Antecedents and consequences of cyberloafing: Evidence from the Malaysian ICT industry
First Monday

Antecedents and consequences of cyberloafing: Evidence from the Malaysian ICT industry by Kian Yeik Koay, Patrick Chin-Hooi Soh, and Kok Wai Chew



Abstract
Cyberloafing poses a serious threat to organizations. Seeking a comprehensive understanding of cyberloafing, this paper tested a model of antecedents and consequences of cyberloafing. A total of 301 ICT employees data were collected and analysed with variance-based structural equation modelling (Smart-PLS). The model revealed that perceived favourable consequences, affect and social factors were significant factors for intention to cyberloaf while private demands were not significant. Furthermore, intention, habit, and perceived favourable facilitating conditions were significant factors for actual cyberloafing behaviour. In regards to consequences of cyberloafing, it was found that cyberloafing has a significant relationship with job stress but not with work performance. The study discusses these findings and concludes with the limitations and future recommendations.

Contents

1. Introduction
2. Theoretical framework
3. Development of hypotheses
4. Methodology
5. Data analysis
6. Discussions and implications
7. Limitations and future recommendations

 


 

1. Introduction

The Internet has been revolutionizing our workplace practices and is a critical tool for communications, knowledge dissemination and global access to markets (Anandarajan, et al., 2000). On the flip side, the Internet is a double-edge sword that has inadvertently opened up new avenues for employees to slack by deceptively engaging in non-work-related online activities (Jia, et al., 2013). Cyberloafing is a term used to describe employees’ non-work-related online activities during office hours (Lim, 2002). Based on Robinson and Bennett’s (1995) typology of deviant workplace behavior, cyberloafing is classified under the category of workplace production deviance (Lim, 2002) in which employees intentionally avoid doing work. While similar to the traditional ways of slacking such as leaving work early, taking excessive breaks, or intentionally walking slowly, cyberloafing is more dangerous in a sense that it allows employees to pretend to be industrious and increases security and litigation risks to corporations. Employees typically use the Internet during office hours for personal non-work related functions such as personal communication, playing online games, watching videos, or surfing social networking sites (Lim, 2002; Lim, et al., 2002).

Despite being considered as an acceptable practice in the workplace (Ramayah, 2010), employees admit cyberloafing to be a morally wrongful behavior (Ahmad and Jamaluddin, 2009). Several studies have reported that the employees spend an average of two hours daily cyberloafing (Greengard, 2000; Gouveia, 2013). In fact, about 60 percent of time online purchases were made during office hours and most of these purchases are non-work-related (StaffMonitoring.com, 2015). Improper Internet use during office hours has serious negative ramifications for individuals, organizations, and society. Firstly, cyberloafing might reduce job productivity (Young, 2001). Secondly, it may result in network bandwidth degradation and congestion hindering the productivity of coworkers (Moody and Siponen, 2013). Thirdly, cyberloafing increases security risks to corporate data (Johnson and Indvik, 2004). Fourth, personal Internet use exposes organizations to unnecessary legal liabilities such as defamation, sexual harassment, disseminating harmful information, and downloading prohibited material (unauthorised or unlicensed software) (Sipior and Ward, 2002; Weatherbee, 2010). It is estimated that US$759 billion is lost annually as a result of employees’ cyberloafing (Martin, et al., 2010). These dangers are present and real; for instance, the government of Singapore is planning to block civil servants from accessing the Internet through their work computers (British Broadcasting Corporation (BBC), 2016).

Various theories have been employed in prior studies in order to understand the antecedents of cyberloafing and these include deterrence theory (Ugrin and Pearson, 2013), social exchange theory (Lim, 2002), theory of planned behavior (TPB) (Seymour and Nadasen, 2007; Askew, et al., 2014), theory of interpersonal behavior (TIB) (e.g., Moody and Siponen, 2013; Betts, et al., 2014), self-control and the general theory of crime (Restubog, et al., 2011), work/family border theory (König and de la Guardia, 2014), personality theory (Everton, et al., 2005), and agency theory (Glassman, et al., 2015). After reviewing these literature, we found that the theory of interpersonal behavior (TIB) to be most suitable theory to study cyberloafing. TIB also provided the highest predictive power with the coefficient of determination r2 on cyberloafing up to 0.867 in Moody and Siponen (2013)’s study.

TIB posits that actual behavior is determined by intention, habit, and facilitating conditions while intention is driven by perceived consequences, affect, and social factors. However, TIB has several limitations that the theory incorporates only the social-psychological aspects of employees (internal antecedents) and does not account the possible impact of the non-work domain on cyberloafing behavior, which is an external factor. According to the family-work border theory, employees are proactively making transitions between non-work and work-domain in order to balance work and family demands (König and Caner de la Guardia, 2014). Employees’ social and/or family demands (from non-work domain) can significantly influence employees’ behavior in the workplace and vice versa. For this reason, this paper investigates the role of private demands on employees’ intention to cyberloaf.

Secondly, the central focus of extant cyberloafing studies has been predominantly on the antecedents of cyberloafing (e.g., Cheng, et al., 2014; König and de la Guardia, 2014), with little attention given to understanding the consequences of cyberloafing. Some scholars suggest that cyberloafing can be a palliative coping strategy against stress and burnout. There are also mixed findings on the impact of cyberloafing on employees’ work performance, with literature reporting negative, positive and no significant effect. Therefore, it is crucial for researchers to also examine the consequences of cyberloafing, so that practitioners can have a holistic view of cyberloafing and be able to formulate an effective strategy to manage it. Hence, the aim of this study is to investigate the antecedents and consequences of cyberloafing.

 

++++++++++

2. Theoretical framework

2.1. Theory of interpersonal behavior (TIB)

TIB provides a theoretical framework to explain different types of behaviors (Triandis, 1977). It is a synthesized model, enhanced from the theory of reasoned action (TRA) and theory of planned behavior (TPB). Various researchers (e.g., Milhausen, et al., 2006; Pee, et al., 2008) have proposed TIB to be a comprehensive behavioral model that offers stronger explanatory power than TRA and TPB. Their reasons are as follows: first, TRA and TPB focuses only on the cognitive aspect of the behavior and neglects the importance of emotional appeals (Triandis, 1977; Moody and Siponen, 2013). Hence, affect was added to TIB in predicting intention. Affect refers to the emotional response, which can be either favorable or unfavorable feelings experienced by a person when performing the particular behavior (Betts, et al., 2014). Emotions can be more compelling in driving behavior than cognitive motivations. Second, according to TIB, intention and habit (previous occurrence of the same behavior) drives behavior (Sheppard, et al., 1988). TRA and TPB failed to incorporate the possibility that a person’s repeated behavior can become an automatic behavior without conscious deliberation (Triandis, 1977). For example when someone arrives at a stop sign, the decision to slow down his or her vehicle does not require much cognitive mental process as this decision has been repeated many times until it becomes an automated response (Moody and Siponen, 2013). Prior studies have shown that habit has strong predictive power for cyberloafing behavior (e.g., Moody and Siponen, 2013; Betts, et al., 2014).

 

++++++++++

3. Development of hypotheses

3.1. Intention

According to TIB, intention is defined as a person’s conscious plan or self-instruction to perform the behavior (Triandis, 1980). Pee et al. [1] added that intention includes “a subjective probability dimension linking individuals to the behavior and indicates how much effort individuals are willing to invest to engage in it”. Intention is a widely accepted cognitive antecedent of behavior. The higher the levels of intention to cyberloaf, the more likely a person will engage in cyberloafing. This relationship is consistently supported by many empirical studies, such as Galletta and Polak (2003); Woon and Pee (2004); Askew, et al. (2014); and Betts, et al. (2014).

Hypothesis 1 (H1): Higher levels of intention to cyberloaf will lead to higher levels of actual cyberloafing behavior.

