Abstract
This study examines the factors that contribute to internet abuse intention among employees of SMEs in the services sector in Malaysia. Modified Theory of Planned Behaviour (TPB) model has been adopted as the theoretical framework. Primary data has been collected from 500 SMEs’ employees who have access to the internet at their workplace through self-administered survey questionnaire. Attitude, subjective norms, perceived behavioural control, moral norms and external locus of control are positively related to internet abuse intention. This research has included the additional variables, moral norms (internal variable) and external locus of control (external variable) to enhance the TPB model. This research will contribute to the employers who intend to reduce the internet abuse behaviour of their employees in the workplace and as a reference for future researchers on similar topics.
Keywords:Internet, abuse, Intention, SMEs, Modified, theory, of, planned, Behaviour, Malaysia.
Received:14 September 2018 / Revised: 17 October 2018 / Accepted: 21 November 2018 / Published:12 December 2018
Contribution/ Originality
This study added two more variables, external locus of control and moral norms to extend the original TPB model. From the management perspective, it would help to enhance the work productivity of Malaysian SMEs by planning various measures to avoid internet abuse intention among employees.
1. INTRODUCTION
Young (2006) has stated that the internet was seen as one of the things that can influence our daily life because internet users are likely to spend most of their leisure time in the cyber community. Internet abuse can be defined as an individual’s inability to restrain his or her internet usage, and may eventually result in a feeling of distress and causes their daily activities to be functionally impaired (Shapira et al., 2003). Small and Medium Enterprises (SMEs) are legally independent companies that have contributed 21.4% to the overall GDP growth in the services sector in Malaysia for the year 2015 according to the official website of SME Corporation Malaysia. The occurrence of internet abuse in a workplace may jeopardize the organization regarding increase in infrastructure and internal security costs, as well as possible risks related to organization’s civil and legal liabilities (Conlin, 2000; Verespej, 2000). This concern has become a serious issue in many countries, of which Malaysia is one of them. It is believed that this ‘plague’ has infected a great number of employees in Malaysia and this situation is worsening day by day. Internet abuse can be a hidden activity and the growing availability of various internet facilities in the workplace is making it easier for misuse of the internet to occur in many different forms.
In this research, internet abuse is defined as the personal use of internet for non-work-related activities during working hours through electronic devices such as computer and mobile phone. This study mainly focuses on investigating the factors that cause the internet abuse intention among employees of SMEs which includes all the micro, small and medium-sized enterprises in the services sector.
1.1. Problem Statement
The internet abuse can be seen as an intensifying issue in Malaysia. Therefore, it is important to acknowledge the existence of this problem. Cheng (2013) reported in the Star Online that employees in Malaysia with regular working hours of 9 a.m. to 5 p.m. spend a total of 4 hours being productive. Rosli et al. (2015) mentioned that the internet abuse phenomenon is affecting small and medium enterprises (SMEs) in Malaysia. By referring to SME Corporation Malaysia (2015/16) there has been a sharp increase in Information and Communication Technology (ICT) adoption among the SMEs in Malaysia in recent years. Rosli et al. (2015) further stated that it is possible for employers to encounter lawsuits if their employees have used the company network to access pornography or to post inappropriate comments in various online forums. Numerous studies have addressed these concerns. However, these studies have some deficiencies. In a similar study, Jamaluddin et al. (2015) have distributed their questionnaires by using the convenience sampling method to only 12 organizations in Kuala Lumpur and Melaka. We have extended the research to cover four States in Malaysia, which were Selangor, WPKL, Johor and Perak as these States have the highest rankings in the total SME establishments.
2. LITERATURE REVIEW
2.1. Theory of Planned Behaviour
Theory of Planned Behaviour (TPB) explains how attitudes and social norms of individuals may affect their intention to behave in a particular way (Pee et al., 2008). TPB is a model extended from the Theory of Reasoned Action (TRA) (Ajzen, 1991). The main assumption of both TRA and TPB models is that individuals are rational in considering their behaviours and actions. TPB has been widely applied and cited by many past researches (Morris et al., 2012b). It is well accepted as a model with a strong predictive function and as a well established model to forecast intention. Attitude, subjective norms and perceived behavioural control are the variables of the TPB model. Attitude is the possibility of carrying out a certain action derived from the belief and the outcome assessment (Henle et al., 2010). Subjective norm is the influence derived from the perception of those who are important to the individual in performing an act (Tan et al., 2015). Perceived behavioural control is which an individual belief that he or she can overmaster the personal resources to perform certain behaviour (Ko and Jin, 2017).
However, the determinants of intention are not bound by the three factors of the TPB model (Ajzen, 1991). According to Huang and Chen (2015) the TPB has overlooked on the explicit motivational content as a possible enticement of intention to perform an actual behaviour. Many other determinants may influence behaviour. Many researchers have modified the TPB model. The addition of external locus of control and moral norms is used in this study to fill the lack of the original TPB model.
2.2. Internet Abuse Intention
Internet abuse intention can be defined as any voluntary act of an employee to use the company’s internet to browse web sites and examine personal e-mail during their working hours (Lim, 2002). Harrison et al. (1997) stated that intention construct has been adopted by many studies, as it can predict the actual behaviour accurately. Theory of Planned Behaviour (TPB) model proposed by Ajzen (1985) suggests that people will behave according to their intentions and perceptions of control over the behaviour. Similarly, the study by Woon and Pee (2004) has examined the effects of job satisfaction, affect, social factors, perceived consequences, habit, and facilitating conditions on internet abuse intention and all the factors were found to influence the internet abuse intention, which eventually lead to the behaviour of internet abuse. Moreover, Francis et al. (2004) mentioned that although the relationship between an individual’s intention and actual behaviour is imperfect, intention can be utilized as an understudy measure of behaviour. Hence, this study proposes to determine what factors will cause the internet abuse intention.
