Ijcsn 2012 1 6 41
Ijcsn 2012 1 6 41
Ijcsn 2012 1 6 41
OnPerceptions Towards On-line Banking Security: An Empirical Investigation of a Developing Country`s Sector, OnBanking Banking Sector, how secure is On-line Banking
1
Dean, faculty of Business, Solusi University Bulawayo, +263, Zimbabwe / 2 Department of Computers and Information Systems Solusi University, Bulawayo, +263, Zimbabwe
Abstract
The increase in computer crime has led to scepticism about the move made by the banks to introduce on-line banking. Some view this as a noble move which has made the banking system more efficient, reliable and secure, while others view it as a risky and insecure way of banking. The aim of this study was to assess whether on-line banking in the developing countries is secure or not. The researcher chose a descriptive-quantitative research design. Data was collected using a self constructed questionnaire. Convenience sampling and stratified random sampling techniques were used to select the main subjects of the study. Generally on average there was no significant difference between the perceptions of management bank personnel and non-management bank personnel on the security of on-line banking. The study recommends further future studies on the security of on-line banking in developing countries based on the perceptions of the customers themselves, who are using on-line banking services, the Common Criteria for Information Technology Security and also a study of the latent dimensions of on-line banking security as extracted by factor analysis, how they differ from elements of information security as derived from the theoretical framework and literature. Keywords: on-line banking; on-line banking security; information security; network services; banking system
1. Introduction
Information systems concentrate data in computer files that have the potential to be accessed by large numbers of people in and outside of organisations. While security breaches and damages of information systems still come from organisational insiders, security breaches are increasing, especially in developing countries because organisations are now open to outsiders through the internet. As a result, automated data are more susceptible to error, destruction, fraud and misuse. The banking sector in Zimbabwe has introduced, of late, on-line banking facilities and these are heavily dependent on the use of internet. According to Laudon and Laudon [1] on-line banking, one of the systems
involved in the movement of funds remains a subject to attack by natural and human threats. E-mail threats such as spam are well known, but there is another major entry point into a network, that is, the Web. As companies have become more adept at stopping e-mail threats, cyber-criminals have discovered new ways to infiltrate corporate networks through the internet. Research and Markets.Com (August 2004)[2] predicted that on-line banking will overtake traditional banking channels and become the single most important consumer banking channel. In internet banking system, the bank has a centralised database that is web enabled. Any organisation in e-business must consistently deliver great performance that matters to customers, raise customers` expectations and force all competitors to respond and generate rapid growth [3]. Furthermore, security and reliability are absolutely critical in systems for banking and financial services because errors, fraud, and disruption of service can lead to large monetary losses and the erosion of customer confidence in those companies and even the entire financial industry [1]. An introduction of on-line banking services in order to improve the efficiency in service delivery in the banking sector is to be welcome as a noble idea in Zimbabwe and the developing world in general, and a great step towards improving the performance of the banking sector in line with advancements in technology. Organisations that seek to make a difference should harness the latest technologies and stay abreast with the industry`s competition if they are to remain profitable with an edge or competitive advantage over their rivals. In the order of consideration, the author first examines some insights from the literature and conceptual framework. Following the examination of the literature and the discussion of the conceptual framework, the researcher proposes research questions and related
73
International Journal of Computer Science and Network (IJCSN) Volume 1, Issue 6, December 2012 www.ijcsn.org ISSN 2277-5420
hypotheses that are tested using the Statistical Package of Social Sciences (SPSS).The research methodology is outlined, followed by the analysis of the data, and discussion. Finally, the conclusions are drawn and recommendations for further future research are made.
Availability ensuring that legitimate users are not unduly denied access to information and resources.
2. Literature framework
review
and
theoretical
This section presents the theoretical framework, developed from literature upon which the concepts or themes of the study were based. The review of related literature cantered mainly on the risks associated with doing business on the internet. It also makes an examination and review of information pertaining computer security, computer crime, and the structure of information systems.
