Differences of Cyberloafing Behavior Outcomes on Men and Women Employees

: The activities of reading news, chatting, viewing YouTube and Facebook or even status updates, Instagram, buying and selling online, to playing online games that are not related to work by using internet facilities are cyberloafing behaviors that are often carried out by employees. The limitations of previous research examining the impact of cyberloafing on work behavior empirically are still very limited. In addition, previous research that explains the occurrence of cyberloafing behavior also shows results that have not been established. Encouraged by these findings, this study aimed to examine the factors that explain cyberloafing behavior and its impact on employees' organizational behavior. To meet these objectives, an empirical model was developed with job characteristics and self-control variables as exogenous variables, job stress and cyberloafing as mediating variables and laziness as endogenous variables. Testing the influence between these variables was carried out with an analytical approach to Structural Equation Modeling (SEM) which used empirical data obtained through questionnaires as an interview guide to employee respondents in various fields of work. The results of data analysis showed that job characteristics, self control, and job stress were statistically proven to have an effect on cyberloafing behavior. Cyberloafing testing on negative organizational behavior, namely laziness also shows a real influence. In addition, the results of this study also show that there are differences in prevalence caused by cyberloafing behavior between male and female employees.


Introduction
Surfing in cyberspace such as reading news, chatting, watching YouTube and Facebook or even updating status, Instagram, online buying and selling, and playing online games is an activity that has absolutely nothing to do with work in the office but is often carried out by employees on the sidelines of office hours. It can be done by using private internet facilities or using WiFi provided by the company. This behavior is called cyberloafing.
Cyberloafing is a phenomenon of new organizational behavior along with the digitization that goes into all sides of life including at work. Cyberloafing behavior can have both positive and negative implications. The positive implication of cyberloafing behavior according to Lynn, Coker & Coker (2011) is if it is done no more than 12% of all work time. Positive implications of cyberloafing behavior are increasing awareness of developing information (Seymour & Nadasen, 2007;Lynn et al, 2011), increasing employee creativity, increasing employee well-being, recreation and recovery solutions for employees (Malhotra, 2013). Balancing life work and personal employees, reducing work stress and workload, and making life more interesting (Anandarajan, Paravastu & Simmers, 2006;Lim & Chen, 2012). Some studies also showed that cyberloafing behavior also leads to counterproductive behavior (Lim, 2002;Lim & Teo, 2005). As productivity declines (Griffiths, 2010;Weatherbee, 2010;Liberman, Seidman, McKenna, & Buffardi, 2011;Malhotra, 2013), increases in corporate finances, indisciplinary behavior, threats to corporate information security (Lim, 2002;Lim & Teo, 2005;Ozler & Polat, 2012;Malhotra, 2013).
The emergence of cyberloafing behavior can be triggered by individual employee factors (Liberman, Seidman & McKenna, 2011;Ozler & Polat, 2012;Malhotra, 2013;Abidin et al, 2014) and organizational factors (Ozler & Polat, 2012;Al-Shuaibi, & Shamsudin, 2013;Malhotra, 2013). One individual factor that can explain cyberloafing behavior is self-control (Ozler, & Polat, 2012). Self-control is related to the ability of employees to hold desires that are contrary to the norms prevailing in the company. Empirically, the effect of self-control on cyberloafing behavior has been investigated by Ramadan & Sari (2018). The results of the study showed a significant positive relationship between self-control variables on cyberloafing behavior. Still on testing the same variable, a study conducted by Sari & Ratnaningsih (2018) showed a significant negative effect of self-control on cyberloafing behavior while organizational factors focus on the work characteristics variable from the study of Malhotra (2013) and Ozler & Polat (2012) which have not been empirically tested in this study.
Although there have been many studies related to cyberloafing behavior, empirical studies are still limited to determinants of cyberloafing behavior. Studies that map aspects of cyberloafing behavior and empirical testing have not been conducted. Referring to these findings, this study aims to develop a comprehensive model of cyberloafing behavior by testing the determinants and outcomes of cyberloafing behavior and mapping the preferences of cyberloafing behavior according to gender.

References and Research Models
Cyberloafing Behavior Lim (2002) conceptualized cyberloafing behavior is as an activity carried out by employees to intentionally utilize internet access provided by the company to carry out actions that do not relate to work during working hours. By Blanchard & Henle (2008), cyberloafing behavior is categorized into two groups, namely minor cyberloafing and serious cyberloafing. Minor cyberloafing behavior includes using the internet and accessing email that is done during working hours such as receiving and sending private messages, visiting news sites, sports, and finance while serious cyberloafing behavior includes gambling online, downloading songs, or just opening or even watching online movie sites.
Cyberloafing Based on Gender Lim and Chen (2009) stated that there are differences in internet usage preferences according to gender. There are indications of the frequency, intensity, and nature of internet use by gender. As stated by Rahman & Abdul-Gader (1993), Anandarajan, Simmers, & Igbaria (2000), Ono & Zavodny (2003), Colley & Maltby (2008), Garrett & Danziger (2008) that the emergence of differences in preferences is caused by differences in preferences in the purpose of using the internet.