3.2. Perceived consequences

Based on TIB, an act is “perceived as having potential outcomes that have either positive or negative value, together with the probability of occurrence of each outcome” [2]. A person’s decision to act in a certain way is motivated by the conduct of the specific acts and by the expected eventualities (Thompson, et al., 1991). In the context of cyberloafing, perceived consequences is “the degree to which cyberloafing behavior is positively and negatively valued by an individual; it is an assessment of the total set of outcomes that come as a result using the Internet at work for non-related purposes” [3]. Similar to the construct of perceived usefulness in theory of acceptance model (Davis, 1989), favourable perceived outcomes of the performed behavior will increase the likelihood of the person performing the behavior. Perceived consequences is a reflective-formative construct, represented by two sub-dimensions namely degree of benefits and severity of negative consequences (Betts, et al., 2014).

The perceived benefits such as entertainment, saving time, information retrieval, social interaction are the motivating factors for employees’ degree to cyberloaf. Prior studies have found that Internet usage is associated with expectations of better job performance (Cheung, et al., 2000; Chang and Cheung, 2001). On the other hand, people are willing to go against the established laws, rules, norms, procedures, and guidelines when the risks pay off. The general deterrence theory suggests that the imposition of punishment can deter people from committing illicit behaviors. The consciousness of penalties for misbehavior may be sufficient to discourage illicit behavior (Ugrin and Pearson, 2013). In order to deter employees from cyberloafing, the severity of sanctions must be sufficiently strong such as job termination, warning letters, and reprimands (Ugrin and Pearson, 2008). In general, perceived positive benefits coupled with low severity levels of negative consequences facilitate people breaking rules and regulation. Extant studies found that perceived favourable consequences are positively related with the intention to cyberloaf (Pee, et al., 2008; Betts, et al., 2014).

H2: Higher levels of perceived favourable consequences will be positively associated with higher levels of employees’ intention to use the Internet for cyberloafing.

3.3. Affect

Triandis (1977) pointed out that decisions made by an individual are often not solely based on the cognitive aspects of a situation. In many cases, affect plays an important role in an individual’s decision-making. Affect may be positive and negative emotional responses with different degrees of strength, and is governed by instinctive behavioral responses to specific situations (Parikh and Gupta, 2010). As such, decisions influenced by emotions are different from rational instrumental evaluations of consequences. Affect represents an emotional state evoked by the act of a particular behavior (Gagnon, et al., 2006). Certain behaviors (conditioned stimulus) are attached with certain emotions (unconditioned stimuli) (Triandis, 1977). The emotions attached to certain behaviors are specific to the individual and cannot be generalized to everyone. According to Pee, et al. (2008), individuals can elicit feelings such as joy, excitement, and depression from cyberloafing activities. Several researchers have reported a significant positive relationship between affect and intention to cyberloaf (Pee, et al., 2008; Betts, et al., 2014). For example, when employees feel enjoyment, excitement, and positive sensations as the result of using the Internet to cyberloaf, the intention to cyberloaf will be higher.

H3: Higher levels of affect will be positively associated with higher levels of employees’ intention to use the Internet for cyberloafing.

3.4. Social factors

Social factors is defined as the individual’s internalization of the reference group’s subjective culture, and specific interpersonal agreements that the individual has made with others in specific social situations [4]. People are social beings seeking for social acceptance, trying to satisfy and please others, and looking for clues in others’ behavior to guide their own behavior in certain settings so that they can “fit in” with reference groups (Solomon, 2013). Reference groups are social groups that people refer to when evaluating their [own] qualities, circumstances, attitudes, values, and behaviors (Thompson and Hickey, 2005) and can be family members, friends, co-workers supervisors, and management. Because of the pressure and the desire to conform to the expectations of their reference groups, social norms play a significant influence on a person’s decision whether to perform a particular activity (Ajzen, 1991).

When cyberloafing is perceived as an unacceptable behavior that goes against the standard norms of the referent groups (management), the intention to cyberloaf will be lowered in order to avoid potential conflicts in the workplace and to avert censure by the reference groups. Galletta and Polak (2003) found that both supportive peer culture and supportive supervisor are significantly related to personal Internet use. When co-workers themselves engage in personal Internet use and supervisors allows or does not enforce the policy of Internet use, employees will be more likely to develop the intention to cyberloaf. Extant studies (e.g., Moody and Siponen, 2013; Betts, et al., 2014) have found that social factors are significant to the intention to cyberloaf.

H4: Higher levels of social factors will be positively associated with higher levels of employees’ intention to use the Internet for cyberloafing.

3.5. Private demands

Studies suggest that cyberloafing behavior can be attributed to the increasingly diffused boundary separating the domains of work and non-work (König and Caner de la Guardia, 2014). Drawing on the work/family border theory that postulates individuals are often proactively trying to juggle work/life demands, employees are using cyberloafing as a means to this end. Employees actively cross the border between work and non-work domains in order to attempt to meet the demands from both the domains. People often hold different roles in society for example being a part-time English teacher of a tuition centre, a personal coach of a badminton club, or a volunteer of a charity club. People who have high private demands are more likely to cyberloaf than those who have less (König and Caner de la Guardia, 2014). Private demands can be defined as “the obligations that people have towards others who do not belong to their work domain” [5].

Given the ease of online access from both during and after office hours, staying online all the time for work-related purposes has become a habit and lifestyle. Furthermore, employees nowadays are also expected to “be available” after office hours. For example, it is common to reply to e-mail messages and receive calls from customers or superiors after office hours. Hence, employees might feel justified by replying to private e-mail messages or paying house bills via online during office hours (König and Caner de la Guardia, 2014). The shift in work patterns has diminished the boundary between the domains of work and home (Lim and Teo, 2005; Weatherbee, 2010). Based on the discussion above, private demands are postulated to have a significant positive relationship with employees’s intention to cyberloaf. This is an augmentation to the original TIB model.

H5: Higher levels of private demands will be positively associated with higher levels of employees’ intention to use the Internet for cyberloafing.

3.6. Habit

Habit is a mental construct developed by repetition of the particular action or behavior (Schneider and Shiffrin, 1977; LaRose, 2010). Habit can be defined as situation-behavior sequences that are or have become automatic without self-instruction or with little conscious awareness in response to specific cues in the environment (Pee, et al., 2008). The frequent behavior in the past, the rewards as a result of the behavior and the constant setting are the important elements for the formation of habit (Ouellette and Wood, 1998). Intentions are formed with a conscious and deliberate mind in response to environment events. However, with constant practise and repetition, the behavior will operate with minimal conscious guidance (Verplanken, et al., 1997; Verplanken and Wood, 2006). Once the habit is formed, intentions no longer require much conscious decision-making as the behavior has become automatic or self-triggered (Ouellette and Wood, 1998). Several studies concluded that habit is the most significant antecedent to predict cyberloafing as compared to intention (Pee, et al., 2008; Moody and Siponen, 2013; Betts, et al., 2014).

H6: Higher levels of habit will be positively associated with higher levels of actual cyberloafing behavior.

3.7. Facilitating condition

Defined by Triandis (1980), facilitating conditions refer to individuals’ environment factors that make a behavior easy to be carried out. In the context of cyberloafing, it is less likely for employees to cyberloaf when the behavior is difficult to be carried out and can result in severe punishment that outweighs the expected benefits. For example, employees may find it difficult to cyberloaf in an open-office environment because of the visibility of their behavior. Similarly, strict Internet usage policy and severe disciplinary punishments may deter employees to cyberloaf (Jia, et al., 2013). However, in order to mitigate against any negative feelings associated with the invasion of their privacy, employees must be informed that they are being monitored (Betts, et al., 2014; Glassman, et al., 2015). Machado, et al. (2014) highlight that having some forms of restrictions on Internet use in the workplace increase self-regulation and reduce cyberloafing. In Pee, et al. (2008), facilitating conditions were found to have significant positive relationship with cyberloafing. However, Moody and Siponen (2013) report that facilitating conditions have no significant relationship with cyberloafing. With such mixed results in the current literature, this research hopes to provide a better understanding of the role of facilitating conditions in cyberloafing.

H7: Higher levels of facilitating conditions will be positively associated with higher levels of actual cyberloafing behavior.