2.3. External Locus of Control
External locus of control is where a person believes that his or her future success is due to external causes (Rotter, 1966). Some researchers have found that it may significantly affect the decision to perform a certain behaviour (Chonko et al., 2002; Hume et al., 2006). According to Poškus (2015) by integrating moral norms, it can aid to curb the lack of regards towards an individual’s moral aspects to behaviour.
2.4. Moral Norms
Moral norms refer to an individual’s belief in moral righteousness when performing certain behaviour. The addition of moral norms has resulted in an increase in intention towards certain behaviours (Taneja, 2006). According to the study of Botetzagias et al. (2015) moral norms have an extensive positive effect on the intention. In his study, moral norms were the second most important predictor to explain the recycling intention. An individual may be determined to act morally when there are internal feelings regarding the moral obligation for particular actions. Gino et al. (2011) explained that moral identity has moderated the relationship between self-regulatory resource depletion and dishonesty, in which such relationship has been weaker for the participants with high moral identity. Moral norms are closely associated with an individual’s consideration in terms of self-moral worth instead of the outcomes of the behaviour (Huang and Chen, 2015).
2.5. The Relationship between Attitudes and Internet Abuse Intention
Attitudes have been found to be a substantial factor in the prediction of behaviour (Glasman and Albarracin, 2006). Lau et al. (2003) found that an individual’s attitude towards internet abuse does play a role in internet abuse intention and behaviours. Individuals with positive attitudes towards internet abuse are more likely to commit internet abuse at work (Liberman et al., 2011; Lim and Chen, 2012; Askew et al., 2014). Hence, the hypothesis is developed as below:
H1: There is a positive relationship between attitudes and internet abuse intention.
2.6. The Relationship between Subjective Norms and Internet Abuse Intention
Based on the past studies of Henle et al. (2010); Moody and Siponen (2013) and Tan et al. (2015) subjective norms is a significant independent variable that will positively influence the intention of an individual to perform certain behaviour. Therefore, the following hypothesisis is formulated:
H2:There is a positive relationship between subjective norms and internet abuse intention.
2.7. The Relationship between Perceived Behavioural Control (PBC) and Internet Abuse Intention
The past studies of Ko and Jin (2017); Liao et al. (2010) have shown that higher level of PBC will result in a greater intention to avoid Internet misuse on the employees’ perspective. This leads to the following hypothesis:
H3: There is a positive relationship between perceived behavioural control and internet abuse intention.
2.8. The Relationship between Moral Norms and Internet Abuse Intention
Several previous empirical studies have concluded that there is a significant relationship between moral norms and the intention to certain behaviour. According to Botetzagias et al. (2015) moral norms have an extensive positive effect on the intention, similar to the study of Taneja (2006). Gino et al. (2011) explained that impaired moral norms are positively related to the intention of performing an immoral act. Hence, the following hypothesis is suggested:
H4: There is a positive relationship between immoral norms and internet abuse intention
2.9. The Relationship between External Locus of Control (ELOC) and Internet Abuse Intention
Past studies by Vitak et al. (2011) and Chak and Leung (2004) have agreed that individuals with external locus of control will have lower self-control ability over internet use and they tend to spend more time on the internet. Similarly, Davis et al. (2002) have concluded that people who possess high ELOC are more likely to be involved in internet abuse at work. With this, the following hypothesis is developed:
H5: There is a positive relationship between external locus of control and internet abuse intention
2.10. Proposed Research Model
The proposed research model is shown in Figure 1.
Figure-1. Proposed research model.
Source: Adopted from: Ajzen (1985); Ajzen and Driver (1992); Jamaluddin et al. (2015)
3. METHODOLOGY
3.1. Research Design
The quantitative data collection technique is applied in this research as numerical data gathered from questionnaire survey are measurable and quantifiable (Zikmund et al., 2010). According to Sedgwick (2014) cross-sectional study is adopted when all the measurements are acquired at a single point in time. Cross-sectional studies are also used to identify commonness, and it is quick and easy (Mann, 2003). Hence, data for this research were collected on 3 months basis from 3rd October 2016 to 10th January 2017 by using the cross-sectional approach.
3.2. Population
The target population of this study is all the employees of SMEs in Malaysia who have the access to the internet in the workplace. As retrieved from the official website of SME Corporation Malaysia, the Economic Census 2011 reported that SMEs accounted for 97.3%, which amounted to 645,136 of total business establishments in 2010. Selangor, WPKL, Johor and Perak are the four States with the highest rankings in the total SMEs establishments. Hence, our focus narrows down to the services sector in these four States.
3.3. Sampling
The sample is used as it is unfeasible to test data from every single element and that it may, at times, produce more reliable results than the entire population (Sekaran and Bougie, 2013). Taking a sample also requires lower cost (Cooper and Schindler, 2008).
The cluster random sampling method is used in this study. The researchers classify the population into groups, called clusters which often naturally occur in the population (Teddlie and Yu, 2007). The population was put into clusters by the States. The sample was further selected from the four clusters by utilizing simple random sampling, where each of the population has equal chances to be selected (Cooper and Schindler, 2008). The Comprehensive Business Directory for Small & Medium Enterprises Malaysia has been used as the sampling frame.
The minimum requirement for population of above 500,000 is 384 samples (Krejcie and Morgan, 1970). Our population size is 645,136. Hence, 500 questionnaires have been distributed to our target respondents.
3.4. Research Instrument
Self-administered questionnaire was used in this study, because it is convenient, less time consuming and does not have any specific skill requirement compared to semi-structured and in-depth interview. In addition, this survey method can also be delivered to a numerous respondents at the same time (Hair et al., 2011).
3.5. Constructs Measurement
Table-1. Measurement of Variables.
