The Influence of Hotel Employee Job Stress towards Cyberloafing

The purpose of this research is to determine the influence of employee’s job stress towards cyberloafing at Four Seasons Hotel Jakarta. Primary data was collected through survey using questionnaire with five-point Likert scale and close-ended question. Sample of this research was 36 employees at Four Seasons Hotel Jakarta in back of the house departments. Data was analyzed with descriptive statistics, correlation, regression analysis, and three hypothesis testing using SPSS. The results indicate that job stress has significantly influenced on cyberloafing with a value of 11%. Job stress and cyberloafing have low correlation, but positive and significant. Furthermore, job ambiguity showed significant influence with cyberloafing, while job conflict showed the contrary of insignificant. It showed that employees tend to do cyberloafing when they felt uncertainty or ambiguity related to their job compared to the conflict that they encountered. From this findings, hospitality companies are expected to always assure that their employees perceived certainty related to their job tasks and job description to prevent any upcoming job stress.


INTRODUCTION
The growing demand for human resources in the hospitality sector has an impact on high levels of employee turnover, whether it is unavoidable or avoidable turnover (Hafid & Prasetio, 2017;Mohsin, Lengler, & Aguzzoli, 2015). The unavoidable turnover's causes are mostly related to personal or family reasons such as retirement, illness, death, pregnancy, and permanently changing domicile (Lestari et al., 2021). On the other hand, avoidable turnover that is usually caused by work stress or job stress can be minimized from within the company (Mohsin, Lengler, & Aguzzoli, 2015).
Job stress is a condition that negatively affects employee performance and has an impact on our physical and mental health (Kotteeswari & Sharief, 2014), while Yuda & Ardana (2017) called it as work stress. Job stress can be caused by work or work-related factors (Yongkang, Weixi, Yalin, Yipeng, & Liu, 2014). It is known that as many as 41% of employees feel that the cause of work stress is unclear job goals, poor cooperation between co-workers, and excessive working hours (Setiawan, 2019). Herdiati, Sujoso, & Hartanti (2015) said that the high number of job stress that occurs has an impact on employees who tend to cyberloaf at work or doing cyberloafing activities. The relationship between job stress and cyberloafing is based on previous researches conducted by Karim Herdiati, Sujoso, & Hartanti (2015), showed a positive and significant influence. However, less study found in the context of hotel industry in Indonesia.
Cyberloafing is considered as one of the negative influences due to internet use (Ramadhan & Nurtjahjanti, 2017) and can be caused by three factors: individual factors, situational factors, and organizational factors (Ardilasari & Firmanto, 2017). In addition, Hardiani, Rahardja, & Yuniawan (2017) argue that cyberloafing activities will be seen when conflicts arise between co-workers and when employees feel that they do not know the role they take. Triana in Kerwin (2015) said that employees spend 2.5 hours for cyberloafing per day which does not only reduce productivity but can cause information system and data security problems. In other words, employees prefer to do cyberloafing to avoid role stressors, when employees feel unclear about the obligations and tasks assigned or when they are not aware of the procedures needed to complete the work (Arshad, Aftab, & Bukhari, 2016). Hotel employees should be well managed as it can become a weakness as well as strength point for hotel in terms of providing excellent service for customer (Choirisa & Armeilia, 2019).
There are three objectives of this research: (1) to determine the employee job stress, (2) to determine the employee cyberloafing, and (3) to determine the influence of employee job stress towards cyberloafing.