3.8 Cyberloafing and job stress

Job stress refers to employees’ emotional experience that is associated with strain, anxiety, and tenseness generated from his employment (Cooke and Rousseau, 1984). Without proper stress management, employees may eventually lose their motivation to work and experience a state of extreme fatigue and lower job performance (Yu, et al., 2015). Previous studies suggest that in order to counteract stress, employees may use cyberloafing to relax their minds (Oravec, 2002). Intermittent breaks during hectic work enables recovery, recharging mental and physical energies (Sluiter, et al., 2003). For example, individuals who check online sports scores while answering telephone calls can relieve frustration (Oravec, 2002). A study discovered that the recreational use of computer games during work hours leads to psychological detachment from work, builds internal resources (e.g., knowledge and new skills), and increases psychological well-being (Reinecke, 2009).

On the other hand, other literature indicate that cyberloafing may result in higher job stress. First, cyberloafing often competes with work for the employees’ limited resources such as time and mental energy. Hence, employees might end up feeling stressed because they are unable to complete job-related tasks on time or bring unfinished tasks home to work (Lim and Teo, 2005). In fact, cyberloafing was found to have a positive association with mental exhaustion (Doorn, 2011). Certain types of cyberloafing activities may not be suitable as a mean of relaxation because they consume employees’ job resources such as energy, time, and concentration (Lim and Chen, 2009). When employees’ job resources are unable to match with job demands, stress will occur. To date, the influence of cyberloafing on job stress has been ambigious with limited empirical studies.

H8: Cyberloafing has a significant influence on job stress.

3.9. Cyberloafing and work performance

Work performance is defined as the extent to which employees are perfoming their assigned tasks (Kuvaas, 2006). According to Young (2001), cyberloafing may negatively affect employees’ work performance such as being unresponsive to customers’ requests, having less interactions with colleagues, and unable to meet promised deadlines. Furthermore, the frequency of making mistakes may increase because employees might lose concentration on their work through the swapping of attention between office work and personal Internet use. In a study of 71 employees from six multinational companies, cyberloafing was found to have a negative influence on work performance (Bock and Ho, 2009). Similarly, the use of online social network sites for personal purposes at work negatively affect work performance in a study of 11,018 working Norwegians (Andreassen, et al., 2014).

On the other hand, cyberloafing is not necessarily detrimental but rather innocuous if appropriately moderated (Blanchard and Henle, 2008). Frequent Internet use in workplace might even lead to higher productivity in comparison with those who use the Internet infrequently or not at all (Stanton, 2002). Results from a study of 268 working adults indicated that those who cyberloaf have higher job productivity compared to those who do not or cannot cyberloaf (Coker, 2011). The Internet can provide sources for inspiration, creativity, flexibility, and new ideas (Stanton, 2002; Anandarajan, et al., 2004; Blanchard and Henle, 2008). Furthermore, excessive restrictions on personal Internet use in the workplace can lead to employees’ dissatisfaction. The impact of cyberloafing on employees’ work performance remains inconclusive. Therefore, a non-directional hypothesis is proposed.

H9: Cyberloafing has a significant influence on work performance.

3.10. Control variables

For this research, several control variables are included into statistical analysis in order to control for the possible variance accounted for the dependent variable (cyberloafing) by unexpected factors. Gender (Stanton, 2002; Garrett and Danziger, 2008; Vitak et al., 2011) and age (Vitak, et al., 2011) are demographic factors that have significant relationships with cyberloafing, which could possibility have confounding effect for the proposed relationships in this study. As a result, this research included age and gender group as control variables.

 

++++++++++

4. Methodology

4.1. Sampling and data collection

The recruitment of respondents was done through distributing survey questionnaire to ICT employees in MSC Malaysia status companies for this study. MSC Malaysia status is a symbol of recognition by the Government of Malaysia for ICT and ICT-facilitated businesses that develop or use multimedia technologies to produce and enhance their products and services (MSC, 2015). ICT employees of MSC Malaysia status companies were selected as target respondents because the nature of their work requires high utilization of information technology. Given the access and facilities to use the Internet and the online resources (e.g., Internet connectivity, computers, laptops, and others) provided by their organizations, these employees would find it easier to cyberloaf (Vitak, et al., 2011). Furthermore, the IT industry is a critical productivity enabler and any impact to productivity of IT industry workers will be amplified through other industries to the nation’s gross domestic product (GDP). To ensure the questionnaire is well-constructed, it was pre-tested on three IT professionals and two academic experts. Slight modifications were made on the questionnaire such as font size, instructions, and wordings for clarity.

Four hundred MSC status companies were approached for data collection but only 41 companies responded positively. Participants were assured of confidentiality and were given a shopping voucher upon completion of the survey questionnaire. The initial sample of 350 respondents were vetted to ensure no substantial missing data or silver lining answers, leaving a final sample of 301 survey responses (Hair, et al., 2017). Table 1 displays the demographic characteristics of the sample.

4.2. Measures

To ensure the validity and reliability of measurement, this study adapted scales from previous studies for all the constructs and some items were modified to suit the context of the research. All questions were answered on 7-point Likert scale by participants. The adaptation of scales is summarised in Table 2. Two constructs, namely cyberloafing and perceived consequences need special attention because of the involvement of second-order constructs. Cyberloafing was modelled as type 1 second-order construct (reflective-reflective construct) (Jarvis, et al., 2003) with two lower-order constructs namely browsing (10 items) and emailing (three items). Perceived consequences was modelled as type 2 second-order construct (reflective-formative construct), with two lower-order constructs namely perceived benefits (five items) and severity of negative consequences (five items). The control variables (gender and age group) are categorical variables which need to be converted to dummy variables.

 

++++++++++

5. Data analysis

Structural equation modelling (SEM) with (Partial Least Square) PLS approach was employed using Smart PLS version 3 to assess the measurement and structural model for this research (Ringle, et al., 2015) The reasons to use PLS-based over CB-based (covariance-based) structural equation modelling are justified as follow (Hair, et al., 2017): (1) A modification of original theory was tested with the inclusion of private demands as predictor of intention; (2) The research model involves reflective, formative constructs and second-order constructs; and, (3) PLS handles non-normal data better (Lowry and Gaskin, 2014).

5.1. Common method assessment

The data for independent and dependent variables were collected from a single data source, hence there is a need to check for common method variance. In order to minimize the common method variance, researchers took additional precaution in developing the survey questionnaire by following the procedural remedies suggested by MacKenzie and Podsakoff (2012). These remedies include ensuring participants’ anonymity, giving token of appreciation, pre-testing the questionnaire carefully, and using a different scale point to measure cyberloafing. Furthermore, several approaches were employed to test for common method variance; Harman’s single-factor test was conducted by running factor analysis on all the measurement items and the results demonstrated that the first factor accounted for 18.56 percent of the variance. Hence, we can conclude that effect of common method variance is not a major issue for this research.

 

Table 1: Demographic characteristics of the sample.
GenderICT competencyInternet experienceAge groupRaceHighest education level
Male50.2%Normal47.2%<1 year3.3%<211.7%Malay55.1%Secondary2.7%
Female49.8%Intermediate19.3%1–5 years14%21–3059.1%Chinese24.9%Post-secondary3.3%
  Expert33.6%6–10 years29.2%31–4030.9%Indian12.6%Diploma19.6%
    >10 years53.5%41–507.3%Others7.3%Undergraduate61.5%
      51–601%  Postgraduate13.0%
      >600%    

 

 

Table 2: Adaptation of scales.
ConstructsSub dimensionsNumber of itemsTypes of scaleSources
Cyberloafing
(Reflective-Reflective)
Web browsing 105-point numerical scaleLim and Teo (2005)
E-mailing35-point numerical scale
Perceived consequences
(Reflective-Formative)
Perceived benefit57-point Likert scaleBetts, et al. (2014)
Severity of negative consequences57-point Likert scale
Affect4 7-point bipolar scaleMoody and Siponen (2013); Betts, et al. (2014)
Social factors6 7-point Likert scaleMoody and Siponen (2013)
Private demands57-point Likert scaleKönig and Caner de la Guardia (2014)
Facilitating conditions67-point Likert scaleBetts, et al. (2014)
Habit127-point Likert scaleMoody and Siponen (2013)
Intention37-point Likert scaleMoody and Siponen (2013)
Job stress97-point Likert scaleJamal and Baba (1992)
Work performance67-point Likert scaleKuvaas (2006)

 

5.2. Measurement model

The assessment of reflective measurement model involves internal consistency of measures, convergent validity and discriminant validity. As seen in Table 3, the values of composite reliability for all the constructs were greater than the recommended value of 0.7; ascertaining internal consistency. In assessing convergent validity, the loadings of each item on its respective constructs must be greater than 0.7. However, items with loadings between 0.4 to 0.7 need to be removed if the particular construct’s average variance extracted (AVE) does not meet the threshold value of 0.5 (Bagozzi, et al., 1991). Based on Table 3, convergent validity was achieved.