Variables |
Measure |
Scale of Measurement |
|
Personal |
Gender |
Nominal |
|
Profile |
Age |
Ordinal |
|
Highest Education Completed |
Nominal |
||
Working Experience |
Ordinal |
||
The Time of Internet Usage during Working Hours |
Ordinal |
||
Company Profile |
Age of Firm |
Ordinal |
|
Category of Services Sector |
Nominal |
||
Occupational Status |
Nominal |
||
Independent Variables |
Attitudes |
Interval |
5-point Likert Scale |
Subjective Norms |
Interval |
5-point Likert Scale |
|
Perceived Behavioural Control |
Interval |
5-point Likert Scale |
|
Moral Norms |
Interval |
5-point Likert Scale |
|
External Locus of Control |
Interval |
5-point Likert Scale |
|
Dependent Variable |
Internet Abuse Intention among Employees of SMEs in the Services Sector |
Interval |
5-point Likert Scale |
Table 1 shows the measurement of each variable contained in the questionnaire. 5-point Likert scale was employed in this study. 5-point Likert scale ranged from (1) “Strongly disagree” to (5) “Strongly agree”. 5-point Likert scale is convenient for the respondents to view and choose based on the list of scale descriptors (Maringka, 2012).
4. DATA ANALYSIS
SAS Enterprise Guide 7.1 was used in this study to analyse the data collected from survey questionnaire.
4.1. Pilot Test
It is important to conduct the pilot test before distributing the questionnaires in a large number. Thabane et al. (2010) defined a pilot test as an investigation way to test the feasibility of procedures and methods that may be useful for the following large scale study. It is sufficient to examine the validity of survey if 20 respondents participate in the pilot test (Arain et al., 2010). Hence, 20 target respondents were selected to conduct the pilot test and this took place at SMEs in Ipoh, Perak State. Table 2 below shows the reliability test result. It shows that the Cronbach’s Alpha values of the variables fell within the range of 0.708978 to 0.908125. Consequently, we can conclude that our survey questionnaires had met the acceptable level of reliability test which is more than 0.7 to test the validity of a survey.
Table-2. Cronbach’s Alpha Values of Pilot Test.
Variables |
Number of Items |
Cronbach’s Alpha |
Attitudes |
4 |
0.781524 |
Subjective Norms |
6 |
0.860152 |
Perceived Behavioural Control |
4 |
0.763889 |
Moral Norms |
4 |
0.708978 |
External Locus of Control |
12 |
0.808045 |
Internet Abuse Intention |
7 |
0.908125 |
Table-3. Summary of demographic profile.
Profile |
Category |
Frequency |
Percent (%) |
Gender |
Female |
226 |
56.5 |
Male |
174 |
43.5 |
|
Age |
Below 20 |
82 |
20.5 |
20-29 |
242 |
60.5 |
|
30-39 |
42 |
10.5 |
|
40 and above |
34 |
8.5 |
|
Highest education completed |
UPSR/PMR/SPR/SPM |
23 |
5.75 |
STPM |
33 |
8.25 |
|
Diploma |
126 |
31.5 |
|
Bachelor’s Degree |
209 |
52.25 |
|
Master’s Degree |
9 |
2.25 |
|
Working experience |
Up to 5 |
215 |
53.75 |
10-Jun |
99 |
24.75 |
|
20-Nov |
56 |
14 |
|
More than 20 |
30 |
7.5 |
|
Time of internet usage |
Up to 1 hour |
150 |
37.5 |
1-5 hours |
227 |
56.75 |
|
More than 5 hours |
23 |
5.75 |
|
Age of firm |
Less than 10 |
179 |
44.75 |
10 or more |
221 |
55.25 |
|
Category of service sector |
Wholesale & retailer |
56 |
14 |
Food & beverages |
27 |
6.75 |
|
Transportation & storage |
5 |
1.25 |
|
Hotels & restaurants |
38 |
9.5 |
|
Professional & ICT services |
118 |
29.5 |
|
Private education & healthcare |
82 |
20.5 |
|
Entertainment & manufacturing services |
74 |
18.5 |
|
Occupational Status |
Managerial |
7 |
1.75 |
Non-managerial |
393 |
98.25 |
5. DESCRIPTIVE ANALYSIS
5.1. Demographic Profile of the Respondents
Table 3 shows the demographic profile of the respondents. Overall, most of the respondents are females aged 20 to 29 with Bachelor’s Degree qualification, who have working experience up to 5 years and are using the internet for 1 to 5 hours at work.
5.2. Central Tendencies Measurement of Constructs
Based on Table 4, the constructs have the mean values in the range of 2.29250 to 3.50500. The results indicate that the constructs are dispersed between disagree to agree. This table also lists out the standard deviations which delegate the dispersion of the data of every item. We can obtain the information that the lowest standard deviation is 0.84482 while the highest figure is 1.17176.
Table-4. Central Tendencies Measurement of Constructs.
