Computers in Human Behavior 36 (2014) 510–519

Contents lists available at ScienceDirect

Computers in Human Behavior

journal homepage: www.elsevier.com/locate/comphumbeh

Explaining cyberloafing: The role of the theory of planned behavior

Kevin Askew a,⇑, John E. Buckner b,2, Meng U. Taing a,1, Alex Ilie c,3, Jeremy A. Bauer a,1,
Michael D. Coovert a,1
a
Department of Psychology, Montclair State University, 1 Normal Avenue, Dickson Hall, Montclair, New Jersey 07043, United States
b
Department of Psychology, Louisiana Tech University, Woodard Hall, Office 122, Ruston, LA 71272, United States
c
Department of Psychology, Illinois State University, Campus Box 4620, Normal, IL 61790, United States

a r t i c l e i n f o a b s t r a c t

Article history: The Internet enables employees to be more productive than ever before, but it also allows employees a
Available online 9 May 2014 new way to escape from work—cyberloafing. In our investigation, we test the validity of the Theory of
Planned Behavior as a model of cyberloafing. In Study 1, the goal is to provide an initial test of the theory.
Keywords: In Study 2, we cross-validate the results from Study 1 in a sample that approaches representing the gen-
Cyberloafing eral working population. Results unanimously support the main TPB model, the model accounting for 32%
Cyberslacking and 37% of the variance in cyberloafing in Studies 1 and 2, respectively. The discussion addresses both the
Personal web usage
theoretical impact and practical implications of our work.
Theory of planned behavior
Cyberdeviancy
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction potential intervention point for increasing productivity

(Naughton, Raymond, & Shulman, 1999). The focus of cyberloafing
Cyberloafing is a set of behaviors at work in which an employee research should not be on trying to eliminate cyberloafing. Rather,
engages in electronically-mediated activities, particularly through the focus should be on understanding cyberloafing so that organi-
the use of the internet, that his or her immediate supervisor would zations can strike a balance between productivity and the needs
not consider job-related (Askew, Coovert, Vandello, Taing, & Bauer, and concerns of employees (de Lara, Tacoronte, & Ding, 2006).
2011; Lim, 2002). Examples of cyberloafing include watching You- Being overly strict concerning internet usage at work could nega-
Tube and checking Facebook (Lim, 2002). Many cyberloafing tively impact employee satisfaction and perceived fairness, as well
behaviors, such as web-browsing, are familiar to most people as hurt the retention of talented employees, whereas being overly
(Lim & Teo, 2005). Other behaviors, such as playing video games lenient with regard to internet usage could have negative effects
at work over the internet, are rare, but are nonetheless cyberloa- on productivity (Case & Young, 2002).
fing (Lim & Teo, 2005). Fundamentally, cyberloafing is about wast- Researchers have proposed a couple different explanations of
ing time at work through a computer – often times under the guise cyberloafing. Lim and her colleagues have suggested perceived jus-
of doing actual work (Blanchard & Henle, 2008). It is this quality of tice as a major cause (Lim, 2002; Lim & Teo, 2006). People cyberloaf
cyberloafing – the ability to look like one is doing work while actu- when they perceive that the company or its members has treated
ally slacking off – that makes cyberloafing especially problematic them unfairly. Cyberloafing is a way to get even, or restore justice,
for organizations (Wagner, Barnes, Lim, & Ferris, 2012). Certain in this perspective (Lim, 2002). Mastrangelo and his colleagues
employees can and do spend entire days cyberloafing (Wallace, have proposed the ABCD model of cyberloafing, which posits that
2004; Wallace, 2004). cyberloafing can be understood by looking at the confluence of
Cyberloafing is important to study because it is the most Access to computers, Breaks from work, Climate, and individual
common way that people waste time at work, and therefore is a Differences (Mastrangelo, Everton, & Jolton, 2006).
Perhaps the theory that has garnered the most support from
⇑ Corresponding author. Address: 2100 Paper Mill Rd., Phoenix, MD 21131, cyberloafing community is the ego depletion model of self-regula-
United States. Tel.: +1 (813) 974 2492. tion (Baumeister, Muraven, & Tice, 2000; Wagner et al., 2012). In
E-mail addresses: [email protected] (K. Askew), [email protected] (J.E. its general form, the ego depletion model of self-regulation posits
Buckner), [email protected] (M.U. Taing), [email protected] (A. Ilie), jbauer3@mail. that self-control is like a muscle: it fatigues with use but will
usf.edu (J.A. Bauer), [email protected] (M. D. Coovert).
1
recover with rest. Applied to cyberloafing, the theory states that
+1 (813) 974 2492.
2
+1 (651) 491 4057.
when an employee’s self-control resources are depleted, he or
3
+1 (309) 438 8068. she engages in cyberloafing to recoup self-control resources

http://dx.doi.org/10.1016/j.chb.2014.04.006
0747-5632/Ó 2014 Elsevier Ltd. All rights reserved.
 K. Askew et al. / Computers in Human Behavior 36 (2014) 510–519 511