LITERATURE REVIEW Job Stress
Job stress or usually known as occupational stress, defined as a form of stress caused by work or work-related factors (Yongkang, Weixi, Yalin, Yipeng, & Liu, 2014). Job stress can cause a negative influence on a person's performance and has an impact on the mind and overall body health (Kotteeswari & Sharief, 2014). Job stress is also a condition of a person in dealing with demands at work that exceed his ability to cope. In the context of hotel industry as a part of service industry, job stress is strongly related to hotel work culture that involves long working hours, shift work, and irregular occupancy (Hotel Tech Report, 2022). In addition, refering to the study done by O'Neill & Davis, hotel employees feel stress on 40 -62% of workdays, which is higher than the average employee across industries (Ko, 2020). In the same vein, study of Lestari et al. (2021) explores on how family-work conflicts can positively affect stress for hotel Front Office employees.
Job stress itself can be defined from its role ambiguity and role conflict (Bowling, et al., 2017). Role ambiguity is a condition of a person who is faced with an unclear work situations (Bowling, et al., 2017), unclear information about the goals related to the role, the way to fulfill the role of goals or the results of the role performance (Irzani & Witjaksono, 2014). Role ambiguity can hinder one's work and cause feelings of confusion and uncertainty (Blanchard & Henle, in Karim, Parawansa, & Jusni, 2019). Meanwhile, role conflict is a condition of a person experiencing job demands that are inappropriate or contradictory to the role being carried out (Bowling, et al., 2017). Role conflict is defined as a condition that occurs as a result of a conflict between the wishes of a person or group due to differences in the roles they carry out (Irzani & Witjaksono, 2014). Role conflict occurs when someone is faced with role expectations that conflict with various social statuses in their position (Yongkang, Weixi, Yalin, Yipeng, & Liu, 2014). Prior study showed that role ambiguity and role conflict as a measured of job stress has a significant impact on cyberloafing (Arshad, Aftab, dan Bukhari, 2016: Koay, Chin-Hooi, & Chew,2017: Gokcearslan, Uluyol, & Sahin,2018.

Cyberloafing
Cyberloafing is defined as the use of the internet by employees to pass the time and occupy themselves during working hours (Utama, et al., 2016), for personal purposes and not related to work (Herdiati, Sujoso, & Hartanti, 2015;(Lim, in Gokcearslan, Uluyol, & Sahin, 2018). Ozler & Polat in Ardilasari & Firmanto (2017) explain three factors that cause cyberloafing: individual factors, situational factors, and organizational factors. Individual factors related to employee characteristics such as employee perceptions and attitudes of using the internet personally, personal traits of employees, habits of employees using the internet, demographic factors, social norms, personal code of ethics. Situational factors usually cause employees to cyberloaf because of conditions that trigger employees to access the internet. Organizational factors influence the tendency of employees to cyberloaf because of restrictions on internet use, consequences applied to employees if cyberloafing occurs, job characteristics and forms of managerial support to inform employees of internet use at work.
Moreever, cyberloafing can be measured from 12 indicators from Lim & Teo dalam Pindek, Krajcevska, & Spector (2018). Prior study showed that job stress that consist of role ambiguity and role conflict has a significant impact on cyberloafing (Karim, Parawansa, & Jusni, 2019;Hardiani, Rahardja, dan Yuniawan, 2017;Herdiati, Sujoso, dan Hartanti, 2015). Hence, this study attempted to test three hypotheses as follows: H1: Role ambiguity has a significant effect on employee cyberloafing H2: Role conflict has a significant effect on employee cyberloafing H3: Job stress that consist of role ambiguity and role conflict has a significant effect on employee cyberloafing

RESEARCH METHOD
This research is categorized as quantitative research that examines a particular population or sample and tests hypotheses based on statistical data to show the relationship between variables (Sugiyono, 2016). There are two variables in this study, job stress (consists of role ambiguity and role conflict) as independent variables and cyberloafing as a dependent variable as shown in Figure 1, with three hypotheses.
Primary data was obtained through a survey with a questionnaire research instrument. The research instrument uses a five-point Likert scale and closed questions where 5 means strongly agree, 4 means agree, 3 means neutral, 2 means disagree, and 1 means strongly disagree. The research instrument consists of three categories. The first category consists of six demographic questions consisting of gender, age range, domicile, status, length of work and departments. The second category consists of three questions related to the general respondent's use of the internet or gadgets.
The third category consist of 12 questions for job stress, adopted from Bowling, et al. (2017), and 12 questions for cyberloafing adopted from Pindek, Krajcevska, & Spector (2018). Questionnaires were distributed directly or face-to-face from May to June 2019.
Sample is used as a procedure that only takes part of the population which is then used to determine the desired characteristics of a population (Sugiyono, 2016). The sample in this study was the 36 employees from four back-of-the-house department at a fivestar hotel in South Jakarta, particularly located in Jalan Gatot Subroto and has been operating for more than 15 years. The hotel is classified as international brand based in United States. According to Memon et al. (2020), sample size can be determined by researchers who are considered to have the best knowledge related to their study, including aside the percentage sample taken.
On top of that, rather than emphasizing on the sample size, it is important for researcher to emphasize on accurately selecting the samples (Mooi et al., 2018). Hence, the selection of the back-office department is determined based on the work environment nature in hotel industry in which front-of-thehouse department is restricted and has limited access to explore mobile phone compared to those in back-of-the-house departments. Moreover, the particular hotel was chosen based on previous presurvey through interview on March 2019 with the Human Resources employee representative showing the data of employee turnover.
The pilot study was conducted on 50% of the total sample (equal to 18 respondents) with the aim of testing the validity and reliability of the research instrument, in which all instrument are valid and reliable. Further, to answer the first and second research questions, descriptive statistics was deployed to find out the average value of all indicators of job stress and cyberloafing (Siyoto & Sodik, 2015). Meanwhile, to answer the third research question, three hypotheses was deployed through the t test and regression analysis (Siregar, 2015). All the statistical data was computed using SPSS. Table 1 shows the profile of respondents in this study, where the majority of respondents are women (55.6%), age between 26 to 30 years old (30.6%), domicile in Jakarta (63.9%), single status (50%), and have worked in the current hotel for more than three years (52.8%) in the Sales & Marketing department (44.9%).