Determinants of Cyberloafing
Determinants of cyberloafing behavior by Ozler & Polat (2012) and Malhotra (2013) are categorized in individual and organizational factors. Individual factors includes relating to individual perceptions and attitudes towards internet use, demographic characteristics, personality, self-control, selfefficacy, locus of control. Organizational factors include restrictions on internet use, job characteristics, manager support, and applicable norms.

Work Stress
Work stress by Salleh, Bakar and Keong (2008) is interpreted as a pressure, strength or a tendency or a mental effort made by someone against their work. According to Robbins and Judge (2008), there are three categories of symptoms that appear to a person who is experiencing work stress, namely physical function disorder, psychological function disorder, and behavior disorder.
A study conducted by Kusumawati & Fransiska (2018) found that work stress proved to have a positive but not significant effect in explaining cyberloafing behavior. However, the effect of work stress on cyberloafing behavior in a study conducted by Koay, Soh & Chew (2017) showed significant positive results.

Self-Control
Self-control according to Baumeister (2002) is conceptualized as the ability possessed by individuals to direct and regulate the feelings, thoughts and behavior needed in order to adapt to the environment or to meet certain needs or other temptations. Selfcontrol is a determinant of cyberloafing behavior that comes from individuals (Ozler & Polat, 2012). Good self-control is thought to be needed to minimize cyberloafing behavior by employees. With good self-control, employees are able to restrain themselves from acting that is not in accordance with the norms at work which in this case is an act of cyberloafing.

Job Characteristics
Ozler & Polat (2012) in their study stated that job characteristics are an explanation of cyberloafing behavior at the organizational level. Arshad & Bukhari (2016) in their study measured job characteristics in five aspects, namely skill variety, task identity, task significance, autonomy and feedback on cyberloafing behavior. The results of the study showed that variety and autonomy skills had a significant negative effect on cyberloafing behavior, task identity had no a significant positive effect on cyberloafing behavior, task significance and feedback had no significant negative effect on cyberloafing behavior.

Cyberloafing Outcomes
The orientation of cyberloafing can be categorized in the following four groups: (1) Development behavior. This behavior assumes that cyberloafing behavior as a potential source for learning. Cyberloafing behavior from this perspective is considered to be able to improve skills that are useful for future employee activities that are able to provide benefits for individuals and organizations (Belanger & Slyke, 2002); (2) Recovery behavior. This behavior assumes that employee cyberloafing behavior can reduce discomfort and provide positive effects for employees and organizations (McLean, Tingley, Scott & Richards, 2001;Lim & Chen, 2009); (3) Deviant behavior. This behavior assumes that cyberloafing behavior is a distortion of unwanted behavior and is directed at the organization; (4) Addiction behavior. Cyberloafing behavior in this context is considered as a habit that leads to the emergence of problematic behavior.

Effect of Job Characteristics on Job Stress and Cyberloafing
Job characteristics are related to the characteristics of the work carried out by employees in the company which includes skill variety/job (skill variety), job identity (task identity), job significance (task significance), autonomy (autonomy) and feedback (feedback). If employees experience skill variety, task identity, task significance, autonomy and feedback, employees experience job stress and cyberloafing behavior.
Based on the description above, the following hypothesis was developed:

The Effect of Self Control on Cyberloafing
Self control is related to the individual ability of employees to control or control their behavior within the organization. Studies on the effect of self-control on cyberloafing behavior have been studied by several previous researchers. A study conducted by Sari & Ratnaningsih (2018) shows that selfcontrol is able to explain the variation in cyberloafing behavior negatively. However, the study conducted by Ramadhan & Sari (2018) shows a different direction where selfcontrol explains cyberloafing behavior positively.
Based on the description above, the following hypothesis was developed: H4: Self control has a negative effect on cyberloafing

The Effect of Cyberloafing on Laziness
In this behavior, cyberloafing has negative consequences for the organization (such as decreased productivity) (Weatherbee, 2010;Young, 2010). Yellowlees & Marks (2007) revealed that severe internet addiction will lead to problems in work behavior. More specifically, Stanton (2002) and Yellowlees & Marks (2007) show that the consequences of Internet addiction can lead to decreased performance.
Based on the description above, the following hypothesis was developed: H5: Cyberloafing has a positive effect on laziness

Differences in Cyberloafing Behavior Models Affect Laziness on Female and Male Employees
The study of Kim and Davis (2009) showed a preference for the use of the internet by women for the reasons of seeking information, establishing relationships and for shopping purposes. While the use of the internet in men for reasons to be able to have relationships with people from all over the world and to get a job. A study conducted by Garrett & Danziger (2008) found differences in the use of the internet for personal reasons, pleasure and communication based on gender.
The comparative study on cyberloafing behavior preferences conducted by Lim & Chen (2012) showed different results. In his study, men tend to do more cyberloafing than women. Although some literature has tested differences in internet use preferences by gender, these results are not conclusive.
Women assume that using the internet will get a wide social network. Meanwhile, men use the internet for relaxation reasons or to show their strength (Colley & Maltby, 2008;Garrett & Danziger, 2008).
Based on the description above, the following hypothesis was developed: H6: The cyberloafing behavior model affects laziness in male employees differently than female employees

Empirical Model
Effect between research variables was visualized in an empirical model developed and tested in this study.