(Wagner et al., 2012). The theory has been tested by three research amount of time an employee spends working to less than what is
teams, all three of which have found general support for the theory expected by the organization (Spector et al., 2006) – and with-
(Prasad, Lim, & Chen, 2010; Restubog et al., 2011; Wagner et al., drawal behaviors (e.g., lateness, absenteeism, extended breaks)
2012). Also consistent with theory of ego-depletion is the fact that have been accurately modeled by the TPB (Brouwer et al., 2009;
self-regulation personality variables such as impulsivity (Everton, Henle et al., 2010). Likewise, technology-related behaviors such
Mastrangelo, & Jolton, 2005), self-control (Restubog et al., 2011), as instant messaging use (Lu, Zhou, & Wang, 2009), technology
and conscientiousness (Jia, 2008) have been implicated in adoption (Mathieson, 1991), and use of social networking sites
cyberloafing. (Pelling & White, 2009) have all been successfully modeled with
Even though ego depletion theory has strong empirical support, the TPB. Given that behaviors similar to cyberloafing have been
the theory is unlikely to fully account for why people cyberloaf. explained by the TPB, we expect that cyberloafing can be under-
First, the theory fails to account for non-resource draining environ- stood within a TPB framework as well.
mental influences, such as social norms and the physical work In additional to the theoretical evidence, empirical evidence for
environment, which have already been established as predictors the TPB as a valid model of cyberloafing also exists. Attitudes have
of cyberloafing (Blanchard & Henle, 2008; Liberman, Seidman, been found to correlate with cyberloafing (Liberman et al., 2011), a
McKenna, & Buffardi, 2011; Ozler & Polat, 2012). Second, the the- construct close to perceived behavioral control (i.e., the ability to
ory of ego depletion cannot account for why people still cyberloaf hide cyberloafing) has been found to relate to cyberloafing
even when they are fully rested. Thus, to fully explain cyberloafing (Askew, Coovert, Taing, Ilie, & Bauer, 2012), and social norms are
an alternative theory is needed that can address the limitations of arguably the best known predictor of cyberloafing (Askew,
the ego depletion model of cyberloafing. Vandello, & Coovert, 2010; Blanchard & Henle, 2008; Restubog
In the present investigation, we sought to identify and test an et al., 2011). Thus, there is empirical evidence in the extant litera-
alternative theory of cyberloafing. We examined the definition of ture that converges with the theoretical evidence for the TPB as a
cyberloafing and the literatures on related constructs, and con- valid model of cyberloafing.
cluded that the Theory of Planned Behavior (TPB) could be a valid
theory of cyberloafing (Ajzen, 1985). The TPB, as applied to 2.2. Theoretical considerations
cyberloafing, posits that cyberloafing is caused by three distal
antecedents – subjective social norms, cyberloafing attitudes, and The TPB posits that one of the antecedents of behavior is per-
perceived behavioral control – which are mediated through ceived behavioral control, a variable conceptually close to the con-
intentions to engage in cyberloafing. The main line of reasoning struct of self-efficacy (Ajzen, 2011). In regards to cyberloafing, one
for identifying the TPB as a potential model was: (a) cyberloafing way to conceptualize perceived behavioral control is the person’s
meets the definition of a withdrawal behavior – that is, cyberloa- self-efficacy to navigate to their favorite websites at work. In theory,
fing reduces the amount of time an employee spends working to this skill depends on three factors: the ability to navigate to desired
less than is what is expected by the organization; (b) withdrawal website by typing in the URL or navigating to the website via a search
behaviors have been modeled successfully with the TPB, and (c) engine (i.e., Google), the presence or lack of website-blocking tech-
therefore, the TPB is likely to be an useful theory in understanding nologies at work, and the ability to circumvent blocking technolo-
cyberloafing. gies if they exist by using a proxy server or some other means. We
We tested the TPB as a model of cyberloafing in two studies. In call this overall skill, which is determined non-linearly by the three
Study 1, the goal is to provide an initial test of the theory. In Study factors above, ‘‘website access self-efficacy’’. Website access self-
2, the goal is to cross-validate our results from Study 1 in a sample efficacy has not been examined by cyberloafing researchers to date
that approaches representing the general working population. In but a related construct, company monitoring, has, and has been
the section below, we expound upon the evidence for the TPB as found to have only a modest relationship with cyberloafing
a model of cyberloafing, and explore possible variations of the (Henle, Kohut, & Booth, 2009; Mastrangelo et al., 2006).
model. The discussion of variations of the model culminates in Another way to conceptualize perceived behavioral control is an
our hypotheses, which are tested in the two studies. We close with employee’s self-efficacy to engage in cyberloafing behavior with-
a discussion of the implications for both research and practice, as out ‘‘getting caught’’. The perceived ability to hide cyberloafing
well as noting future directions. refers to how well an employee can hide his or her computer activ-
ity from their coworkers and supervisors (Askew et al., 2011).
Employees who are high on the ability to hide cyberloafing might
2. Hypothesis development have some or all of the following conditions: (a) their computer
screens are not easily visible to coworkers or supervisors, (b) they
2.1. The theory of planned behavior can hear or see people approaching their work station, (c) they
work in isolation or (d) their computer activity is not monitored.
The Theory of Planned Behavior posits that behavior is caused by In contrast to website access self-efficacy, the ability to hide cyber-
three main antecedents: subjective social norms, attitudes, and per- loafing has been established as a predictor of cyberloafing (Askew
ceived behavioral control (Ajzen, 1985). The theory also posits that et al., 2011; Askew et al., 2012). As such, we consider the ability to
the influence of these three antecedents is mediated by the forma- hide cyberloafing to be the stronger candidate for the perceived
tion of intentions to engage in the behavior (Ajzen, 1985). Thus, behavioral control variable.
applied to cyberloafing, the theory posits that perceptions of refer- A second conceptual issue concerning the TPB that needs to be
ent others’ cyberloafing behaviors, attitudes towards personal com- addressed is subjective norms. There are two broad types of norms:
puter use at work, and perceived behavioral control in regards to what referent others say is acceptable behavior (prescriptive
cyberloafing contribute or inhibit to the formation of intentions norms), and what referent others actually do (descriptive norms)
to cyberloaf. Further, for people who form intentions to cyberloaf, (Cialdini, Reno, & Kallgren, 1990; Park & Smith, 2007). In regards
their intentions lead directly to actual cyberloafing. to cyberloafing, prescriptive norms would be the extent that
The TPB has shown to be a valid model for predicting behaviors coworkers and supervisors would approve of the employee cyber-
conceptually similar to cyberloafing (Henle, Reeve, & Pitts, 2010; loafing and descriptive norms would be the extent to which cowork-
Pelling & White, 2009). For example, cyberloafing can be consid- ers and supervisors themselves cyberloaf (Askew et al., 2010). Both
ered a type of withdrawal behavior – behaviors that reduce the types of subjective norms have been implicated in cyberloafing
512 K. Askew et al. / Computers in Human Behavior 36 (2014) 510–519

(Askew et al., 2010; Blanchard & Henle, 2008; Restubog et al., 2011). The first hypothesis will examine the main assumption that led
Studies that have examined both norms in conjunction have found to the identification of the TPB as a potential model of cyberloafing
that descriptive norms account for a larger amount of unique vari- – the assumption that cyberloafing is a type of withdrawal
ance in cyberloafing than prescriptive norms (Askew et al., 2010). behavior. If cyberloafing is a withdrawal behavior, it should corre-
Given the greater predictive power of descriptive norms over pre- late with other withdrawal behaviors such as absenteeism, late-
scriptive norms, we consider descriptive norms to be the stronger ness, leaving early, and taking extended breaks. Thus, Hypothesis
candidate for the subjective norms variable. 1 is:

2.3. Hypotheses H1. Cyberloafing will correlate positively with overall withdrawal.
Cyberloafing will also correlate positively with the specific with-
Having discussed the details of the TPB, we now present our drawal behaviors absenteeism, lateness, leaving early, and taking
main model of cyberloafing, possible alternative models, and extended breaks.
hypotheses derived from these models. The main model that will
be tested is a TPB model with the ability to hide cyberloafing and The second set of hypotheses focus on testing predictions
descriptive norms as the perceived behavioral control and subjec- derived from our main model of cyberloafing – a TPB model con-
tive norms variables, respectively. The alternative models will sisting of the variables descriptive norms, the ability to hide cyber-
incorporate prescriptive norms and/or website access self-efficacy. loafing, and cyberloafing attitudes. We will test the following
The four models are shown in Fig. 1. predictions derived from the main model:

Descriptive
Norms Main Model

Cyberloafing Behavioral
Attitudes Cyberloafing
Intentions

Ability to
Hide

Prescriptive
Norms Alternative Model 1

Cyberloafing Behavioral
Attitudes Cyberloafing
Intentions

Ability to
Hide

Descriptive
Norms Alternative Model 2

Cyberloafing Behavioral
Attitudes Cyberloafing
Intentions

Web Access
Self-Efficacy

Prescriptive
Norms Alternative Model 3

Cyberloafing Behavioral
Attitudes Cyberloafing
Intentions

Web Access
Self-Efficacy

Fig. 1. Main and alternative models predicting cyberloafing.