Variables |
Items |
Mean |
Standard Deviation |
Attitudes |
AT 1 |
2.5075 |
1.10135 |
AT 2 |
2.4975 |
1.1416 |
|
AT 3 |
2.6925 |
1.05865 |
|
AT 4 |
2.3725 |
1.06846 |
|
Subjective Norms |
SN 1 |
3.505 |
1.01368 |
SN 2 |
3.47 |
0.99578 |
|
SN 3 |
2.9325 |
1.03713 |
|
SN 4 |
2.78 |
0.95859 |
|
SN 5 |
2.2925 |
0.84482 |
|
SN 6 |
2.34 |
0.93626 |
|
Perceived Behavioural Control |
PBC 1 |
3.0225 |
1.09086 |
PBC 2 |
2.555 |
0.85663 |
|
PBC 3 |
3.1875 |
1.14701 |
|
PBC 4 |
3.1975 |
1.12546 |
|
Moral Norms |
MN 1 |
2.69 |
1.07087 |
MN 2 |
2.3775 |
1.04053 |
|
MN 3 |
2.61 |
0.95927 |
|
MN 4 |
2.63 |
0.97235 |
|
External Locus of Control |
ELOC 1 |
2.87 |
1.04922 |
ELOC 2 |
2.6125 |
0.94847 |
|
ELOC 3 |
2.6875 |
1.01608 |
|
ELOC 4 |
3.18 |
1.02236 |
|
ELOC 5 |
2.6075 |
1.00547 |
|
ELOC 6 |
2.6025 |
0.93363 |
|
ELOC 7 |
2.6175 |
0.96878 |
|
ELOC 8 |
2.7525 |
1.02414 |
|
ELOC 9 |
2.7925 |
1.04267 |
|
External Locus of Control |
ELOC 10 |
2.825 |
0.90078 |
ELOC 11 |
2.58 |
0.92777 |
|
ELOC 12 |
2.61 |
1.04673 |
|
Internet Abuse Intention |
IAI 1 |
3.02 |
1.17176 |
IAI 2 |
3.295 |
0.99773 |
|
IAI 3 |
3.345 |
0.98915 |
|
IAI 4 |
2.98 |
1.15453 |
|
IAI 5 |
2.995 |
1.13499 |
|
IAI 6 |
2.6775 |
0.97513 |
|
IAI 7 |
2.735 |
0.95502 |
5.3. Normality and Reliability Test
The outcome of the variable’s items based on the normality test and reliability test is shown in Table 5.
Table-5. Normality and reliability test results.
Variables |
Items |
Skewness |
Kurtosis |
Cronbach’s Alpha |
Attitudes |
AT1 |
0.399462 |
-0.54214 |
0.768405 |
AT2 |
0.575033 |
-0.43664 |
||
AT3 |
0.359104 |
-0.43853 |
||
AT4 |
0.601569 |
-0.16065 |
||
Subjective Norms |
SN1 |
-0.28209 |
-0.68466 |
0.863805 |
SN2 |
-0.16169 |
-0.69574 |
||
SN3 |
-0.1625 |
-0.56016 |
||
SN4 |
-0.16517 |
-0.45618 |
||
SN5 |
0.844824 |
0.713728 |
||
SN6 |
0.176543 |
-0.6062 |
||
Perceived Behavioural Control |
PBC1 |
-0.03309 |
-0.72448 |
0.716514 |
PBC2 |
0.681698 |
0.659201 |
||
PBC3 |
-0.21184 |
-0.70177 |
||
PBC4 |
-0.18293 |
-0.6569 |
||
Moral Norms |
MN1 |
0.310637 |
-0.4297 |
0.873437 |
MN2 |
0.72496 |
0.115489 |
||
MN3 |
0.300489 |
-0.04223 |
||
MN4 |
0.43564 |
0.159255 |
||
External Locus of Control |
ELOC1 |
-0.07811 |
-0.60684 |
0.917609 |
ELOC2 |
0.010448 |
-0.73367 |
||
ELOC3 |
-0.03506 |
-0.85639 |
||
ELOC4 |
-0.6212 |
-0.53951 |
||
ELOC5 |
0.163952 |
-0.64657 |
||
ELOC6 |
0.107931 |
-0.46867 |
||
ELOC7 |
0.097231 |
-0.51152 |
||
ELOC8 |
0.060495 |
-0.57765 |
||
ELOC9 |
0.05022 |
-0.39442 |
||
ELOC10 |
-0.24774 |
0.400469 |
||
ELOC11 |
-0.15024 |
-0.81792 |
||
ELOC12 |
0.226674 |
-0.40561 |
||
Internet Abuse Intention |
IAI1 |
0.008026 |
-0.74694 |
0.922997 |
IAI2 |
0.005625 |
-0.76597 |
||
IAI3 |
-0.00125 |
-0.85879 |
||
IAI4 |
0.058783 |
-0.74639 |
||
IAI5 |
0.009837 |
-0.68913 |
||
IAI6 |
0.047496 |
-0.73082 |
||
IAI7 |
-0.00361 |
-0.76128 |
Past studies have suggested that the satisfactory range of skewness and kurtosis in every variable is between ±1 (Byrne, 2001; Fotopoulos and Psomas, 2009). The skewness values range of our research is between -0.6212038 to 0.84482428 whereas the kurtosis values range is between -0.8587949 to 0.71372807. The absolute value for skewness and kurtosis test is ±1. Hence, all the variables in this research have passed the normality test.
This study has employed reliability test for each variable. According to Webb et al. (2006) variables are treated as reliable if the Cronbach’s Alpha has reached a value of 0.7. Therefore, the assumptions of reliability for all the items in table 5 are considered as reliable.
5.4. Pearson’s Correlation Analysis
Table 6 shows that the correlation values between the independent variables show a positive correlation ranging from 0.11364 to 0.42751. The dependent variable has a positive relationship with all the independent variables which ranged from 0.29019 to 0.61047. Since none of the correlation values are higher than 0.90, no multicollinearity problem has been revealed in this research (Hair et al., 2009). Moreover, significant relationships exist between the variables as all the variables have p-values that are less than 0.05.
Table-6. Pearson Correlation Coefficients.
Variables |
AT |
SN |
PBC |
MN |
ELOC |
IAI |
AT |
1 |
|||||
SN |
0.20041 |
|||||
<.0001 |
||||||
PBC |
0.15042 |
0.21417 |
||||
0.0026 |
<.0001 |
|||||
MN |
0.16382 |
0.42751 |
0.11364 |
|||
0.001 |
<.0001 |
0.023 |
||||
ELOC |
0.20124 |
0.39967 |
0.20757 |
0.24381 |
||
<.0001 |
<.0001 |
<.0001 |
<.0001 |
|||
IAI |
0.46626 |
0.60904 |
0.29019 |
0.61047 |
0.539 |
1 |
<.0001 |
<.0001 |
<.0001 |
<.0001 |
<.0001 |
5.5. Multiple Linear Regression
R2 value is presented in table 7 with a value of 0.6932. This signifies that five factors adopted in this study can explain about 69.32% of the changes in the intention of abusing internet among employee and the other 30.68% is comprised of other variables that are not included in this research.