RESULTS AND DISCUSSIONS Respondent Profiles
In addition, referring to Figure 2 on how respondents access the internet at work, 27.8% of respondents use personal data packages, followed by 13.9% of respondents use free hotel Wi-Fi for employees, 2.8% of respondents use hotel Wi-Fi (with password) for employee-only, 19.4% of respondents use personal data packages and free hotel Wi-Fi for employees-only, and 36.1% of respondents use personal data and hotel Wi-Fi (with passwords) for employeesonly. In addition, refering to Figure 3, it can be seen that 19.4% of respondents access the internet outside of work content in working hours for less than one hour, followed by 19.4% of respondents access the internet for one to two hours, and 61.1% of respondents access the internet for more than two hour. Because the sample is an employee in the back office, most of them use a desktop computer provided by the hotel.
Moreover, it is necessary to know whether the respondent is also cyberloafing through a device intended for work. Further, Figure 4 showed that 66.7% of respondents actively use the internet network via LAN located on their respective desktop computers to access the internet during working hours outside the context of work.   Table 2 shows descriptive statistics of job stress where the standard deviation value is in the range of 0.577 to 0.874 which means that the respondents' answers are quite diverse. The descriptive statistic of job stress has a mean value of 3.87, which means that the respondents posit between neutral to agree on the job stress measurement items or indicators. Job stress is measured through two dimensions and 12 indicators adopted from Bowling, et al. (2017). The first dimension, namely role ambiguity, has a mean value of 4.02, which means that respondents feel unclear in their work. This is indicated by respondents who are not completely sure (mean 3.78) and do not fully understand (mean 3.75) what is expected of them to work, and feel that the demands of work are quite unclear (mean 3.69). On the other hand, respondents clearly understand the work that must be done (mean 4.31), including job descriptions of each job section (mean 4.17) and the clarity of the work given (mean 4.42). The second dimension, namely role conflict, has a mean value of 3.69, which means that respondents feel that they have experienced enough conflict in their work. This is indicated by respondents who feel that the assign task are adequate in causing conflict with other employee (mean 3.44). In addition, they accept jobs with different directions (mean 4.08) and are quite in situations that cause conflicts (mean 4.08). Respondents tend to agree that they received orders from their superiors to do two difficult jobs simultaneously (mean 3.92), felt compelled to do opposite jobs (mean 3.69), and the assigned work rarely caused conflict with others (mean 3.72). Sending personal e-mail (non-office e-mail)