 K. Askew et al. / Computers in Human Behavior 36 (2014) 510–519 513

Table 1
Study one – descriptive statistics, correlations, and reliabilities.

Variable Mean SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1. Cyberloafing 2.14 .94 .92
2. Descriptive norms 3.68 1.28 .36** .85
3. Prescriptive norms 2.69 .76 .19** .34** .85
4. Cyberloafing attitudes 4.92 1.50 .41** .26** .26** .87
5. Ability to hide 3.59 1.72 .41** .23** .09 .25** .92
6. Web access self-efficacy 4.05 1.85 .18** .15** .14** .15** .32** .86
7. Behavioral intentions 4.99 1.40 .53** .44** .29** .51** .29** .19** .79
8. Withdrawal 1.67 .59 .28** .17** .00 .14** .12* .04 .22** .65
9. Lateness 1.86 .93 .17** .15** .08 .14** .01 .01 .18** .69** –
10. Absenteeism 1.52 .64 .16** .07 .13* .04 .08 .05 .11* .61** .20** –
11. Extended break 1.83 .97 .20** .15** .07 .07 .09 .09 .14** .78** .33** .37** –
12. Leaving early 1.46 .79 .25** .09 .10* .14** .17** .01 .17** .71** .31** .31** .40** –
13. Conscientiousness 3.95 .60 .01 .00 .08 .06 .01 .01 .06 .28** .22** .12* .20** .23** –
14. Gender 1.75 .43 .08 .06 .01 .01 .07 .01 .00 .11* .08 .03 .09 .13** .02 –
15. Age 23.78 6.78 .03 .04 .07 .04 .06 .19** .07 .05 .03 .04 .03 .04 .04 .12* –

Note: Reliabilities bolded in the diagonal. Gender coded male = 1, female = 2.

N = 432.
*
p < .05.
**
p < .01.

H2. Descriptive norms, cyberloafing attitudes, and the ability to 3. Study 1

hide cyberloafing will predict cyberloafing incremental to each
other. 3.1. Method

3.1.1. Sample and procedure

H3. Descriptive norms, cyberloafing attitudes, and the ability to
For our initial test of the theory, we used a sample consisting of
hide cyberloafing will predict intentions incremental to each other.
both student and non-student employees. Student employees were
recruited via college classrooms and offered extra credit for their
H4. Intentions will predict cyberloafing incremental to descriptive participation. Non-student employees were recruited via the
norms, cyberloafing attitudes, and the ability to hide cyberloafing. snowballing method. To qualify for the study, participants had to
indicate that they worked at a job that involved working with a
computer. To prevent participants who did not meet this
H5. The predictive power of the three antecedents – descriptive
requirement from participating, a research questionnaire of equal
norms, cyberloafing attitudes, and ability to hide cyberloafing –
length was provided as an alternative. In total, 429 employees
to predict cyberloafing will be reduced when intentions is included
participated in the study (72.70% female, Mage = 23.78, Cauca-
as predictor in the regression model.
sian = 53.20%, African American = 15.10%, Asian = 3.80%,
Other = 23.90%).
The third set of hypotheses focus on testing alternative models
of cyberloafing – TPB models with subjective prescriptive norms
3.1.2. Measures
and/or website access self-efficacy. Note that these alternative
Below each of the measures used in Studies 1 and 2 are
models are not mutually exclusive to each other or the main TPB
described. The reliabilities for each scale in Study 1 are presented
model. To aid in presentation, we grouped predictions from each
in Table 1, and the reliabilities for each scale in Study 2 are
alternative model into single hypotheses.
presented in Table 3. The items for each of the measures used in
Studies 1 and 2 can be found in Appendix A.
H6. Prescriptive norms, cyberloafing attitudes, and the ability to
hide cyberloafing will predict cyberloafing incremental to each
3.1.2.1. Cyberloafing. Cyberloafing was measured using an
other. Regression mediation analyses will support intentions as the
extended 19-item version of Lim’s (2002) cyberloafing scale
mediator.
(Blanchard & Henle, 2008; Lim, 2002). Participants rated the fre-
quency they engaged in a list of cyberloafing behaviors on a 6-
H7. Descriptive norms, cyberloafing attitudes, and website access point scale (1 = Never, 4 = Once a day, 6 = Constantly).
self-efficacy will predict cyberloafing incremental to each other.
Regression mediation analyses will support intentions as the 3.1.2.2. Subjective social norms. Subjective prescriptive norms were
mediator. measured using Blanchard and Henle’s (2008) social norms scale.
Participants were asked to rate the extent that their supervisors
and coworkers would approve of them engaging in non-work
H8. Prescriptive norms, cyberloafing attitudes, and website access
related emailing and web-browsing. Subjective descriptive norms
self-efficacy will predict cyberloafing incremental to each other.
were measured using a similar scale derived from Blanchard and
Regression mediation analyses will support intentions as the
Henle’s scale (Askew et al., 2010), which asks participants to rate
mediator.4
the extent to which their supervisors or coworkers engage in
4
non-work related activities online.
We chose to use regression analyses to test our models over SEM techniques
following the recommendation from the APA Task Force on Statistical Inference to use
simpler analyses over complex ones when possible (i.e., the ‘‘minimally sufficient 3.1.2.3. Cyberloafing attitudes and intentions. Attitudes and inten-
analysis’’) (Wilkinson, 1999). tions were measured using scales developed for this study based
514 K. Askew et al. / Computers in Human Behavior 36 (2014) 510–519

on an article by Ajzen (2002), ‘‘Constructing a TPB Questionnaire: Table 2

Conceptual and Methodological Considerations’’. For both attitudes Study one – multiple regression analyses.