Table-7. Model summary.
Model |
R2 |
Adjusted R2 |
Std. error of the estimate |
1 |
0.6932 |
0.6893 |
0.4876 |
As shown in table 8, the F-value demonstrates a value of 178.03 with the significance level of <0.001 which is greater than the value obtained from the F-table, 2.175. Large F-value indicates that the model employed fits this research. Furthermore, p-value with a figure less than 0.05 indicates that at least one of the five independent variables can be used to explain and model the dependent variable. As a conclusion, the relationship between all the five factors and IAI (DV) in this research study is significant.
Table-8. ANOVA results.
Model |
Sum of Squares |
df |
Mean Square |
F |
Sig. |
Regression |
211.6335 |
5 |
42.32669 |
178.03 |
<.0001 |
Residual |
93.67465 |
394 |
0.23775 |
||
Total |
305.3081 |
399 |
Table 9 shows that AT (p < 0.0001), SN (p < 0.0001), PBC (p = 0.0015), MN (p < 0.0001) and ELOC (p < 0.0001) have a significant relationship with internet abuse intention. All the independent variables have an impact to influence the internet abuse intention as their p-value is less than 0.05. These results have confirmed that all the hypotheses are supported. The MLR equation for the model is formulated below:
IAI = - 0.81243 + 0.29616 (AT) + 0.31218 (SN) + 0.10413 (PBC) + 0.38088 (MN) + 0.32241(ELOC)
Table-9. Multiple linear regression analysis results.
Unstandardized Coefficients |
Supported/Not supported |
|||||||
Model |
Par. Est |
Std. Est |
t |
Sig. |
Tolerance |
Var.Inf. |
Hypo. |
|
Intercept |
-0.81243 |
0.14186 |
-5.73 |
<.0001 |
- |
0 |
||
AT |
0.29616 |
0.03018 |
9.81 |
<.0001 |
0.92828 |
1.07726 |
H1 |
Supported |
SN |
0.31218 |
0.03895 |
8.02 |
<.0001 |
0.70698 |
1.41447 |
H2 |
Supported |
PBC |
0.10413 |
0.03248 |
3.21 |
0.0015 |
0.92781 |
1.0778 |
H3 |
Supported |
MN |
0.38088 |
0.03156 |
12.07 |
<.0001 |
0.80602 |
1.24066 |
H4 |
Supported |
ELOC |
0.32241 |
0.03781 |
8.53 |
<.0001 |
0.80783 |
1.23789 |
H5 |
Supported |
6. DISCUSSION, CONCLUSION AND IMPLICATIONS
6.1. Discussion of Major Findings
The results show that attitudes, subjective norms, perceived behavioural control, external locus of control and immoral norms are positively related to internet abuse intention. Attitudes achieved a p-value of <0.0001 which illustrates attitudes is positively related to the intention of abusing the internet among employees of SMEs in the services sector in Malaysia. The results statistically prove that there is a significant relationship between attitudes and internet abuse intention since the p-value is below 0.05. According to Rosli et al. (2015) an undesirable attitude led by emotional instability, introversion and non-conscientious are significantly correlated with internet abuse intention. In accordance to the study of Liberman et al. (2011) attitudes towards internet abuse behaviours are positively related to internet abuse. In addition, past studies have also determined that involvement, beliefs and commitment as several aspects of attitudes can be a strong determinant to explain the willingness of employees to commit an act of computer abuse (Lee and Lee, 2002).
Subjective norms show a positive relationship and significant impact on the internet abuse intention towards the employees of SMEs in the services sector of Malaysia. The result of the p-value for this variable is less than 0.0001. This implies that the internet abuse intention is affected by subjective norms of the respondents since the p-value is less than 0.05. Generally, the higher an individual’s subjective norm is, the greater the internet abuse intention. The result generated by this study conforms to the researches by Moody and Siponen (2013) and Henle et al. (2010). Moody and Siponen (2013) results showed that the personal self-concept is important for the intention of internet abuse, and the self-concept is mostly influenced by the supervisor and peer culture. Moreover, Henle et al. (2010) have found that subjective norms are one of the causes for organizational misbehaviour, including time theft.
Perceived behavioural control (PBC) as one of the independent variables in our research also satisfied the requirement of p-value < 0.05. Thus, the result shows that PBC has a positive relationship with IAI among employees of services sector in Malaysia. However, it is the highest p-value figure comparable to other variables due to the p-value figure for other independent variables are <.0001.The findings for this variable are aligned with Liao et al. (2010) which indicates that the more the user comprehends the mechanisms of internet abuse, the higher probability that the user gets the thoughts to internet abuse. Research by Taneja (2006) also established that this variable had a significant relationship towards internet abuse intention.
In the context of this study, Immoral norm has a positive and significant relationship with IAI. The p-value generated from the MLR analysis has shown a value of <.0001 which is below 0.05. This illustrates the significant relationship between the MN and the intention to internet abuse by the employees of services sector in Malaysia. Generally, the greater the MN of employees, the lower the IAI. Several past studies can back this result. In the study of Taneja (2006) it has been found that MN of user has a significant effect on the intention of usage of information system (IS) assets adversely. According to the research done by Godin et al. (2005) MN has an important impact on whether people endorse their intentions. In another study, MN has been proved to have a significant impact on the intention of knowledge sharing (Huang and Chen, 2015).