Typically, there are three ways to assess discriminant validity, namely crossing loadings criterion, Fornell-Larcker criterion, and HTMT criterion (Henseler, et al., 2016). However, cross loadings criterion was not used in this research due to the lack of sensitivity to detect discriminant validity problems (Henseler, et al., 2015). Following Fornell-Lacker criterion, the square root of each construct’s AVE should not be higher than its highest correlation with any other construct. In addition, HTMT criterion is that the HTMT ratio should not be higher than 0.85 (Kline, 2011). Based on Table 4 and Table 5, discriminant validity is not an issue.

The assessment of formative measurement models is different from a reflective measurement model with different set of criteria (Hair, et al., 2017). In this research, perceived consequences are the only formative construct. First, we checked for collinearity by examining the variance inflated factor (VIF) value to ensure items are conceptually distinct. Since the VIF values for both items are below than 5, they are not highly correlated with each other. Furthermore, we assessed the significance of outer weighs for both items. Perceived benefits (t = 16.29.924, <0.01) and Severity of negative consequences (t = 2.78, <0.01) are statistically significant for perceived consequences.

 

Table 3: Measurement model.
First order constructsSecond order constructsMeasurement itemLoading/WeightComposite reliabilityAVE
Web browsing
(WB)
 C30.6940.8820.518
C40.668
C50.64
C70.768
C80.736
C90.761
C100.759
E-mail
(EM)
 C110.9080.9270.809
C120.917
C130.872
 Cyberloafing (CL)
Reflective
Web browsing0.8570.8960.812
E-mail0.943
Perceived benefits (PB) PCA10.7450.9040.653
PCA20.81
PCA30.815
PCA40.843
PCA50.826
Severity of negative consequences (NC) PCB10.8710.9420.766
PCB20.92
PCB30.924
PCB40.886
PCB50.765
 Perceived consequences (PC)
Formative
BN0.573  
SNC0.211
Affect (AFF) Aff10.8720.9480.821
Aff20.903
Aff30.936
Aff40.913
Social factors (SF) SF10.9100.9110.720
SF20.889
SF30.888
SF40.689
Private demands (PD) P10.8330.880.648
P20.883
P30.757
P40.738
Habit (HB) Hab10.7870.9610.673
Hab20.847
Hab30.877
Hab40.832
Hab50.836
Hab60.767
Hab70.77
Hab80.835
Hab90.817
Hab100.771
Hab110.849
Hab120.847
Intention (IN) Int10.9480.9690.911
Int20.964
Int30.952
Facilitating conditions (FC) FCA10.7870.9120.635
FCA20.816
FCA30.846
FCA40.828
FCA50.701
Job stress (JS) JS10.610.9180.556
JS20.736
JS30.755
JS40.781
JS50.639
JS60.812
JS70.784
JS80.791
JS90.775
Work performance (WP) WP10.8780.9390.718
WP20.871
WP30.772
WP40.868
WP50.809
WP60.882

 

 

Table 4: Fornell-Larcker criteria.
Note: AFF=Affect; CL=Cyberloafing; EM=E-email; FC=Facilitating conditions; GD=Gender; HB=Habit; IN=Intention; JS=Job stress; NC=Severity of negative consequences; PB=Perceived benefits; PC=Perceived consequences; PD=Private demands; SF=Social factors; WB=Web browsing; WP=Work performance.
 AFFAgeCLEMFCGDHBINJSNCPBPCPDSFWBWP
AFF0.906               
Age0.016              
CL0.3960.0900.708             
EM0.3270.0770.8570.899            
FC-0.0430.0790.1860.1440.796           
GD0.0460.0070.0120.070-0.035          
HB0.4450.0640.4860.3820.117-0.1030.820         
IN0.4640.0760.4340.392-0.028-0.0570.5300.955        
JS0.1810.0550.2300.1770.0540.0240.2520.1310.746       
NC0.168-0.0900.0560.059-0.248-0.015-0.0120.158-0.0730.875      
PB0.3030.0410.2460.219-0.0420.0210.3680.3860.1210.0380.808     
PC0.308-0.0480.1860.173-0.2220.0000.2020.3480.0110.8200.603    
PD0.2130.0050.3110.2360.061-0.0270.2950.2180.289-0.0220.3790.2000.805   
SF0.3990.0580.3040.2170.0180.0040.3320.2990.321-0.0020.2660.1510.1980.849  
WB0.3820.0850.9430.6380.186-0.0280.4810.3950.2280.0460.2260.1670.3140.3150.720 
WP0.098-0.0180.0780.106-0.173-0.0240.0370.0630.1380.0950.2220.2030.1700.0980.0480.848

 

 

Table 5: HTMT results.
Note: AFF=Affect; CL=Cyberloafing; EM=E-email; FC=Facilitating conditions; GD=Gender; HB=Habit; IN=Intention; JS=Job stress; NC=Severity of negative consequences; PB=Perceived benefits; PC=Perceived consequences; PD=Private demands; SF=Social factors; WB=Web browsing; WP=Work performance.
 AFFCLFCHBINJSPCPDSFWF
AFF          
CL0.436
[0.342, 0.524]
         
FC0.083
[0.104, 0.271]
0.207
[0.135, 0.344]
        
HB0.473
[0.389, 0.560]
0.526
[0.451, 0.614]
0.137
[0.117, 0.284]
       
IN0.493
[0.409, 0.576]
0.47
[0.380, 0.555]
0.085
[0.092, 0.277]
0.554
[0.471, 0.636]
      
JS0.187
[0.102, 0.320]
0.228
[0.200, 0.358]
0.104
[0.180, 0.253]
0.254
[0.163, 0.374]
0.136
[0.079, 0.268]
     
PC0.364
[0.254, 0.453]
0.275
[0.267, 0.418]
0.284
[0.246, 0.417]
0.322
[0.298, 0.454]
0.414
[0.314, 0.492]
0.178
[0.224, 0.350]
    
PD0.229
[0.125, 0.354]
0.354
[0.251, 0.461]
0.09
[0.168, 0.309]
0.329
[0.207, 0.430]
0.231
[0.121, 0.346]
0.358
[0.272, 0.467]
0.333
[0.326, 0.503]
   
SF0.445
[0.345, 0.546]
0.35
[0.235, 0.460]
0.072
[0.166, 0.296]
0.361
[0.259, 0.456]
0.325
[0.209, 0.432]
0.363
[0.272, 0.477]
0.239
[0.237, 0.419]
0.23
[0.117, 0.360]
  
WP0.111
[0.062, 0.242]
0.112
[0.141, 0.294]
0.216
[0.132, 0.320]
0.076
[0.148, 0.226]
0.078
[0.057, 0.224]
0.158
[0.107, 0.307]
0.267
[0.180, 0.394]
0.169
[0.093, 0.294]
0.111
[0.083, 0.268]
 

 

5.3. Structural model

The examination of statistical significance was done through a non-parametric bootstrapping procedure of 5,000 iterations to generate the path coeficients and standard errors (Chin, 2010). Although PLS does not have goodness-of-fit indices to indicate how well the empirical data fit the proposed model, Henseler, et al. (2016) proposed an approximate model fit criterion by applying the standardized root mean square residual (SRMR). A value less than 0.10 suggests that the PLS path models provide an adequate fit of the empirical data (Hu and Bentler, 1999). For this study, the SRMR=0.088 indicating acceptable model fit. R2 refers to the explanatory power of a set of predictors on the respective dependent variables. For this study, the R2 for intention to cyberloafing=0.288, cyberloafing=0.309, job stress=0.053 and work performance=0.006. Following Hair, et al.’s (2017) interpretation of R2, the R2 for this model ranges from moderate to weak. Results of path analysis were shown in Figure 1. The results of structural model analysis coupled with effect size and predictive relevance of each hypothesized relationship were also presented in Table 6.