and intentions items are rated on a seven-point scale. The attitudes Criterion/predictor b 95% CI b R2
scale consists of four items asking participants to rate the extent to Main model
which they think cyberloafing is valuable, enjoyable, beneficial, H2: Cyberloafing .32**
and good. The intentions scale consists of six items asking partici- Cyberloafing attitudes .17 .12–.23 .28**
pants to rate their intentions to engage in six common cyberloafing Ability to hide .15 .11–.20 .28**
Descriptive norms .18 .12–.24 .24**
behaviors (e.g., web-browsing, sending email) in the coming
month. H3: Behavioral Intentions .38**
Cyberloafing attitudes .38 .30–.45 .41**
Descriptive norms .33 .25–.42 .31**
3.1.2.4. Perceived ability to hide cyberloafing. The perceived ability Ability to hide .10 .03–.16 .12**
to hide cyberloafing was measured using Askew et al.’s (2011). H4–5: Cyberloafing .39**
Ability to Hide Cyberloafing scale. Participants rated the extent to Behavioral intentions .22 .16–.29 .33**
which they can get away with cyberloafing on a seven-point scale. Ability to hide .13 .09–.17 .24**
An example item is: ‘‘I COULD hide what I do on my work com- Descriptive norms .11 .04–.17 .14**
Cyberloafing attitudes .09 .03–.15 .14**
puter from other employees’’.
Alternative Model 1
3.1.2.5. Website access self-efficacy. Website access self-efficacy was H6: Cyberloafing .28**
measured using a three item scale developed for this study. Since Cyberloafing attitudes .20 .14–.25 .32**
all employees with computers are effectively capable of navigating Ability to hide .17 .13–.22 .32**
Prescriptive norms .10 .00 to .20 .08
to a website, the scale focused on perceived behavioral control
based on whether the company blocks the websites they want to H6: Behavioral Intentions .32**
Cyberloafing attitudes .41 .33–.49 .44**
visit and their ability to circumnavigate this obstacle. An example
Prescriptive norms .29 .14–.44 .16**
item is: ‘‘I can get to any website I want to at work’’. Ability to hide .13 .07–.20 .16**
H6: Cyberloafing .38**
3.1.2.6. Withdrawal. Withdrawal was measured using four items Behavioral intentions .26 .19–.32 .38**
from Spector et al.’s (2006) CWB checklist. Participants were asked Ability to hide .14 .10–.18 .26**
to rate the frequency in which they engaged in four withdrawal Cyberloafing attitudes .09 .03–.15 .15**
Prescriptive norms .03 .07 to .12 .02
behaviors (absence, lateness, leaving early, and extended breaks)
on a 5-point scale (1 = Never, 5 = Every Day). Alternative Model 2
H7: Cyberloafing .25**
3.2. Results Cyberloafing attitudes .20 .15–.26 .32**
Descriptive norms .21 .14–.27 .28**
Web access self-efficacy .04 .00 to .08 .08
Bivariate correlations are shown in Table 1. Supporting Hypoth-
esis 1, cyberloafing was significantly correlated with all with- H7: Behavioral Intentions .38**
Cyberloafing attitudes .40 .32–.47 .43**
drawal behaviors and overall withdrawal (r = .28, p < .01). All six
Descriptive norms .34 .26–.43 .32**
of the predictors of cyberloafing posited by the TPB were signifi- Web access self-efficacy .06 .00–.12 .08
cantly correlated with cyberloafing. The strongest predictors of
H7: Cyberloafing .34**
cyberloafing in this study were intentions (r = .53, p < .01), ability Behavioral intentions .24 .18–31 .36**
to hide cyberloafing (r = .41, p < .01), attitudes (r = .41, p < .01), Descriptive norms .12 .06–.19 .17**
and descriptive norms (r = .36, p < .01). Age and gender were not Cyberloafing attitudes .11 .05–.16 .17**
Web access self-efficacy .03 .02 to .07 .05
significantly correlated with cyberloafing.
Regression analyses are shown in Table 2. The second set of Alternative Model 3
hypotheses (H2–5) tested our main model of cyberloafing – the H8: Cyberloafing .19**
TPB model with subjective descriptive norms, cyberloafing atti- Cyberloafing attitudes .23 .18–.29 .38**
tudes, and the ability to hide cyberloafing as predictors of cyberloa- Web access self-efficacy .05 .01–.10 .11*
Prescriptive norms .10 .01 to .21 .08
fing. Hypothesis 2 was supported; the three antecedents predicted
cyberloafing incremental to each other (bd.norms = .24, p < .01; batti- H8: Behavioral Intentions .31**
Cyberloafing attitudes .44 .36–.51 .47**
tudes = .28, p < .01; bATHC = .28, p < .01), accounting for almost a third
Prescriptive norms .27 .12–.42 .15**
of the variance in cyberloafing (R2 = .32, p < .01). Hypotheses exam- Web access self-efficacy .07 .01–.13 .09*
ining intentions as the mediator were examined based on Baron
H8: Cyberloafing .32**
and Kenny’s (1986) criteria, and were also supported: The three Behavioral intentions .29 .22–.35 .43**
antecedents predicted intentions incremental to each other (H3); Cyberloafing attitudes .11 .05–.17 .18**
intentions predicted cyberloafing incremental to the three anteced- Web access self-efficacy .03 .01 to .08 .07
ents (H4); and including intentions in the regression model with Prescriptive norms .02 .09 to .12 .02

cyberloafing as the criterion reduced the predictive power of the *

p < .05.
**
three antecedents (H5). Though the predictive power of the three p < .01.
antecedents of cyberloafing was reduced they remained significant
when behavioral intentions were entered, suggesting partial medi-
ation. In sum, our main model of cyberloafing consisting of the TPB However, none of the alternative models received complete sup-
variables descriptive norms, attitudes, and ability to hide cyberloa- port, which was the case for the main model. Prescriptive norms
fing was completely supported in Study 1. did not predict cyberloafing significantly in any of the regression
The third set of hypotheses (H6–8) focused on testing alterna- models with cyberloafing as a criterion; and models involving
tive models, TPB models including prescriptive norms and/or web- the website access self-efficacy variable had only two significant
site self-efficacy as predictors. Some support was found for all the predictors when cyberloafing was the criterion. Thus, only partial
models, as they shared similar predictions with the main model. support was found for Hypotheses 6–8.
 K. Askew et al. / Computers in Human Behavior 36 (2014) 510–519 515