There is a significant relationship between ELOC and the IAI among the employees of services sector in Malaysia. The p-value generated from the MLR analysis shows a value of <.0001 which is lower than 0.05. This indicates that the hypothesis is supported and it means ELOC has a positive relationship with IAI. The greater the ELOC of employees, the higher the IAI among employees of services sector in Malaysia. Our result is supported by the study of Blanchard and Henle (2008) which proved that employees with high ELOC will have higher chance of abusing internet as they believed that the probability of getting caught on the action of internet abuse action during working hours was purely by chance or because they were unlucky from the employees’ perspective. Moreover, past researchers also concluded that ELOC will influence employees’ internet addictions and finally lead to internet abuse (Chen et al., 2008). Bellamy and Hanewicz (2002) did a similar study and found that ELOC is related to an increased risk of internet abuse.
6.2. Theoretical Implications
Theoretically, this research examined the importance of Modified Theory of Planned Behaviour (TPB) model in finding out the factors that will contribute to Internet Abuse Intention among employees of SMEs in the services sector in Malaysia. The proposed conceptual framework focused on how the five independent variables, Attitudes, Subjective Norms, Perceived Behavioural Control, Moral Norms, and External Locus of Control could affect the Internet Abuse Intention. Undoubtedly, the Modified TPB model is the valid theory in explaining the Internet Abuse Intention of the employees of SMEs in Malaysia. Consequently, the Modified TPB model is suggested to the future researchers in conducting any research relevant to Internet Abuse Intention.
According to Hasbullah et al. (2014) the existing TPB model (Attitudes, Subjective Norms, Perceived Behavioural Control) lacks external factors. Hence, this research has included the additional variables, moral norms (internal variable) and external locus of control (external variable) to enhance the TPB model.
6.3. Practical Implication
Firstly, many of the past studies have analysed the internet abuse intention among adolescents, generation X and generation Y but not in the workplace. Therefore, here is a study about internet abuse intention in the workplace and it had narrowed down the scope into only on SMEs in the services sector so that this research is more specific to this target respondents.
Internet abuse intention among employees had created serious productivity matters and legal issues towards the organizations. From the consequences of this research, employers will have a clearer guideline to handle the occurrence of employee’s internet abuse intention. On the other hand, employers will also able to manage this issue in a proper and efficient way. By understanding the factors of internet abuse intention among employees, employers also can plan appropriate measures to deal with this issue.
6.4. Limitations and Recommendations for Future Research
Several limitations have been noticed in this study. Firstly, our research focused on the services sector and four States in Malaysia only due to the cost and time saving reasons. There are many sectors such as construction and manufacturing sectors in Malaysia which have not been considered. Besides that, Malaysia consists of thirteen States and only the employees in four States were covered in our research. It is suggested that future researchers study the internet abuse intention among employees in other sectors and other States in Malaysia, as employees from different sectors and different States may show different internet abuse intention.
In addition, our study has generally focused on all the micro, small and medium-sized SMEs. Future researchers are suggested to carry out their study by specializing on the size of SMEs selected.
Employees in different sizes of SMEs may behave differently for the intention of abusing internet in the workplace. Besides, the targeted respondents in this study were mainly located in the middle and south Malaysia. It is strongly suggested that similar research be conducted to investigate the factors of internet abuse intention of employees in north Malaysia as they might have a different mindset for internet abuse compared to employees of the middle and south Malaysia.
Moreover, self-administered survey questionnaire method has been adopted to collect all the data for this research. It is a self-report method as target respondents are assumed to respond to the questionnaires accurately. However, there is a possibility that people might answer the questionnaires incorrectly. Besides, collecting back the self-administered survey questionnaire may be time-consuming and there is also a chance that researchers cannot collect back their survey questionnaire. The face-to-face interview method is recommended to collect reliable and accurate data as interviewers are allowed to take advantage of social or nonverbal cues to gain additional information from the interview (Emans, 1986).
Also, a cross-sectional study was applied to this research due to the insufficiency of research time and low-cost budget. Cross-sectional study is adopted when all the measurements are acquired at a single point in time (Sedgwick, 2014). The result can be irrelevant in the future as the information obtained from this research may be outdated. Future researchers can adopt longitudinal study to investigate the similar issue as the research can be carried out in a longer period instead of just focusing on a single point in time. Hedeker and Gibbons (2006) commented that information can be provided with more detail by applying longitudinal study.
7. CONCLUSION
In brief, this research has achieved and fulfilled all of its objectives and research questions. It proved that the relationship between all variables are statistically significant and all of the proposed hypotheses have been accepted. This research also concludes that all of the independent variables (i.e AT, SN, PBC, MN and ELOC) have significant influence on the internet abuse intention among employees of SMEs in the services sector in Malaysia.
Funding: This study received no specific financial support. |
Competing Interests: The authors declare that they have no competing interests. |
Contributors/Acknowledgement: All authors contributed equally to the conception and design of the study. |
REFERENCES
Ajzen, I., 1985. From intentions to actions: A theory of planned behavior. In Action control. Berlin, Heidelberg: Springer. pp: 11-39.
Ajzen, I., 1991. The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2): 179-211.
Ajzen, I. and B.L. Driver, 1992. Application of the theory of planned behavior to leisure choice. Journal of Leisure Research, 24(3): 207-224. Available at: https://doi.org/10.1080/00222216.1992.11969889.
Arain, M., M.J. Campbell, C.L. Cooper and G.A. Lancaster, 2010. What is a pilot or feasibility study? A review of current practice and editorial policy. Medical Research Methodology, 10(1): 1-7. Available at: https://doi.org/10.1186/1471-2288-10-67.
Askew, K., J.E. Buckner, M.U. Taing, A. Ilie, J.A. Bauer and M.D. Coovert, 2014. Explaining cyberloafing. Computers in Human Behavior, 36(C): 510-519.
Bellamy, A. and C. Hanewicz, 2002. An exploratory analyses of the social nature of internet addiction: A research paper submitted to the electronic journal of sociology. Electronic Journal of Sociology.