 

PLS results  Path estimates
 
Figure 1: PLS results — Path estimates.
Note: * p<0.05; ** p<0.01.

 

 

Table 6: Structural estimates (Hypotheses testing).
Note: f2: Effect size; q2: Predictive relevance.
 Direct effectConfidence interval (95%)Standard errort statsf2q2Decision
PC→INT0.211[0.004, 0.331]0.0832.5520.0550.044Supported
AFF→INT0.335[0.223, 0.454]0.0595.6710.1200.105Supported
SF→INT0.118[0.003, 0.234]0.0621.9010.0160.012Supported
PD→INT0.081[0.001, 0.219]0.0561.4450.0080.000Not supported
INT→CL0.262[0.140, 0.373]0.0574.5880.0700.028Supported
HB→CL0.334[0.206, 0.441]0.0595.6920.1120.045Supported
FC→CL0.154[0.086, 0.267]0.0473.2850.0330.012Supported
CL→JS0.230[0.183, 0.387]0.0504.5920.0700.021Supported
CL→WP0.078[0.051, 0.284]0.1090.7170.0060.001Not supported
Age→CL0.037[-0.107, 0.166]0.0790.4650.642N/AN/A
GD→CL0.066[-0.028, 0.160]0.0481.3940.164N/AN/A

 

Perceived consequences (β=0.273, t=2.552), affect (β=0.335, t=5.671) and social factors (β=0.118, t=1.901) were significant to intention to cyberloafing except for private demands (β=0.081, t=1.445). Furthermore, intention to cyberloaf (β=0.262, t=4.588), habit (β=0.334, t=5.692) and facilitating conditions (β=0.154, t=3.285) were significant to cyberloafing. The results also indicated that cyberloafing has a significant positive impact on job stress (β=0.154, t=3.285) but no impact on work performance (β=0.154, t=3.285). Lastly, control variables, age (β=0.037, t=0.465) and gender (β=0.066, t=1.394) were not significant influences on cyberloafing.

 

++++++++++

6. Discussions and implications

The findings in this research suggest that higher levels of perceived consequences result in higher levels of intention to cyberloaf. This is consistent with previous studies (Pee, et al., 2008; Betts, et al., 2014). When employees perceive cyberloafing to be rewarding, they will be motivated to engage in cyberloafing (Cheng, et al., 2014). Conversely, if employees perceive cyberloafing not to be worthwhile in view of the risks of negative consequences (imposed by employers), their intention to cyberloaf will be reduced (Moody and Siponen, 2013). Implementing punishments such as probation, suspension, reprimands, or warnings for contravening policies related to Internet usage in the workplace would be effective in stopping cyberloafing. In addition, employers may want to terminate or suspend employees who cyberloaf excessively in spite of having received several warnings or reprimands.

The result of this study is congruent with extant studies reporting that emotions play a critical role in employees’ decisions to cyberloaf (Pee, et al., 2008; Moody and Siponen, 2013; Betts, et al., 2014). Affect has the strongest effect on employees’ intention to cyberloaf among other predictors. However, it is still unclear which aspects of cyberloafing generate positive effects. For example, do employees feel enjoyable due to the process of cyberloafing or the outcomes as a result of cyberloafing (e.g., satisfied private demands from a non-work domain). As a result, it may be difficult for organizations to influence the emotions of employees indulging in cyberloafing (Betts, et al., 2014).

Consistent with previous studies (Pee, et al., 2008; Askew, et al., 2014), social factors is positively related to intention to cyberloaf. This indicates that the approval of employees’ referent groups regarding cyberloafing behavior has a strong influence on their intention to cyberloaf. As organizational norms are developed and established over time, it will not be easy to alter the climate towards cyberloafing. Top and middle level managers need to be exemplary role models by not using the Internet for personal reasons, giving an impression to subordinates that cyberloafing is acceptable (Askew, et al., 2014). In addition, organizations need to constantly communicate with employees that excessive cyberloafing is unacceptable through indirect or direct media, such as formal Internet usage policy, education, and punishment for violators.

Surprisingly, private demands are not significantly related to intentions to cyberloaf. This result contrasts with König and Caner de la Guardia (2014) that private demands drive employees to use the Internet to solve personal needs during work hours. The insignificant result signifies high levels of discipline and dedication of employees to their work, such that they would not respond to (non-urgent) private demands via cyberloafing during work hours.

Intentions to cyberloaf are significant to actual cyberloafing behavior, consistent with many extant studies and theories (Pee, et al., 2008; Moody and Siponen, 2013; Askew, et al., 2014; Betts, et al., 2014), showing that intention is an important precursor of behavior. Higher levels of intention to cyberloaf result in higher levels of actual cyberloafing behavior. This research found that habit was the most salient predictor of actual cyberloafing behavior — a finding that corroborated with prior studies (Moody and Siponen, 2013; Betts, et al., 2014). Furthermore, perceived favorable facilitating conditions were significant to actual cyberloafing behavior (Pee, et al., 2008). Hence, in order to effectively deter employees from cyberloafing, organizations need to consider implementing policies and procedures such as Internet usage policies, Internet monitoring systems, productivity measurements, and restrictions to Internet resources. Naturally, such interventions must complement each other. For instance, having an Internet usage policy is ineffective without the presence of an Internet monitoring system and the threat of punishment (Young, 2010).

This research supports the hypothesized relationship that cyberloafing results in higher job stress. Employees often spend too much of their time on non-work-related online activities at work leading to fatigue, distractions from work, and an inability to complete tasks on time. Resources (time, energy, attention) that should have been allocated for work are expended on cyberloafing, resulting in higher job stress (Bakker, et al., 2003; Lim and Teo, 2005). Despite studies suggesting cyberloafing can reduce job stress (Oravec, 2002), time spend cyberloafing should not be excessive. Furthermore, cyberloafing activities should not require high levels of cognitive effort that indirectly drain attentional energy (Lim and Teo, 2005). Employers may consider blocking Internet access to certain Web sites related gaming, investments, or shopping. Furthermore, companies can install applications that automatically inform employees after a certain time period that they are being monitored.

The findings of this research reveal that cyberloafing does not have significant impacts on work performance. One reason for this non-significant impact of cyberloafing on work performance could be multi-tasking capabilities of ICT employees because they can perform their tasks while engaging in non-work-related Internet use (Kenyon, 2008). Another possible reason is that employees could also have finished tasks before engaging in cyberloafing (Ivarsson and Larsson, 2011). The net effect on work performance of employees cyberloafing activities could be cancelled as some cyberloafing activities may increase work productivity while others may reduce it. Contrary to conventional recommendations (e.g., imposing punishment, monitoring Internet usage), we encourage employers to allow their employees to cyberloaf for two reasons. First, cyberloafing has several positive benefits to employees such as reducing stress, improving creativity, and increasing job satisfaction. Second, it is impractical and impossible for employers to monitor employees’ Internet activities all the time, if possible, it would create a sense of distrust. Furthermore, employees can easily hide their cyberloafing activities by accessing the Internet through personal digital gadgets. Nonetheless, it cannot be denied that cyberloafing entails higher security risks for an organization, congests bandwidth, and increases legal liabilities. Hence, employers may introduce some forms of Internet usage policies and procedures as discussed in Young (2010).