4. Study 2 4.1.2. Measures

The same measures used in Study 1 were used in Study 2. To
4.1. Method keep the survey on a single page and therefore more attractive
to potential participants, we shortened the cyberloafing scale to
4.1.1. Sample and procedure seven items based on factor loadings and item content.
Participants were recruited in the downtown area of a major
U.S. metropolitan city. The first author and/or trained research
assistants approached people on the sidewalk who were walking 4.2. Results
alone and appeared to not be in a hurry (the walking-alone crite-
rion was employed to avoid non-independent observations that Study 2 bivariate correlations are shown in Table 3. All with-
would result from two workers in the same work-group complet- drawal behaviors were significantly correlated with cyberloafing,
ing the survey). The script used when approaching potential supporting Hypothesis 1. As expected, all TPB predictors were sig-
participants was: ‘‘Excuse me. I’m a student and I’m working on nificantly related to cyberloafing. Compared to the patterns of rela-
a study. Would you mind filling out a short 1-page survey?’’ Those tionships in Study 1, the relationships among the variables were
who agreed to participate were asked a follow-up question to higher. The relationships between cyberloafing and its predictors
ensure they were currently employed in a position that involves followed a similar rank-order as the observed magnitudes in Study
working with computers with Internet access; those who declined 1, with the exception of the ability to hide cyberloafing having a
our invitation to participate were politely thanked for their consid- relatively lower rank order in Study 2. This is perhaps due to the
eration. The participants would then fill out the survey in the street fact that Study 2 likely had more employees from higher levels
while the first author or a research assistant waited at a moderate in their organizations, employees who have more power, who
distance (about 40–50 feet) to allow the participant to feel a sense may have relatively less need to hide their cyberloafing (Vitak,
of privacy. After the participant completed the survey, we collected Crouse, & LaRose, 2011).
the survey and thanked him or her. This procedure allowed us to Study 2 regression results are shown in Table 4. The second set
collect data from participants in a wide-range of industries (e.g., of hypotheses (H2–5) focused on testing our main model of cyber-
legal, government, medical, non-profit, banking, communications, loafing – the TPB model with descriptive norms and the ability to
real estate, healthcare). The participation rate was about 20%. In hide cyberloafing as predictors of cyberloafing. Again, Hypotheses
total, 202 employees were surveyed. 54.7% of the sample was male, 2–5 were supported. The three variables predicted cyberloafing
and age was approximately normally distributed (see Fig. 2 for a incremental to each other (bd.norms = .25, p < .01; battitudes = .35,
distribution of age). p < .01; bATHC = .16, p < .01) and all mediation analyses were consis-
tent with intentions as the mediator. Descriptive norms, cyberloa-
fing attitudes, and the ability to hide cyberloafing accounted for
37% of the variance in cyberloafing, even higher than the 32% of
70
62 variance accounted for in Study 1. Thus, using a sample more rep-
60 resentative of the general working population, Study 2 cross-vali-
50 43 dates and replicates the findings of Study 1.
Frequency

40 34
The third set of hypotheses (H5–7) tested alternative TPB mod-
28 els – models including the prescriptive norms and/or website
30
access self-efficacy. All three alternative models were supported
20 17
by regression analyses with cyberloafing as the criterion, but not
10 6 with behavioral intentions as the criterion. Thus, for all three mod-
0 els, there was support for the predictors as distal causes of cyber-
18-25 26-35 36-45 46-55 56-65 65+ loafing, but some analyses were inconsistent with intentions as the
Age mediator. This is in contrast to our main model, in which all
hypotheses were supported. In sum, partial support was found
Fig. 2. Distribution of participant age in Study 2. for Hypotheses 6–8.

Table 3
Study two – descriptive statistics, correlations, and reliabilities.

Variable Mean SD 1 2 3 4 5 6 7 8 9 10 11 12 13
1. Cyberloafing 2.18 .99 .76
2. Descriptive norms 3.16 1.48 .49** .92
3. Prescriptive norms 2.40 1.00 .43** .52** .90
4. Cyberloafing attitudes 4.14 2.08 .52** .51** .47** .91
5. Ability to hide 3.46 1.88 .36** .31** .31** .36** .89
6. Web access self-efficacy 3.70 1.56 .23** .13 .16* .11 .17* .48
7. Behavioral intentions 2.73 1.07 .53** .55** .48** .60** .36** .01 .75
8. Withdrawal 1.64 .59 .31** .37** .22** .28** .30** .04 .27** .73
9. Absenteeism 1.51 .57 .24** .19** .06 .27** .30** .01 .20** .67** –
10. Extended break 1.76 .80 .21** .31** .20** .18* .24** .06 .13 .88** .42** –
11. Leaving early 1.65 .80 .31** .38** .25** .23** .22** .06 .31** .86** .36** .65** –
12. Gender 1.47 .56 .12 .08 .10 .17* .14 .07 .15 .21** .07 .21** .21** –
13. Age 3.45 1.28 .14* .00 .04 .02 .02 .05 .02 .05 .02 .04 .09 .16* –

Note: Reliabilities bolded in the diagonal. Gender coded male = 1, female = 2.

N = 200.
*
p < .05.
**
p < .01.
516 K. Askew et al. / Computers in Human Behavior 36 (2014) 510–519