Blanchard, A.L. and C.A. Henle, 2008. Correlates of different forms of cyberloafing: The role of norms and external locus of control. Computers in Human Behavior, 24(3): 1067-1084. Available at: https://doi.org/10.1016/j.chb.2007.03.008.
Botetzagias, I., A.-F. Dima and C. Malesios, 2015. Extending the theory of planned behavior in the context of recycling: The role of moral norms and of demographic predictors. Resources, Conservation and Recycling, 95: 58-67. Available at: https://doi.org/10.1016/j.resconrec.2014.12.004.
Byrne, B., 2001. Structural equation modeling with AMOS: Basic concepts, applications, and programming. Mahwah, USA: Erlbaum.
Chak, K. and L. Leung, 2004. Shyness and locus of control as predictors of internet addiction and internet use. Cyber Psychology & Behavior, 7(5): 559-570. Available at: https://doi.org/10.1089/1094931042403073.
Chen, J.V., C.C. Chen and H.-H. Yang, 2008. An empirical evaluation of key factors contributing to internet abuse in the workplace. Industrial Management & Data Systems, 108(1): 87-106. Available at: https://doi.org/10.1108/02635570810844106.
Cheng, N., 2013. Malaysian workers lack performance culture, say experts. The Star Online. Available from http://www.thestar.com.my/news/nation/2013/02/27/malaysian-workers-lack-performance-culture-say-experts/ .
Chonko, L.B., E. Jones, J.A. Roberts and A.J. Dubinsky, 2002. The role of environmental turbulence, readiness for change, and salesperson learning in the success of sales force change. Journal of Personal Selling & Sales Management, 22(4): 227-245.
Conlin, M., 2000. Workers, surf at your own risk. Bloomberg News. Available at: https://www.bloomberg.com/news/articles/2000-06-11/workers-surf-at-your-own-risk .
Cooper, D.R. and P.S. Schindler, 2008. Business research methods. 10th Edn., New York: McGraw-Hill Irwin.
Davis, R.A., G.L. Flett and A. Besser, 2002. Validation of a new scale for measuring problematic internet use: Implications for pre-employment screening. Cyber Psychology & Behavior, 5(4): 331-345. Available at: https://doi.org/10.1089/109493102760275581.
Emans, B., 1986. Interviewen; theorie, techniek en training. Groningen: Wolters-Noordhoff.
Fotopoulos, C.B. and E.L. Psomas, 2009. The impact of “soft” and “hard” TQM elements on quality management results. International Journal of Quality & Reliability Management, 26(2): 150-163. Available at: https://doi.org/10.1108/02656710910928798.
Francis, J.J., M.P. Eccles, M. Johnston, A.E. Walker, J.M. Grimshaw, R. Foy, E.F.S. Kaner, L. Smith and D. Bonetti, 2004. Constructing questionnaires based on the theory of planned behaviour: A manual for health services researchers. Newcastle-Upon-Tyne: University of Newcastle.
Gino, F., M.E. Schweitzer, N.L. Mead and D. Ariely, 2011. Unable to resist temptation: How self-control depletion promotes unethical behavior. Organizational Behavior and Human Decision Processes, 115(2): 191-203. Available at: https://doi.org/10.1016/j.obhdp.2011.03.001.
Glasman, L. and D. Albarracin, 2006. Forming attitudes that predict future behavior: A meta-analysis of the attitude-behavior relation. Psychological Bulletin, 132(5): 778-822. Available at: https://doi.org/10.1037/0033-2909.132.5.778.
Godin, G., M. Conner and P. Sheeran, 2005. Bridging the intention–behaviour gap: The role of moral norm. British Journal of Social Psychology, 44(4): 497-512. Available at: https://doi.org/10.1348/014466604x17452.
Hair, J.F., W.C. Black, B.J. Babin and R.E. Anderson, 2009. Multivariate data analysis. 7th Edn., NJ: Pearson Prentice-Hall.
Hair, J.F., M.W. Celsi, A.H. Money, P. Samouel and M. Page, 2011. Essentials of business research methods. 2nd Edn., Armonk, New York: M. E. Sharpe.
Harrison, D.A., P.P. Mykytyn Jr and C.K. Riemenschneider, 1997. Executive decisions about adoption of information technology in small business: Theory and empirical tests. Information Systems Research, 8(2): 171-195. Available at: https://doi.org/10.1287/isre.8.2.171.
Hasbullah, N., A.J. Mahajar and M.I. Salleh, 2014. Extending the theory of planned behavior: Evidence of the arguments of its sufficiency. International Journal of Humanities and Social Science, 4(14): 101-105.
Hedeker and Gibbons, 2006. Longitudinal data analysis. New York:: Wiley.
Henle, C.A., C.L. Reeve and V.E. Pitts, 2010. Stealing time at work: Attitudes, social pressure, and perceived control as predictors of time theft. Journal of Business Ethics, 94(1): 53-67. Available at: https://doi.org/10.1007/s10551-009-0249-z.
Huang, C.-C. and T.-H. Chen, 2015. Moral norm and the two-component theory of planned behavior model in predicting knowledge sharing intention: A role of mediator desire. Psychology, 6(13): 1685-1699. Available at: https://doi.org/10.4236/psych.2015.613165.
Hume, E.C., A. Smith and F. Stephen, 2006. University student ethics: The differential explanatory effect of locus of control. Academy of Educational Leadership Journal, 10(3): 49-58.
Jamaluddin, H., Z. Ahmad, M. Alias and M. Simun, 2015. Personal internet use: The use of personal mobile devices at the workplace. Procedia-Social and Behavioural Sciences, 172: 495-502. Available at: https://doi.org/10.1016/j.sbspro.2015.01.391.
Ko, B.S. and B. Jin, 2017. Predictors of purchase intention toward green apparel products: A cross-cultural investigation in the USA and China. Journal of Fashion Marketing and Management: An International Journal, 21(1): 70-87.k
Krejcie, R.V. and D.W. Morgan, 1970. Determining sample size for research activities. Educational and Psychological Measurement, 30(3): 607-610. Available at: https://doi.org/10.1177/001316447003000308.
