Worker perceptions and managerial prescriptions. S.A.M. Advanced Management Journal, 57(3): 35-42.
Huws, U., Korte, W. B., & Robinson, S. 1990. Telework: Towards the elusive office. Chichester, UK: Wiley.
316 Journal of Management / April 2005
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
Igbaria, M., & Guimaraes, T.1999. Exploring differences in employee turnover intentions and its determinants among
telecommuters and non-telecommuters. Journal of Management Information Systems, 6(1): 147-164.
Jehn, K. A. 1995. A multimethod examination of the benefits and detriments of intragroup conflict. Administrative
Science Quarterly, 40: 256-282.
Jex, S. M., & Gudanowski, D. M. 1992. Efficacy beliefs and work stress: An exploratory study. Journal of Organiza-
tional Behavior, 13(5): 509-517.
Kiggundu, M. N. 1981. Task interdependence and the theory of job design. Academy of Management Review, 6: 499-
508.
Kiggundu, M. N. 1983. Task interdependence and job design: Test of a theory. Organizational Behavior and Human
Performance, 31: 145-172.
Kohler,S. S., & Mathieu, J. E. 1993. Individual characteristics, work perceptions, and affective reactions influences on
differentiated absence criteria. Journal of Organizational Behavior, 14: 515-530.
Langfred, C. W. 2000. The paradox of self-management: Individual and group autonomy in work groups. Journal of
Organizational Behavior, 21: 563-585.
McCloskey, D. W., & Igbaria, M. 1998. A review of the empirical research on telecommuting and directions for future
research. In M. Igbaria & M. Tan (Eds.), The virtual workplace: 338-358. Hershey, PA: Idea Group.
Nilles, J. M. 1994. Making telecommuting happen: A guide for telemanagers and telecommuters. New York: Van
Nostrand Reinhold.
Norman, P., Collins, S., Conner, M., Martin, R., & Rance, J. 1995. Attributions, cognitions, and copying styles:
Teleworkers’reactions to work-related problems. Journal of Applied Social Psychology, 25(2): 117-128.
Olson, M. H. 1989. Organizational barriers to professionaltelework. In E. Boris & C. R. Daniels (Eds.), Historical and
contemporary perspectives on paid labor at home: 217-237. Urbana: University of Illinois Press.
Parasuraman, S., & Greenhaus, J. H. 2002. Toward reducing some critical gaps in work-family research. Human
Resource Management Review, 12: 299-312.
Pearce, J. L., & Gregersen, H. B. 1991. Task interdependence and extra role behavior: A test of the mediating effects of
felt responsibility. Journal of Applied Psychology, 76(6): 838-844.
Pierce, J. L., & Newstrom, J. W. 1980. Toward a conceptual clarification of employee responses to flexible working
hours: A work adjustment approach. Journal of Management, 6: 117-134.
Pierce, J. L., & Newstrom, J. W. 1983. The design of flexible work schedules and employee responses: Relationships
and process. Journal of Organizational Behavior, 4: 247-262.
Pierce, J. L., Newstrom, J. W., Dunham, R. B., & Barber, A. E. 1989. Alternative work schedules. Boston: Allyn &
Bacon.
Pinsonneault, A., & Boisvert, M. 2001. The impacts of telecommuting on organizations and individuals: A review of
the literature. In N. J. Johnson (Ed.), Telecommuting and virtual offices: Issues and opportunities: 163-185. Her-
shey, PA: Idea Group.
Piskurich, G. M. 1998. An organizational guide to telecommuting: Setting up and running a successful telecommuter
program: 1-175. Alexandria, VA: American Society for Training and Development.
Podsakoff, P. M., MacKenzie, S. B., Lee, J., & Podsakoff, N. P. 2003. Common method bias in behavioral research: A
critical review of the literature and recommended remedies. Journal of Applied Psychology, 88: 879-903.
Podsakoff, P. M., & Organ, D. W. 1986. Self-reports in organizational research: Problems and prospects. Journal of
Management, 12: 531-544.
Pool, I. 1990. Technologies without boundaries: On telecommunications in a global age. Cambridge, MA: Harvard
University Press.
Ramsower, R. M. 1983. Telecommuting: The organizational and behavioral effects of working at home. Ann Arbor,
MI: University Microfilms International.
Rice, R. E., & Gattiker, U. E. 2001. New media and organizational structure. In F. Jablin & L. Putnam (Eds.), Hand-
book of organizationalcommunication: Advances in theory, researchand methods: 544-581. Thousand Oaks, CA:
Sage.
Riley, F., & McCloskey, D. W. 1997. Telecommuting as a response to helping people balance work and family. In S.
Parasuraman & J. H. Greenhaus (Eds.), Integrating work and family: Challenges and choices for a changing
world: 133-142. Westport, CT: Quorum.
Golden, Veiga / Telecommuting 317
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
Ruppel, C. P., & Harrington, S. J. 1995. Telework: An innovation where nobody is getting on the bandwagon? The
DATABASE for Advances in Information Systems, 26(2-3): 87-104.
Shapiro, D. L., Furst, S. A., Spreitzer, G. M., & Von Glinow, M. A. 2002. Transnational teams in the electronic age:
Are team identity and high performance at risk? Journal of Organizational Behavior, 23: 455-467.
Simsek, Z., & Veiga, J. F. 2001. A primer on Internet organizational surveys. OrganizationalResearch Methods, 4(3):
218-235.
Straus, S. G., & McGrath, J. E. 1994. Does the medium matter? The interaction of task type and technology on group
performance and member reactions. Journal of Applied Psychology, 79: 87-97.
Thompson, J. D. 1967. Organizations in action. New York: McGraw-Hill.
Van de Ven, A. H., Delbecq, A. L., & Koenig, R. 1976. Determinants of coordination modes within organizations.
American Sociological Review, 41: 322-338.
Von Hipple, E. 1994. Sticky information and the locus of problem solving: Implications for Innovation. Management
Science, 40(4): 429-439.
Xie, J. L., & Johns, G. 1995. Job scope and stress: Can job scope be too high? Academy of Management Journal, 38(5):
1288-1309.
Yap, C. S., & Tng, H. 1990. Factors associated with attitudes towards telecommuting. Information and Management,
19: 227-235.
Biographical Notes
Timothy D. Golden is an assistant professor of management in the Lally School of Management and Technology at
Rensselaer Polytechnic Institute. He earned his Ph.D. in organizational behaviorfrom the University of Connecticut.
His current research interests focus on telecommuting and other virtual forms of interaction, work-family conflict,
cross-cultural behavior, and dispersed leadership.
John F. Veiga earned his D.B.A. from Kent State University; he holds the Northeast Utilities Chair in Business Ethics,
is a Board of Trustees Distinguished Professor, and is head of the management department at the University of Con-
necticut. His current research interests include top management teams, telecommuting and virtual work, entrepreneur-
ship, managerial ethics, and cross-cultural behavior.
318 Journal of Management / April 2005
© 2005 Southern Management Association. All rights reserved. Not for commercial use or unauthorized distribution.
at PENNSYLVANIA STATE UNIV on February 6, 2008 http://jom.sagepub.comDownloaded from