Table 4 the alternative models. All three predictors of the main model –
Study two – multiple regression analyses. descriptive norms, attitudes, and the ability to hide cyberloafing
Criterion/predictor b 95% CI b R2 – were found to predict cyberloafing above and beyond the other
Main model predictors in both studies. The three predictors accounted for
H2: Cyberloafing .37** 32% and 37% of the variance in cyberloafing in Studies 1 and 2,
Cyberloafing attitudes .17 .10–.24 .35** respectively. Thus, results support the validity of the main TPB
Descriptive norms .17 .07–.27 .25** model as a model of cyberloafing.
Ability to hide .09 .02–.16 .16*
H3: Behavioral Intentions .44** 5.1. Contributions to the literature
Cyberloafing attitudes .19 .12–.26 .37**
Descriptive norms .22 .12–.32 .31**
Ability to hide .08 .00–.15 .13* Our investigation makes three substantial contributions to the
H4–5: Cyberloafing .41 ** theoretical side of understanding cyberloafing. First we identified,
Behavioral intentions .25 .10–.41 .27** tested, and supported a theory of cyberloafing that can address the
Cyberloafing attitudes .12 .04–.20 .25** short-comings of the Ego Depletion Model of Cyberloafing (Wagner
Descriptive norms .11 .00–.21 .16* et al., 2012). Thus, we believe we have contributed a second major
Ability to hide .07 .00–.14 .14*
theory of cyberloafing to the cyberloafing literature. The cyberloa-
Alternative Model 1 fing TPB can explain why people cyberloaf when they are fully-
H6: Cyberloafing .33** rested, and incorporates one of the most robust known predictors
Cyberloafing attitudes .18 .11–.25 .37** of cyberloafing, social norms (Askew et al., 2010; Blanchard &
Prescriptive norms .19 .05–.33 .19** Henle, 2008; Restubog et al., 2011). The theory is also part of a lar-
Ability to hide .09 .02–.16 .17*
ger framework, and thus cyberloafing researchers now have a long
H6: Behavioral Intentions .41** history from which to draw (Ajzen, 2005).
Cyberloafing attitudes .23 .16–.30 .45**
Second, our investigation places cyberloafing into the wider
Prescriptive norms .25 .10–.40 .22*
Ability to hide .06 .01 to .14 .11 counterproductive work behavior literature. We found that cyber-
loafing met the definition of withdrawal behavior (it is a behavior
H6: Cyberloafing .37**
Behavioral intentions .25 .10–.39 .27** that reduces the amount of time working to less than is expect by
Cyberloafing attitudes .13 .05–.20 .27** the organization) (Spector et al., 2006). Further, we found empirical
Ability to hide .08 .00–.15 .14* links between cyberloafing and withdrawal behaviors (absentee-
Prescriptive norms .09 .06 to .24 .09
ism, lateness, extended breaks, leaving early, and a composite
Alternative Model 2
withdrawal variable). As such, we conclude that cyberloafing
H7: Cyberloafing .36** appears to be a type of withdrawal behavior. If this is supported
Cyberloafing attitudes .19 .12–.26 .38** with future research, then it places cyberloafing as a lower-order
Descriptive Norms .18 .08–.27 .26** construct of counterproductive work behavior.
Web access self-efficacy .09 .01–.18 .17*
Our third contribution to the literature is the different perspec-
H7: Behavioral Intentions .43** tive of cyberloafing that arises when considering the nature of two
Cyberloafing attitudes .21 .14–.28 .42**
constructs in the model: social norms and the ability to hide cyber-
Descriptive norms .24 .14–.35 .34**
Web access self-efficacy -.05 .13 to .04 .07 loafing. Collectively, these two variables suggest that people are
motivated to cyberloaf but avoid doing so to the extent that they
H7: Cyberloafing .42**
Behavioral intentions .29 .14–.45 .31** believe it would be socially disapproved of and the extent to which
Cyberloafing attitudes .13 .05–.20 .26** they would get caught for doing it. This perspective of cyberloafing
Web access self-efficacy .10 .02–.18 .15* is exactly opposite of the current dominant self-control perspective
Descriptive norms .10 .00–.21 .15
of cyberloafing. From the self-control perspective, employees are
Alternative Model 3
trying to work but may lack self-control (or become depleted of
H8: Cyberloafing .33** their self-control resources) (Prasad et al., 2010; Restubog et al.,
Cyberloafing attitudes .20 .13–.26 .41** 2011). In the perspective adopted here, employees are trying to
Prescriptive norms .20 .06–.34 .20** cyberloaf and only work to the extent they have to. We believe
Web access self-efficacy .11 .03–.18 .17**
our theory complements the current dominant perspective and
H8: Behavioral Intentions .41** helps provide an understanding of cyberloafing across a wider
Cyberloafing attitudes .25 .18–.31 .48**
set of circumstances.
prescriptive norms .29 .14–.44 .26**
Web access self-efficacy -.06 .14 to .03 .08
5.2. Implications for practice
H8: Cyberloafing .38**
Behavioral intentions .29 .14–.43 .32**
Cyberloafing attitudes .13 .06–.20 .28** The current investigation also makes two substantial contribu-
Web access self-efficacy .11 .03–.19 .17** tions to organizational practice in regards to cyberloafing. First, the
Prescriptive norms .09 .06 to .23 .08 present situation is that there are many decision makers in organi-
*
p < .05. zations who are concern about cyberloafing (Scheuermann &
**
p < .01. Langford, 1997; Stewart, 2000) but are apprehensive to use
heavy-handed practices such as restricting all computer use or
monitoring all internet traffic (de Lara et al., 2006). These decision
5. Discussion makers are in a difficult situation because the high frequency in
which cyberloafing occurs (Harris Interactive., 2006; Wallace, P.,
In this investigation, we tested the TPB as a model of cyberloa- 2004; Wallace, P.M., 2004) means that there is high potential for
fing. In Study 1, we conducted an initial test of the theory. In Study loss of productivity – if indeed cyberloafing reduces productivity
2, we attempted to replicate our findings using a sample that is (Stewart, 2000). Our model, and more specifically, the ability to
fairly representative of the general working population. Results hide cyberloafing, suggests a non-harmful way to reduce cyberloa-
unanimously supported our main model and partially supported fing: structure the work environment in a way that increases the
 K. Askew et al. / Computers in Human Behavior 36 (2014) 510–519 517