Lau, V.C., W.T. Au and J.M. Ho, 2003. A qualitative and quantitative review of antecedents of counterproductive behavior in organizations. Journal of Business and Psychology, 18(1): 73-99.
Lee, J. and Y. Lee, 2002. A holistic model of computer abuse within organizations. Information Management & Computer Security, 10(2): 57-63. Available at: https://doi.org/10.1108/09685220210424104.
Liao, Q., X. Luo, A. Gurung and L. Li, 2010. Workplace management and employee misuse: Does punishment matter? Journal of Computer Information Systems, 50(2): 49-59.
Liberman, B., G. Seidman, K.Y. Mckenna and L.E. Buffardi, 2011. Employee job attitudes and organizational characteristics as predictors of cyberloafing. Computers in Human Behavior, 27(6): 2192-2199. Available at: https://doi.org/10.1016/j.chb.2011.06.015.
Lim, V.K., 2002. The it way of loafing on the job: Cyberloafing, neutralizing and organizational justice. Journal of Organizational Behavior, 23(5): 675-694. Available at: https://doi.org/10.1002/job.161.
Lim, V.K. and D.J. Chen, 2012. Cyberloafing at the workplace: Gain or drain on work? Behaviour & Information Technology, 31(4): 343-353. Available at: https://doi.org/10.1080/01449290903353054.
Mann, C., 2003. Observational research methods. Research design II: cohort, cross sectional, and case-control studies. Emergency Medicine Journal, 20(1): 54-60. Available at: https://doi.org/10.1136/emj.20.1.54.
Maringka, J., 2012. 5 or 7-point likert scale dilemma. Methodspace Discussion. [Accessed January 17, 2017].
Moody, G.D. and M. Siponen, 2013. Using the theory of interpersonal behavior to explain non-work-related personal use of the internet at work. Information & Management, 50(6): 322-335. Available at: https://doi.org/10.1016/j.im.2013.04.005.
Morris, J., M. Marzano, N. Dandy and L. O’Brien, 2012b. Forestry, sustainable behaviours and behaviour change - setting the scene. Summary Report. Forest Research, Farnham.
Pee, L.G., I.M. Woon and A. Kankanhalli, 2008. Explaining non-work-related computing in the workplace: A comparison of alternative models. Information & Management, 45(2): 120-130. Available at: https://doi.org/10.1016/j.im.2008.01.004.
Poškus, M.S., 2015. Predicting recycling behavior by including moral norms into the theory of planned behavior. Psychology, 52(52): 22-32.
Rosli, R.M., A.A. Aziz, A.M. Lokman and S.R. Hamidi, 2015. Factors contributing to internet abuse in the workplace: Behaviour in SMEs. Proceedings of the International Symposium on Research of Arts, Design and Humanities. pp: 101-112.
Rotter, J., 1966. Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs, 80(1): 1–28. Available at: https://doi.org/10.1037/h0092976.
Sedgwick, P., 2014. Cross sectional studies: Advantages and disadvantages. British Medical Journal, 348(mar262): g2276. Available at: https://doi.org/10.1136/bmj.g2276
Sekaran, U. and R. Bougie, 2013. Research methods for business: A skill building approach. 6th Edn., Chichester, UK: Wiley.
Shapira, N.A., M.C. Lessig, T.D. Goldsmith, S.T. Szabo, M. Lazoritz, M.S. Gold and D.J. Stein, 2003. Problematic internet use: Proposed classification and diagnostic criteria. Depression and Anxiety, 17(4): 207-216. Available at: https://doi.org/10.1002/da.10094.
SME Corporation Malaysia, 2015/16. Annual Report. Kuala Lumpur Sentral, Kuala Lumpur.
Tan, F.Y., J. Johari and A.F. Sukery, 2015. The influence of attitude, subjective norms, and perceived behavioural control on intention to return to work: A case of SOCSO's insured employees. Kajian Malaysia, 33(1): 141-154.
Taneja, A., 2006. Determinants of adverse usage of information systems assets: A study of antecedents of IS exploit in organizations. Unpublished Master Dissertation, University of Texas at Arlington.
Teddlie, C. and F. Yu, 2007. Mixed methods sampling: A typology with examples. Journal of Mixed Methods Research, 1(1): 77-100. Available at: https://doi.org/10.1177/1558689806292430.
Thabane, L., J. Ma, R. Chu, J. Cheng, A. Ismaila, L.P. Rios, R. Robson, M. Thabane, L. Giangregorio and C.H. Goldsmith, 2010. A tutorial on pilot studies: The what, why and how. Medical Research Methodology, 10(1): 1-10. Available at: https://doi.org/10.1186/1471-2288-10-1.
Verespej, M.A., 2000. Internet surfing. Industry Week, 249(3): 58-64.
Vitak, J., J. Crouse and R. LaRose, 2011. Personal internet use at work: Understanding cyberslacking. Computers in Human Behavior, 27(5): 1751-1759. Available at: https://doi.org/10.1016/j.chb.2011.03.002.
Webb, N.M., R.J. Shavelson and E.H. Haertel, 2006. 4 reliability coefficients and generalizability theory. Handbook of Statistics, 26: 81-124. Available at: https://doi.org/10.1016/s0169-7161(06)26004-8.
Woon, I.M. and L.G. Pee, 2004. Behavioral factors affecting Internet abuse in the workplace-an empirical investigation. SIGHCI 2004 Proceedings, 5.
Young, B., 2006. A study on the effect of internet use and social capital on the academic performance. Development and Society, 35(1): 107-123.
Zikmund, W.G., B.J. Babin, J.C. Carr and M. Griffin, 2010. Business research methods. 8th Edn., Mason, HO: Cengage Learning.