 

++++++++++

7. Limitations and future recommendations

This research has several limitations. Firstly, respondents may provide social desirable answers by under-reporting their cyberloafing behavior and over-reporting their work performance in spite of promises of anonymity (Bock and Ho, 2009). To obtain a more bias-free score of employees’ performance, future researchers should consider asking respondents’ immediate superiors to evaluate their work performance. Furthermore, the generalizability of these research findings may be constrained only to ICT employees in a developing country like Malaysia. Researchers might consider validating the research model with data samples from various industries and other locales. Lastly, the research design in this study was crosssectional rather than longitudinal. Hence, drawing casual inferences on the relationships is not possible. Future studies are encouraged to collect longitudinal data to confirm the causeeffectrelationship among the variables. End of article

 

About the authors

Kian Yeik Koay is a master’s student at the Faculty of Management, Multimedia University, Malaysia. His research interests are organizational behavior, service quality, and branding.
E-mail: koaydarren [at] hotmail [dot] com

Patrick Chin-Hooi Soh is a senior lecturer at the Faculty of Management, Multimedia University, Malaysia. His research interests include Internet usage, addiction, electronic commerce and business. He received his Ph.D. in Internet usage from Multimedia University and Master’s of Science degree in Information Systems at Malaysia University of Science and Technology.
E-mail: chsoh [at] mmu [dot] edu [dot] my

Kok Wai Chew is an associate professor at the Faculty of Management, Multimedia University, Malaysia. He is the ambassador for Malaysia, Academy of Management Human Resources Division’s Ambassadors Program. His research interests are management, generation issues, and ecommerce.
E-mail: kwchew [at] mmu [dot] edu [dot] my

 

Acknowledgements

This work was supported by a research grant from the Malaysian Higher Ministry of Education under the scheme of Fundamental Research Grant Scheme (FRGS) (Grant number: FRGS/2/2014/SS03/MMU/03/1).

 

Notes

1. Pee et al., 2008, p. 123.

2. Ibid.

3. Betts, et al., 2014, p. 28.

4. Triandis, 1980, p. 210.

5. König and Caner de la Guardia, 2014, p. 356.

 

References

Z. Ahmad and H. Jamaluddin, 2009. “Employees’ attitude toward cyberloafing in Malaysia,” Creating Global Economies through Innovation and Knowledge Management: Theory and Practice — Proceedings of the 12th International Business Information Management Association Conference, pp. 409–418.

I. Ajzen, 1991. “The theory of planned behavior,” Organizational Behaviour and Human Decision Processes, volume 50, number 2, pp. 179–211.
doi: http://dx.doi.org/10.1016/0749-5978(91)90020-T, accessed 16 February 2017.

M. Anandarajan, P. Devine and C. Simmers, 2004. “A multidimensional scaling approach to personal Web usage in the workplace,” In: M. Anandarajan and C. Simmers (editors). Personal Web usage in the workplace: A guide to effective human resources management. Hershey, Pa.: Information Science Publishing, pp. 28–45.
doi: http://dx.doi.org/10.4018/978-1-59140-148-3.ch004, accessed 16 February 2017.

M. Anandarajan, C. Simmers and M. Igbaria, 2000. “An exploratory investigation of the antecedents and impact of internet usage: An individual perspective,” Behaviour & Information Technology, volume 19, number 1, pp. 69–85.
doi: http://dx.doi.org/10.1080/014492900118803, accessed 16 February 2017.

C. Andreassen, T. Torsheim and S. Pallesen, 2014. “Use of online social network sites for personal purposes at work: Does it impair self-reported performance?” Comprehensive Psychology.
doi: http://dx.doi.org/10.2466/01.21.CP.3.18, accessed 16 February 2017.

K. Askew, J. Buckner, M. Taing, A. Ilie, J. Bauer and M. Coovert, 2014. “Explaining cyberloafing: The role of the theory of planned behavior,” Computers in Behavior, volume 36, pp. 510–519.
doi: http://dx.doi.org/10.1016/j.chb.2014.04.006, accessed 16 February 2017.

R. Bagozzi, Y. Yi and L. Phillips, 1991. “Assessing construct validity in organizational research,” Administrative Science Quarterly, volume 36, number 3, pp. 421–458.
doi: http://dx.doi.org/10.2307/2393203, accessed 16 February 2017.

A. Bakker, E. Demerouti, E. de Boer and W. Schaufeli, 2003. “Job demands and job resources as predictors of absence duration and frequency,” Journal of Vocational Behavior volume 62, number 2, pp. 341–356.
doi: http://dx.doi.org/10.1016/S0001-8791(02)00030-1, accessed 16 February 2017.

T. Betts, A. Setterstrom, J. Pearson and S. Totty, 2014. “Explaining cyberloafing through a theoretical integration of theory of interpersonal behavior and theory of organizational justice,” Journal of Organizational and End User Computing, volume 26, number 4, pp. 23–42.
doi: http://dx.doi.org/10.4018/joeuc.2014100102, accessed 16 February 2017.

A. Blanchard and C. Henle, 2008. “Correlates of different forms of cyberloafing: The role of norms and external locus of control,” Computers in Human Behavior, volume 24, number 3, pp. 1,067–1,084.
doi: http://dx.doi.org/10.1016/j.chb.2007.03.008, accessed 16 February 2017.

G.-W. Bock and S. Ho, 2009. “Non-work related computing (NWRC),” Communications of the ACM, volume 52, number 4, pp. 124–128.
doi: http://dx.doi.org/10.1145/1498765.1498799, accessed 16 February 2017.

British Broadcasting Corporation (BBC), 2016. “No Internet for Singapore public servants” (8 June), at http://www.bbc.com/news/world-asia-36476422, accessed 9 June 2016.

M. Chang and W. Cheung, 2001. “Determinants of the intention to use Internet/WWW at work: A confirmatory study,” Information & Management, volume 39, number 1, pp. 1–14.
doi: http://dx.doi.org/10.1016/S0378-7206(01)00075-1, accessed 16 February 2017.

L. Cheng, W. Li, Q. Zhai and R. Smyth, 2014. “Understanding personal use of the Internet at work: An integrated model of neutralization techniques and general deterrence theory,” Computers in Human Behavior, volume 38, pp. 220–228.
doi: http://dx.doi.org/10.1016/j.chb.2014.05.043, accessed 16 February 2017.

W. Cheung, M. Chang and V. Lai, 2000. “Prediction of Internet and World Wide Web usage at work: A test of an extended Triandis model,” Decision Support Systems, volume 30, number 1, pp. 83–100.
doi: http://dx.doi.org/10.1016/S0167-9236(00)00125-1, accessed 16 February 2017.

W. Chin, 2010. “How to write up and report PLS analyses,” In: V. Vinzi, W. Chin, J. Henseler and H. Wang (editors). Handbook of partial least squares: Concepts, methods and applications. Berlin: Springer-Verlag, pp. 655–690.
doi: http://dx.doi.org/10.1007/978-3-540-32827-8_29, accessed 16 February 2017.

B. Coker, 2011. “Freedom to surf: The positive effects of workplace Internet leisure browsing,” New Technology, Work, and Employment, volume 26, number 3, pp. 238–247.
doi: http://dx.doi.org/10.1111/j.1468-005X.2011.00272.x, accessed 16 February 2017.

R. Cooke and D. Rousseau, 1984. “Stress and strain from family roles and work-role expectations,” Journal of Applied Psychology, volume 69, number 2, pp. 252–260.
doi: http://dx.doi.org/10.1037/0021-9010.69.2.252, accessed 16 February 2017.

F. Davis, 1989. “Perceived usefulness, perceived ease of use, and user acceptance of information technology,” MIS Quarterly, volume 13, number 3, pp. 319–340.
doi: http://dx.doi.org/10.2307/249008, accessed 16 February 2017.

O. van Doorn, 2011. “Cyberloafing: A multi-dimensional construct placed in a theoretical framework,” Master’s thesis in innovation management, Eindhoven University of Technology, at http://www.innovatiefinwerk.nl/sites/innovatiefinwerk.nl/files/field/bijlage/cyberloafing_a_multi-dimensional_construct_placed_in_a_theoretical_framework_-_odin_van_doorn_0547224.pdf, accessed 16 February 2017.

W. Everton, P. Mastrangelo and J. Jolton, 2005. “Personality correlates of employees’ personal use of work computers,” CyberPsychology & Behavior, volume 8, number 2, pp. 143–153.
doi: http://dx.doi.org/10.1089/cpb.2005.8.143, accessed 16 February 2017.

M.-P. Gagnon, E. Sánchez and J. Pons, 2006. “From recommendation to action: Psychosocial factors influencing physician intention to use health technology assessment (HTA) recommendations,” Implementation Science, volume 1, at http://implementationscience.biomedcentral.com/articles/10.1186/1748-5908-1-8, accessed 16 February 2017.
doi: http://dx.doi.org/10.1186/1748-5908-1-8, accessed 16 February 2017.