transparency of computer-mediated activities. Orienting computer relatively preserved, this limitation should not have influenced
screens so that they face hallways instead of walls, setting up cubi- our results.
cles in ways that do not allow the employee to see people A fourth limitation is that our study does not address the con-
approaching their work station, and not isolating workers are three sequences of cyberloafing. This precludes us from making strong
things organizational decision makers can do to reduce cyberloa- recommendations in which to give organizational decision makers
fing without being too heavy-handed. In all likelihood, having such tasked with setting internet usage policies. Nonetheless, our
transparency will lead to a reduction of copious amounts of cyber- studies provide a means to influence cyberloafing once the conse-
loafing, while still allowing employees to engage in some of the quences of different cyberloafing behaviors are better understood.
online activities they find enjoyable (Askew et al., 2011). Future research should investigate the consequences of cyberloa-
The second contribution to practice from this investigation fing. After that, research should focus on testing internet usage pol-
comes from the fact that TPB is a framework that has already been icies, and the results of those studies should be used to inform
used extensively to change behaviors. For example, the TPB has practice.
been used to promote effective job search behaviors (Ryn &
Vinokur, 1992), discourage car use (Bamberg, Ajzen, & Schmidt, 5.4. Summary and conclusion
2003), limit sugar intake (Beale & Manstead, 1991), conserve
energy (Harland, Staats, & Wilke, 1999) and reduce drug-use In this investigation, we aimed to fill a critical theoretical gap in
(Armitage, Armitage, Conner, Loach, & Willetts, 1999). At the the cyberloafing literature – to understand why people cyberloaf
moment, caution should be used in applying TPB interventions to when they are fully rested – and to provide and test a theory of
the reduction of cyberloafing, simply because we do not under- cyberloafing. From the literature we identified a potential theory
stand the consequences of cyberloafing in organizations very well of cyberloafing and tested that theory in two studies. Results from
(Askew et al., 2012). However, once we do understand the conse- the two studies unanimously supported the TPB as a theory of
quences of cyberloafing – and what cyberloafing behaviors we cyberloafing. Specifically, we found evidence that cyberloafing is
want to encourage, discourage and tolerate – we have a valid the- a withdrawal behavior, and that the three predictors – subjective
ory to serve as the basis for interventions (Ajzen, 2011). descriptive norms, cyberloafing attitudes, and perceived ability to
hide cyberloafing – predict cyberloafing above and beyond each
other. We also examined other possible variations of the theory,
5.3. Limitations and future directions helping pin down the exact nature of the theory as it relates to
cyberloafing.
The current studies have a number of limitations. First, the data In short, we have provided evidence for the second major the-
are all cross-sectional, limiting our ability to make causal infer- ory of cyberloafing. The first major theory, the ego depletion model
ences. In our studies, we found relationships that are consistent of cyberloafing, explains why people cyberloaf when they are moti-
with the posited causal model, but it is possible that the directions vated to get work done (Baumeister et al., 2000; Wagner et al.,
of causation are reversed or reciprocal. For example, perhaps 2012). The present theory explains why people cyberloaf when
cyberloafing attitudes do not cause cyberloafing, but rather people they are not drained of self-control resources. The next step for
develop attitudes to justify their cyberloafing. This limitation is cyberloafing researchers is to pin down the consequences of cyber-
somewhat mitigated by the fact that the TPB antecedents have loafing in actual organizations. After that, research should focus on
been shown to be causal in other domains (Ajzen, 2011), and the testing internet usage policies and the results of those studies
fact that TPB interventions have been successful in changing should be used to inform practice.
behaviors (Ajzen, 2011). Nonetheless, future studies should use a
longitudinal design to help establish the temporal precedent of Appendix A. Scales and items used
the exogenous variables.
A second limitation is that the results of the studies might not Descriptive norms
generalize to all work populations. Our sample in Study 2, while How often do your COWORKERS do each of the following
higher on the representative continuum, is no doubt not a fully things during work hours?
representative sample of people who use a computer at work. Even – Visit non-job related websites.
if the sample was perfectly representative of the general working – Check non-work related email.
population, our results still might not generalize to specific working – Visit social networking sites (Facebook, etc.).
populations. For example, in companies with high levels of nepo- How often do your SUPERVISORS do each of the following
tism, organizational justice could be the major driver of cyberloa- things during work hours?
fing (Ahmadi, Bagheri, Ebrahimi, Rokni, & Kahreh, 2011). The – Visit non-job related websites.
generalizability of our results are boosted greatly by the consis- – Check non-work related email.
tency across two very different samples, but nevertheless future – Visit social networking sites (Facebook, etc.).
studies using a variety of populations will be necessary to ulti-
mately examine the generalizability of the results. Prescriptive norms
Third, we used a cyberloafing scale to measure cyberloafing My coworkers would approve of me. . .
instead of directly recording cyberloafing behavior. Thus, the pre- – . . .visiting non-job related websites.
cision in which we measured cyberloafing depends on people’s – . . .sending/receiving non-work related emails.
willingness and ability to recall their cyberloafing behavior in a – . . .visiting social networking sites (Facebook, etc).
typical work week. It was not possible for us to measure cyberloa- My supervisors would approve of me. . .
fing directly in all of the various organizations sampled here. Thus, – . . .visiting non-job related websites.
we had to resort to using an established cyberloafing scale (Lim, – . . .sending/receiving non-work related emails.
2002). However, the use of a scale over directly measuring behav- – . . .visiting social networking sites (Facebook, etc).
ior is only problematic to the extent that the use of the scale results
in different covariances among the constructs than the use of a
direct measure. As long as the rank order of cyberloafers was (continued on next page)
518 K. Askew et al. / Computers in Human Behavior 36 (2014) 510–519