Population and Sample
The population studied in this study were company employees from several industries. The industries studied in this study were: automotive, education, banking, garment and pharmaceutical.
Based on the results of data collection, a sample of 228 respondents was obtained with the distribution of the number of samples used in this study as follows:

Data collection
Data was collected through interviews using interview guides that had been prepared. The interview guide contains statement items which are measurements of the variables studied. Respondents were asked to answer

Operationalization of Research Variables
Operationalization of variables is needed for the purpose of measuring research variables. Operationalization of variables was done by setting indicators adopted from the results of previous relevant studies.

Analysis Techniques
For the sake of testing the research model and testing the effect between the research variables, this study used Structural Equation Modeling (SEM) as a data analysis tool.

Result and Discussion
There were two analysis processes carried out in this study, namely research model testing and comparative testing. The process of the two tests is described below:

Research Model Testing
Empirical models and the effect between variables developed in this study used Structural Equation Modeling (SEM) as an analytical technique approach. In SEM there are two stages of testing carried out. These stages are described below.

Confirmatory Analysis
Confirmatory analysis is the stage carried out to confirm the accuracy of the indicators in measuring the research variables. There are two outputs used as a reference in measuring the accuracy of indicators, namely by analyzing the value of Standardized Regression Weight which is presented in the following table: Referring to the results of the confirmatory analysis presented in the Table   above, the indicators that produce a Standardized Regression Weight below 0.5 are declared null because they cannot statistically reflect measurements on the research variable.

Hypothesis Testing
At this stage, there are two tests conducted, namely testing the serviceability of the research model and testing the research hypothesis.
The following is an evaluation of the empirical model testing developed in this study.
Source: Primary Data Processed, 2020 The empirical model testing developed in this study produced a Chi Square value of 262,043 and a probability value of 0,230. The Chi Square value is smaller than the Chi Square table value (283,585) with a probability value greater than 0,05 so that it can be concluded that the research model is the right model.
After testing the feasibility of the model, it can be tested for the effect of the variables presented in the Table 6.
Testing the effect between variables was done by analyzing the probability value and the value of the Critical Ratio (CR). The probability value in job characteristics and work stress testing was 0,008 < 0,05 and CR was 2,674, meaning that the job characteristics Probability value in work stress and cyberloafing testing was 0,005 < 0,05 and CR was 2,793, meaning that work stress statistically had a positive significant effect in explaining cyberloafing.
Probability value in self-control and cyberloafing testing was 0,021 < 0.05 and CR was -2,309, meaning that self-control statistically had a negative significant c.effect in explaining cyberloafing.
Probability value in job characteristics and cyberloafing testing was 0,969 > 0,05 and CR was -0,039, meaning that job characteristics statistically had a negative non significant effect in explaining cyberloafing.
Probability value in cyberloafing and laziness testing was 0,044 < 0,05 and CR was 2,013, meaning that cyberloafing had a statistically a positive effect in explaining laziness.

Comparative Testing
Comparative testing is a test carried out to determine differences in research models and the effect of variables based on gender.  Table 7 showed that the Chi Square value for each group was smaller than the critical Chi Square value as well as a significance value greater than 0,05. These results indicated that there was no difference between the sample covariance matrixes with the estimated population covariance matrix or in other words the model was accepted or fit.
This subsection provides an explanation of the effect between variables based on each gender group.
The results of testing the effect between variables based on gender groups showed that in men, work stress had a significant positive effect on cyberloafing. It means the higher work stress experienced by male employees will increase cyberloafing. Male employees become happy to linger surfing in cyberspace during work hours. But different in the findings produced by female employees, stress did not actually cause cyberloafing. The results of testing the effect on women's groups indicated that the increase in cyberloafing was caused by the increasing characteristics of work. This test also showed that cyberloafing by female employees can in fact increase creativity.

Conclusion
Empirical phenomena indicated that cyberloafing conducted by employees during working hours by using internet facilities provided by the company is getting higher. Cyberloafing carried out was indicated to have no relevance to the work process. Some previous studies mention many factors that trigger the increasing cyberloafing with the results of studies that are not yet conclusive. In addition, previous research also has not conducted empirical testing of the effects arising from cyberloafing. From these findings, this study developed a comprehensive research model by examining variables at the organizational level, namely the job characteristics and variables at the individual level, namely self-control as a cyberloafing explanation. This study also seeks to empirically examine cyberloafing outcomes both positively and negatively.
The results of overall model tests showed that there was a real effect of job characteristics, work stress and self-control in explaining cyberloafing. Furthermore, this study also found that cyberloafing could trigger creativity and laziness in employees. However, comparative testing based on gender showed that in male employees, cyberloafing was only explained by work stress while in female employees it was explained by job characteristics. Outcome from cyberloafing was only obtained for female employees where in fact cyberloafing could increase the creativity of female employees.   (2019)