D. Galletta and P. Polak, 2003. “An empirical investigation of antecedents of Internet abuse in the workplace,” Proceedings of the Second Annual Workshop on HCI Research in MIS, pp. 47–51, and at http://sighci.org/prototype/uploads/published_papers/ICIS2003/HCI03_06.pdf, accessed 16 February 2017.

R. Garrett and J. Danziger, 2008. “Disaffection or expected outcomes: Understanding personal Internet use during work,” Journal of Computer-Mediated Communication, volume 13, number 4, pp. 937–958.
doi: http://dx.doi.org/10.1111/j.1083-6101.2008.00425.x, accessed 16 February 2017.

J. Glassman, M. Prosch and B. Shao, 2015. “To monitor or not to monitor: Effective of a cyberloafing countermeasure,” Information & Management, volume 52, number 2, pp. 170–182.
doi: http://dx.doi.org/10.1016/j.im.2014.08.001, accessed 16 February 2017.

A. Gouveia, 2013. “2013 wasting time at work survey,” Salary.com, at http://www.salary.com/2013-wasting-time-at-work-survey/slide/2/, accessed 16 February 2017.

S. Greengard, 2000. “The high cost of cyberslacking,” Workforce, volume 79, number 12, pp. 22–24, and at http://www.workforce.com/2000/12/01/the-high-cost-of-cyberslacking/, accessed 16 February 2017.

J. Hair, G. Hult, C. Ringle and M. Sarstedt, 2017. A primer on partial least squares structural equation modeling (PLS-SEM). Second edition. Thousand Oaks, Calif.: Sage.

J. Henseler, G. Hubona and P. Ray, 2016. “Using PLS path modeling in new technology research: Updated guidelines,” Industrial Management & Data Systems, volume 116, number 1, pp. 2–20.
doi: http://dx.doi.org/10.1108/IMDS-09-2015-0382, accessed 16 February 2017.

J. Henseler, C. Ringle and M. Sarstedt, 2015. “A new criterion for assessing discriminant validity in variance-based structural equation modeling,” Journal of the Academy of Marketing Science, volume 43, number 1, pp. 115–135.
doi: http://dx.doi.org/10.1007/s11747-014-0403-8, accessed 16 February 2017.

L. Hu and P. Bentler, 1999. “Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives,” Structural Equation Modeling, volume 6, number 1, pp. 1–55.
doi: http://dx.doi.org/10.1080/10705519909540118, accessed 16 February 2017.

L. Ivarsson and P. Larsson, 2011. “Personal Internet usage at work: A source of recovery,” Journal of Workplace Rights, volume 16, number 1, pp. 63–81.
doi: http://dx.doi.org/10.2190/WR.16.1.e, accessed 16 February 2017.

M. Jamal and V. Baba, 1992. “Shiftwork and department-type related to job stress, work attitudes and behavioral intentions: A study of nurses,” Journal of Organizational Behavior, volume 13, number 5, pp. 449–464.
doi: http://dx.doi.org/10.1002/job.4030130503, accessed 16 February 2017.

C. Jarvis, S. MacKenzie and P. Podsakoff, 2003. “A critical review of construct indicators and measurement model misspecification in marketing and consumer research,” Journal of Consumer Research, volume 30, number 2, pp. 199–218.
doi: https://doi.org/10.1086/376806, accessed 16 February 2017.

H. Jia, R. Jia and S. Karau, 2013. “Cyberloafing and personality: The impact of the big five traits and workplace situational factors,” Journal of Leadership & Organizational Studies, volume 20, number 3, pp. 358–365.
doi: https://doi.org/10.1177/1548051813488208, accessed 16 February 2017.

P. Johnson and J. Indvik, 2004. “The organizational benefits of reducing cyberslacking in the workplace,” Journal of Organizational Culture, Communications and Conflict, volume 8, number 2, pp. 53–59.

S. Kenyon, 2008. “Internet use and time use: The importance of multitasking,” Time & Society, volume 17, numbers 2–3, pp. 283–318.
doi: https://doi.org/10.1177/0961463X08093426, accessed 16 February 2017.

R. Kline, 2011. Principles and practice of structural equation modeling. Third edition. New York: Guilford Press.

C. König and M. Caner de la Guardia, 2014. “Exploring the positive side of personal Internet use at work: Does it help in managing the border between work and nonwork?” Computers in Human Behavior, volume 30, pp. 355–360.
doi: http://dx.doi.org/10.1016/j.chb.2013.09.021, accessed 16 February 2017.

B. Kuvaas, 2006. “Work performance, affective commitment, and work motivation: The roles of pay administration and pay level,” Journal of Organizational Behavior, volume 27, number 3, pp. 365–385.
doi: https://doi.org/10.1002/job.377, accessed 16 February 2017.

R. LaRose, 2010. “The problem of media habits,” Communication Theory, volume 20, number 2, pp. 194–222.
doi: http://dx.doi.org/10.1111/j.1468-2885.2010.01360.x, accessed 16 February 2017.

V. Lim, 2002. “The IT way of loafing on the job: Cyberloafing, neutralizing and organizational justice,” Journal of Organizational Behavior, volume 23, number 5, pp. 675–694.
doi: http://dx.doi.org/10.1002/job.161, accessed 16 February 2017.

V. Lim and D. Chen, 2009. “Browsing and emailing: Impact of cyberloafing on work attitudes,” Proceedings of 23rd Australia and New Zealand Academy of Management, at http://www.anzam.org/wp-content/uploads/pdf-manager/1022_ANZAM2009-182.PDF, accessed 16 February 2017.

V. Lim and T. Teo, 2005. “Prevalence, perceived seriousness, justification and regulation of cyberloafing in Singapore: An exploratory study,” Information & Management, volume 42, number 8, pp. 1,081–1,093.
doi: http://dx.doi.org/10.1016/j.im.2004.12.002, accessed 16 February 2017.

V. Lim, T. Teo and G. Loo, 2002. “How do I loaf here? Let me count the ways,” Communications of the ACM, volume 45, number 1, pp. 66–70.
doi: http://dx.doi.org/10.1145/502269.502300, accessed 16 February 2017.

P. Lowry and J. Gaskin, 2014. “Partial least squares (PLS) structural equation modeling (SEM) for building and testing behavioral causal theory: When to choose it and how to use it,” IEEE Transactions on Professional Communication, volume 57, number 2, pp. 123–146.
doi: http://dx.doi.org/10.1109/TPC.2014.2312452, accessed 16 February 2017.

C. Machado, J. Machado and M. Sousa, 2014. “Human resource management and the Internet: Challenge and/or threat to workplace productivity?” In: C. Machado and J. Davim (editors). Human resource management and technological challenges. Cham, Switzerland: Springer International, pp. 149–168.
doi: http://dx.doi.org/10.1007/978-3-319-02618-3_8, accessed 16 February 2017.

S. MacKenzie and P. Podsakoff, 2012. “Common method bias in marketing: Causes, mechanisms, and procedural remedies,” Journal of Retailing, volume 88, number 4, pp. 542–555.
doi: http://dx.doi.org/10.1016/j.jretai.2012.08.001, accessed 16 February 2017.

L. Martin, M. Brock, M. Buckley and D. Ketchen, Jr., 2010. “Time banditry: Examining the purloining of time in organizations,” Human Resource Management Review, volume 20, pp. 26–34.
doi: http://dx.doi.org/10.1016/j.hrmr.2009.03.013, accessed 16 February 2017.

R. Milhausen, M. Reece and B. Perera, 2006. “A theory-based approach to understanding sexual behavior at Mardi Gras,” Journal of Sex Research, volume 43, number 2, pp. 97–106.
doi: http://dx.doi.org/10.1080/00224490609552304, accessed 16 February 2017.

G. Moody and M. Siponen, 2013. “Using the theory of interpersonal behavior to explain non-work-related personal use of the Internet at work,” Information & Management, volume 50, number 6, pp. 322–335.
doi: http://dx.doi.org/10.1016/j.im.2013.04.005, accessed 16 February 2017.

MSC, 2015. “What is MSC Malaysia?” at http://www.mscmalaysia.my/what_is_msc_malaysia, accessed 20 August 2015.

J. Oravec, 2002. “Constructive approaches to Internet recreation in the workplace,” Communications of the ACM, volume 45, number 1, pp. 60–63.
doi: http://dx.doi.org/10.1145/502269.502298, accessed 16 February 2017.