Behavioral intentions Blanchard, A. L., & Henle, C. A. (2008). Correlates of different forms of cyberloafing:
The role of norms and external locus of control. Computers in Human Behavior,
– I intend to shop online while at work at least once in the 24(3), 1067–1084.
forthcoming month. Brouwer, S., Krol, B., Reneman, M. F., Bültmann, U., Franche, R. L., van der Klink, J. J.,
– I will use my phone for personal reasons while at work at et al. (2009). Behavioral determinants as predictors of return to work after long-
term sickness absence: An application of the theory of planned behavior. Journal
least once in the forthcoming month. of Occupational Rehabilitation, 19(2), 166–174.
– I will send at least a few text messages while at work in the Case, C. J., & Young, K. S. (2002). Employee Internet management: Current business
forthcoming month. practices and outcomes. CyberPsychology & Behavior, 5(4), 355–361.
Cialdini, R. B., Reno, R. R., & Kallgren, C. A. (1990). A focus theory of normative
– I intend to send a non-work related email at least once in conduct: Recycling the concept of norms to reduce littering in public places.
the forthcoming month. Journal of Personality and Social Psychology, 58(6), 1015–1026.
– I plan to browse non-work related websites at work at de Lara, P. Z. M., Tacoronte, D. V., & Ding, J. M. T. (2006). Do current anti-
cyberloafing disciplinary practices have a replica in research findings? A study
least a few times in the forthcoming month. of the effects of coercive strategies on workplace Internet misuse. Internet
– I plan to use a social networking site (ex. Facebook) while Research, 16(4), 450–467.
at work at least once in the forthcoming month. Everton, W. J., Mastrangelo, P. M., & Jolton, J. A. (2005). Personality correlates of
employees’ personal use of work computers. CyberPsychology & Behavior, 8(2),
143–153.
Cyberloafing attitudes Harland, P., Staats, H., & Wilke, H. A. (1999). Explaining proenvironmental intention
For me, using the internet at work for personal reasons is. . . and behavior by personal norms and the Theory of Planned Behavior. Journal of
– Worthless/valuable Applied Social Psychology, 29(12), 2505–2528.
Harris Interactive (2006). IT decision-makers survey results. Websense, Inc. Retrieved
– Unenjoyable/enjoyable from <http://www.securitymanagement.com/archive/library/websense_
– Harmful/beneficial technofile0906.pdf>.
– Bad/good Henle, C. A., Kohut, G., & Booth, R. (2009). Designing electronic use policies to
enhance employee perceptions of fairness and to reduce cyberloafing: An
empirical test of justice theory. Computers in Human Behavior, 25(4),
Ability to hide cyberloafing 902–910.
– I COULD hide what I do on my work computer from other Henle, C. A., Reeve, C. L., & Pitts, V. E. (2010). Stealing time at work: Attitudes, social
pressure, and perceived control as predictors of time theft. Journal of Business
employees Ethics, 94(1), 53–67.
– I COULD pretend to be working on my computer and peo- Jia, H.H. (2008). Relationships between the big five personality dimensions and
ple would never know cyberloafing behavior. (Doctoral Dissertation). Retrieved from Proquest Digital
Dissertations.
– I COULD hide my computer activity if I wanted to Liberman, B., Seidman, G., McKenna, K. Y., & Buffardi, L. E. (2011). Employee job
attitudes and organizational characteristics as predictors of cyberloafing.
Website access self-efficacy Computers in Human Behavior, 27(6), 2192–2199.
Lim, V. K. G. (2002). The IT way of loafing on the job: Cyberloafing, neutralizing and
– My favorite websites are blocked at work.
organizational justice. Journal of Organizational Behavior, 23(5), 675–694.
– My company blocks access to certain sites. Lim, V. K. G., & Teo, T. S. H. (2005). Prevalence, perceived seriousness, justification
– I can get to any website I want to at work. and regulation of cyberloafing in Singapore: An exploratory study. Information
& Management, 42(8), 1081–1093.
Lim, V. K., & Teo, T. S. (2006). Cyberloafing and organizational justice. The Internet
and Workplace Transformation, 241–258.
Lu, Y., Zhou, T., & Wang, B. (2009). Exploring Chinese users’ acceptance of
References instant messaging using the theory of planned behavior, the technology
acceptance model, and the flow theory. Computers in Human Behavior, 25(1),
Ahmadi, H., Bagheri, F., Ebrahimi, S. A., Rokni, M. A. N., & Kahreh, M. S. (2011). 29–39.
Deviant work behavior: explaining relationship between organizational justice Mastrangelo, P. M., Everton, W., & Jolton, J. A. (2006). Personal use of work
and cyber-loafing as a deviant work behavior. American Journal of Scientific computers: Distraction versus destruction. CyberPsychology & Behavior, 9(6),
Research, 24, 103–116. 730–741.
Ajzen, I. (1985). From intentions to actions: A theory of planned behavior (pp. 11–39). Mathieson, K. (1991). Predicting user intentions: Comparing the technology
Berlin Heidelberg: Springer. acceptance model with the theory of planned behavior. Information Systems
Ajzen, I. (2002). Constructing a TPB questionnaire: Conceptual and methodological Research, 2(3), 173–191.
considerations. Retrieved from <http://www.unibielefeld.de/ikg/zick/ Naughton, K., Raymond, J., & Shulman, K. (1999). Cyberslacking. Newsweek, 134(22),
ajzen%20construction%20a%20tpb%20questionnaire.pdf>. 62–65. November.
Ajzen, I. (2005). Attitudes, Personality and Behavior 2e. McGraw-Hill International. Ozler, D. E., & Polat, G. (2012). Cyberloafing phenomenon in organizations:
Ajzen, I. (2011). The theory of planned behavior. Handbook of Theories of Social Determinants and impacts. International Journal of eBusiness and eGovernment
Psychology: Volume One, 1, 438. Studies, 4(2). 2146–0744.
Armitage, C. J., Armitage, C. J., Conner, M., Loach, J., & Willetts, D. (1999). Different Park, H. S., & Smith, S. W. (2007). Distinctiveness and influence of subjective norms,
perceptions of control: Applying an extended theory of planned behavior to personal descriptive and injunctive norms, and societal descriptive and
legal and illegal drug use. Basic and Applied Social Psychology, 21(4), 301–316. injunctive norms on behavioral intent: A case of two behaviors critical to
Askew, K., Coovert, M. D., Taing, M. U., Ilie, A., & Bauer, J. (2012). Work environment organ donation. Human Communication Research, 33(2), 194–218.
factors and cyberloafing: A follow-up to Askew. San Diego, CA: Poster presented at Pelling, E. L., & White, K. M. (2009). The theory of planned behavior applied to young
SIOP. people’s use of social networking web sites. CyberPsychology & Behavior, 12(6),
Askew, K., Vandello, J.A., & Coovert, M.D. (2010). Cyberloafing and Social Norms: 755–759.
The Role of Subjective Prescriptive and Descriptive Norms, unpublished Prasad, S., Lim, V.K., & Chen, D.J. (2010). Self-regulation, individual characteristics
manuscript. and cyberloafing. PACIS 2010 Proceedings. Paper 159. Retrieved from http://
Askew, K., Coovert, M.D., Vandello, J.A., Taing, M.U., & Bauer, J.A. (August, 2011). aisel.aisnet.org/pacis2010/159.
Work environment factors predict cyberloafing. Poster presented at the Annual Restubog, S. L. D., Garcia, P. R. J. M., Toledano, L. S., Amarnani, R. K., Tolentino, L. R., &
Meeting of the American Psychological Society. Washington D.C. Tang, R. L. (2011). Yielding to (cyber)-temptation: Exploring the buffering role
Bamberg, S., Ajzen, I., & Schmidt, P. (2003). Choice of travel mode in the theory of of self-control in the relationship between organizational justice and
planned behavior: The roles of past behavior, habit, and reasoned action. Basic cyberloafing behavior in the workplace. Journal of Research in Personality,
and Applied Social Psychology, 25(3), 175–187. 45(2), 247–251.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in Ryn, M., & Vinokur, A. D. (1992). How did it work? An examination of the
social psychological research: Conceptual, strategic, and statistical mechanisms through which an intervention for the unemployed promoted
considerations. Journal of Personality and Social Psychology, 51(6), 1173–1182. job-search behavior. American Journal of Community Psychology, 20(5),
Baumeister, R. F., Muraven, M., & Tice, D. M. (2000). Ego depletion: A resource 577–597.
model of volition, self-regulation, and controlled processing. Social Cognition, Scheuermann, L. E., & Langford, H. P. (1997). Perceptions of Internet abuse, liability,
18(2), 130–150. and fair use. Perceptual and Motor Skills, 85(3), 847–850.
Beale, D. A., & Manstead, A. S. (1991). Predicting mothers’ intentions to limit Spector, P. E., Fox, S., Penney, L. M., Bruursema, K., Goh, A., & Kessler, S. (2006). The
frequency of infants’ sugar intake: Testing the theory of planned behavior. dimensionality of counterproductivity: Are all counterproductive behaviors
Journal of Applied Social Psychology, 21(5), 409–431. created equal? Journal of Vocational Behavior, 68(3), 446–460.
 K. Askew et al. / Computers in Human Behavior 36 (2014) 510–519 519

Stewart, F. (2000). Internet acceptable use policies: Navigating the management, Wallace, P. M. (2004). The internet in the workplace. New York, NY:
legal, and technical issues. Information Systems Security, 9(3), 1–7. Cambridge Press.
Vitak, J., Crouse, J., & LaRose, R. (2011). Personal Internet use at work: Wallace, P. (2004). The Internet in the workplace: How new technology is transforming
Understanding cyberslacking. Computers in Human Behavior, 27(5), 1751–1759. work. Cambridge University Press.
Wagner, D. T., Barnes, C. M., Lim, V. K. G., & Ferris, D. L. (2012). Lost sleep and Wilkinson, L., & Task force on statistical inference, APA Board of Scientific Affairs
cyberloafing: Evidence from the laboratory and a daylight saving time quasi- (1999). Statistical methods in psychology journals: Guidelines and
experiment. Journal of Applied Psychology, 97, 1068–1076. explanations. American Psychologist, 54, 594–604.