Job Stress and Cyberloafing
,931 3,14 Y1 2 Accepting personal e-mail (non-office e-mail) ,990 3,14 Source: Data Processing Results (2019) Table 2 also shows descriptive statistics of cyberloafing where the standard deviation value is in the range of 0.810 to 1.166, which means that respondents' answers are vary. The descriptive statistic of cyberloafing has a mean value of 3.69, which means that the respondents posit between neutral to agree on the cyberloafing measurement items or indicators. Cyberloafing is measured through 12 indicators adopted from Lim & Teo in (Pindek, Krajcevska, & Spector, 2018).
It can be seen that respondents carried out activities to visit websites that are not related to work during working hours (mean 4.06), visited news website (mean 3.97), visited entertainment websites (mean 3.92), downloaded information that was not related with work (mean 3.92), played online games (mean 3.92), and searched or browsed job vacancy website (mean 3.94). In addition, respondents tend to agree that they also visited sports website (mean 3.81) and use short messaging applications such as WhatsApp (mean 3.72) and online shopping (mean 3.69). However, it is slightly different from email where respondents rarely check personal email outside of work email (mean 3.11) including sending email (mean 3.14). This result is slightly different from the research of  where front of the house employees at hotels are more likely to cyberloaf in terms of opening social media and instant messaging applications, followed by opening music and watching videos. Table 3 shows the results of the correlation test between job stress and cyberloafing with a coefficient value of 0.332 which means a low relationship with a positive direction. In addition, the Sig value was obtained at 0.048 which is smaller than 0.05 which indicates the relationship between job stress and cyberloafing is significant.

The Relationship between Job Stress and Cyberloafing
Further, Table 4 shows the results of the regression test between job stress and cyberloafing. There is a constant of 18.270 which means that if the value of work stress does not increase or is equal to zero, then the value of cyberloafing is 18.270. In addition, the regression coefficient of 0.560 states that each additional work stress value, the cyberloafing coefficient value will increase by 0.560. Thus, the higher job stress will have an impact on increasing cyberloafing. The findings show coefficient of determination value of 0.110, which means that job stress has an effect of 11% on cyberloafing. In summary, 11% of cyberloafing can be explained through job stress, while 89% of cyberloafing is explained by factors other than job stress.   Table 5 shows the tcount value of 2.405 which is greater than the ttable value of 3.032 with a Sig value. 0.022 which is smaller than 0.05. Thus, it can be concluded that H1 is accepted which is role ambiguity has a significant effect on employee cyberloafing. Table 5 shows the tcount value of 1.034 which is greater than the ttable value of 2.032 with a Sig value. 0.308 which is greater than 0.05. Thus, it can be concluded that H2 is accepted which is role conflict has a significant effect on employee cyberloafing. Table 5 shows the tcount value of 2.054 which is greater than the ttable value of 2.032 with a Sig value. 0.048 which is smaller than 0.05, so it can be stated that H3 is accepted, namely work stress consisting of role ambiguity and role conflict has a significant effect on cyberloafing of employees. The findings is inline with past study of Arshad, Aftab, dan Bukhari (2016), Koay, Chin-Hooi, & Chew (2017), and Gokcearslan, Uluyol, & Sahin, (2018).

CONCLUSION
There are three conclusions in this study. First, hotel employees are moderately stress, measured from role ambiguity and role conflict. Employees feel that they perceived ambiguity during their daily working situation with an unclear instruction and uncertainty of what are expected from them by the company or management. Second, employees are quite agreed with cyberloafing activities during working hours. Employees were doing various cyberloafing activities in browsing nonwork-related websites such as news and job vacancy. Third, job stress has a significant effect of 11% on cyberloafing with weak but positive relationship. The higher the employee job stress, the higher the cyberloafing activities carried out. From these findings, hotel companies will be able to acknowledge how employee job stress could affect their cyberloafing behavior. With that in mind, it is important for hotel to minimize employee job stress by increasing socialization and explanation in providing a particular job, including job descriptions and applicable standard operational procedures. In addition, the management is also expected to provide jobs that match the employee's position. Supervisor can also motivate staff to do their job properly because the higher the work motivation, the higher the level of performance it produces. In addition, to minimize employee cyberloafing during working hours, management need to maximize the function of mobile phones as media and communication tools during work, to restrict Wi-Fi and LAN networks for non-work related website. Future research is advised to examine the determinants of cyberloafing from other variables such as self-control (Sari & Ratnaningsih, 2018) or job characteristics (Arshad, Aftab, & Bukhari, 2016). Broader research can also be done by studying how job stress can affect turnover intention (Kim, Im, & Hwang, 2015) or work engagement (Olugbade & Karatepe, 2018).