J. Ouellette and W. Wood, 1998. “Habit and intention in everyday life: The multiple processes by which past behavior predicts future behavior,” Psychological Bulletin, volume 124, number 1, pp. 54–74.
doi: http://dx.doi.org/10.1037/0033-2909.124.1.54, accessed 16 February 2017.

M. Parikh and R. Gupta, 2010. Organizational behaviour. New Delhi: Tata McGraw Hill Education.

L. Pee, I. Woon and A. Kankanhalli, 2008. “Explaining non-work-related computing in the workplace: A comparison of alternative models,” Information & Management, volume 45, number 2, pp. 120–130.
doi: http://dx.doi.org/10.1016/j.im.2008.01.004, accessed 16 February 2017.

T. Ramayah, 2010. “Personal Web usage and work inefficiency,” Business Strategy Series, volume 11, number 5, pp. 295–301.
doi: http://dx.doi.org/10.1108/17515631011080704, accessed 16 February 2017.

L. Reinecke, 2009. “Games at work: The recreational use of computer games during working hours,” CyberPsychology & Behavior, volume 12, number 4, pp. 461–465.
doi: http://dx.doi.org/10.1089/cpb.2009.0010, accessed 16 February 2017.

S. Restubog, P. Garcia, L. Toledano, L. Tolentino, R. Amarnani, L. Tolentino and R. Tang, 2011. “Yielding to (cyber)-temptation: Exploring the buffering role of self-control in the relationship between organizational justice and cyberloafing behaviour in the workplace,” Journal of Research in Personality, volume 45, number 2, pp. 247–251.
doi: http://dx.doi.org/10.1016/j.jrp.2011.01.006, accessed 16 February 2017.

C. Ringle, S. Wende and J. Becker, 2015. SmartPLS 3. Bönningstedt, Germany: SmartPLS GmbH, at http://www.smartpls.com, accessed 9 August 2016.

S. Robinson and R. Bennett, 1995. “A typology of deviant workplace behaviors: A multidimensional scaling study,” Academy of Management Journal, volume 38, number 2, pp. 555–572.
doi: http://dx.doi.org/10.2307/256693, accessed 16 February 2017.

W. Schneider and R. Shiffrin, 1977. “Controlled and automatic human information processing. I. detection, search, and attention,” Psychological Review, volume 84, number 1, pp. 1–66.
doi: http://dx.doi.org/10.1037/0033-295X.84.1.1, accessed 16 February 2017.

L. Seymour and K. Nadasen, 2007. “Web access for IT staff: A developing world perspective on Web abuse,” Electronic Library, volume 25, number 5, pp. 543–557.
doi: http://dx.doi.org/10.1108/02640470710829532, accessed 16 February 2017.

B. Sheppard, J. Hartwick and P. Warshaw, 1988. “The theory of reasoned action: A meta-analysis of past research with recommendations for modifications and future research,” Journal of Consumer Research, volume 15, number 3, pp. 325–343.

J. Sipior and B. Ward, 2002. “A strategic response to the broad spectrum of internet abuse,” Information Systems Management, volume 19, number 4, pp. 71–79.
doi: http://dx.doi.org/10.1201/1078/43202.19.4.20020901/38837.9, accessed 16 February 2017.

J. Sluiter, E. de Croon, T. Meijman and M. Frings-Dresen, 2003. “Need for recovery from work related fatigue and its role in the development and prediction of subjective health complaints,” Occupational & Environmental Medicine, volume 60, number i, pp. 62–70.
doi: http://dx.doi.org/10.1136/oem.60.suppl_1.i62, accessed 16 February 2017.

M. Solomon, 2013. Consumer behavior: Buying, having, and being. Tenth edition. Harlow, England: Pearson Education.

StaffMonitoring.com, 2015. “Office slacker stats,” at http://www.staffmonitoring.com/P32/stats.htm, accessed 16 February 2017.

J. Stanton, 2002. “Company profile of the frequent Internet user,” Communications of the ACM, volume 45, number 1, pp. 55–59.
doi: http://dx.doi.org/10.1145/502269.502297, accessed 16 February 2017.

R. Thompson, C. Higgins and J Howell, 1991. “Personal computing: Toward a conceptual model of utilization,” MIS Quarterly, volume 15, number 1, pp. 125–143.
doi: http://dx.doi.org/10.2307/249443, accessed 16 February 2017.

W. Thompson and J. Hickey, 2005. Society in focus: An introduction to sociology. Fifth edition. Boston: Pearson.

H. Triandis, 1980. “Values, attitudes, and interpersonal behavior,” Nebraska Symposium on Motivation, volume 27, pp. 195–259.

H. Triandis, 1977. Interpersonal behavior. Monterey, Calif.: Brooks/Cole.

J. Ugrin and M. Pearson, 2013. “The effects of sanctions and stigmas on cyberloafing,” Computers in Human Behavior, volume 29, number 3, pp. 812–820.
doi: http://dx.doi.org/10.1016/j.chb.2012.11.005, accessed 16 February 2017.

J. Ugrin and J. Pearson, 2008. “Exploring Internet abuse in the workplace: How can we maximize deterrence efforts?” Review of Business Journal, volume 28, number 2, pp. 29–40.

B. Verplanken and W. Wood, 2006. “Interventions to break and create consumer habits,” Journal of Public Policy & Marketing, volume 25, number 1, pp. 90–103.
doi: http://dx.doi.org/10.1509/jppm.25.1.90, accessed 16 February 2017.

B. Verplanken, H. Aarts and A. Van Knippenberg, 1997. “Habit, information acquisition, and the process of making travel mode choices,” European Journal of Social Psychology, volume 27, number 5, pp. 539–560.

J. Vitak, J. Crouse and R. LaRose, 2011. “Personal Internet use at work: Understanding cyberslacking,” Computers in Human Behavior, volume 27, number 5, pp. 1,751–1,759.
doi: http://dx.doi.org/10.1016/j.chb.2011.03.002, accessed 16 February 2017.

T. Weatherbee, 2010. “Counterproductive use of technology at work: Information & communications technologies and cyberdeviancy,” Human Resource Management Review, volume 20, number 1, pp. 35–44.
doi: http://dx.doi.org/10.1016/j.hrmr.2009.03.012, accessed 16 February 2017.

I. Woon and L. Pee, 2004. “Behavioural factors affecting Internet abuse in the workplace — An empirical investigation,” Proceedings of the Third Annual Workshop on HCI Research in MIS, pp. 80–84, and at http://sighci.org/uploads/published_papers/ICIS2004/SIGHCI_2004_Proceedings_paper_13.pdf, accessed 16 February 2017.

K. Young, 2010. “Policies and procedures to manage employee Internet abuse,” Computer in Human Behavior, volume 26, number 6, pp. 1,467–1,471.
doi: https://doi.org/10.1016/j.chb.2010.04.025, accessed 16 February 2017.

K. Young, 2001. “Managing employee Internet abuse: A comprehensive plan to increase your productivity and reduce liability,” at http://www.netaddiction.com/articles/business.pdf, accessed 16 February 2017.

X. Yu, P. Wang, X. Zhai, H. Dai and Q. Yang, 2015. “The effect of work stress on job burnout among teachers: The mediating role of self-efficacy,” Social Indicators Research, volume 122, number 3, pp. 701–708.
doi: https://doi.org/10.1007/s11205-014-0716-5, accessed 16 February 2017.

 


Editorial history

Received 3 January 2017; accepted 17 February 2017.


Public Domain Mark
This paper (“Antecedents and consequences of cyberloafing: Evidence from the Malaysian ICT industry”, by Kian Yeik Koay, Patrick Chin-Hooi Soh, and Kok Wai Chew) is free of known copyright restrictions.

Antecedents and consequences of cyberloafing: Evidence from the Malaysian ICT industry
by Kian Yeik Koay, Patrick Chin-Hooi Soh, and Kok Wai Chew.
First Monday, Volume 22, Number 3 - 6 March 2017
http://www.firstmonday.dk/ojs/index.php/fm/article/view/7302/5968
doi: http://dx.doi.org/10.5210/fm.v22i13.7302





A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2017. ISSN 1396-0466.