74
International Journal of Computer Science and Network (IJCSN) Volume 1, Issue 6, December 2012 www.ijcsn.org ISSN 2277-5420
a workstation. According to Hurley [12] the unending string of data breaches and laptop thefts in recent months has shown, today`s threat landscape comprises far more than Distributed Denial of Service (DDoS) attacks, viruses and worms.
information security. This challenges the security as they may replace them with untrained ones who may compromise the security standards. Other human limitations increasing system vulnerability include complacency and carelessness, greed, and limited ability to understand complex systems.
2.1.10 Compliance
According to Layton [13] people in the organisation need to be informed of the risks that are possible in the organisation and the control measures put in place and also what is expected of them so that they can do it. If the responsible people in the organisation are aware of the risks and the control measures that have been implemented within their environment, it is reasonable to assume that there will be an environment of heightened awareness with the promise of fewer information security incidents [10].
75
International Journal of Computer Science and Network (IJCSN) Volume 1, Issue 6, December 2012 www.ijcsn.org ISSN 2277-5420
telecommunications providers with computers and switching centres in or nearby the World Trade Center (WTC) lost service and were stalled with busy signals for at least three days when the WTC and the Pentagon were destroyed on the morning of September 11, 2001. However, Merrill was able to resume its business later in the day. The firm did not suffer as much as others because it had redundant telecommunications capabilities and a rock-solid disaster recovery plan. Whether provided by the financial institutions or a third party, management should plan for recovery of critical ebanking technology and business functions and develop alternate operating processes for use during service disruptions [15].
platforms...computer failures, interruptions, and downtime can translate into disgruntled customers [1].
76
International Journal of Computer Science and Network (IJCSN) Volume 1, Issue 6, December 2012 www.ijcsn.org ISSN 2277-5420
3. Research Methodology
Questionnaires were distributed to 25 managers and 93 non managers from 8 commercial banks that offer ebanking services in Zimbabwe. Zimbabwe has 11 commercial banks; three of which currently do not offer e-banking services. The research used a quantitative descriptive method to assess the security of e-banking in a developing country. The samples used for the questionnaires were picked and chosen according to their years of experience and qualification in order to guarantee that they had the necessary computer skills and knowledge of e-banking. Evaluation and scoring of responses on the questionnaires is as shown in Figure 1 below.
`Scale 0 Responses Very Strongly Disagree Strongly Disagree Disagree Verbal interpretation Not implemented at all (1 chance in 100) Hardly ever the practice (1 chance in 10) Rarely the practice (2 chances in 10) Infrequent practice (3 chances in 10) Seldom practice (4 chances in 10) Neutral (5 chances in 10) Occasional practice (6 chances in 10) Often practice (7 chances in 10) Usual practice (8 chances in 10) Regular practice (9 chances in 10) Consistent Usual Practice (99 chances in 100) Mean Interval 0.00 0.50 0.50 1.50 1.51 2.50 2.51 3.50 3.51 4.50 4.51 5.50 5.51 6.50 6.51 7.50 7.51 8.50 8.51 9.50 9.51 10.00
respondents strongly or very strongly agreed that the elements of information security were adhered to.
4. Analysis of Data
The table 1, below shows the demographic characteristics of the respondents in terms of their work experience.
Cumulativ e Percentage 12.5 45.8 83.3 93.8 100.0
Frequenc y Vali d 0-3 years 46 years 79 years 10 12 years 13 years or more Total 6 16 18 5 3 48
Slightly Disagree Very Slightly Disagree Undecided Very Slightly Agree Slightly Agree Agree Strongly Agree Very Strongly Agree
4 5 6
7 8 9 10
Any score in the range of 8 and above were accepted as representing the perceptions of the respondents on the banks adherence to elements of information security in implementing e-banking in Zimbabwe as the respondents expressed that they agreed, whilst scores in the range of 7 and below were taken to mean that the elements of information security were not adhered to. The test value of 8.0 was therefore used as it is the minimum of the acceptable range of 8 to 10. Any values deviating from the test value were checked if they were on the upper or lower end. The higher values were showing that
The majority of the employees were experienced as the results show that at least 87.5% of the respondents had worked for the banks at least 4 years and they therefore understood better how the bank operates. They had observed at least how work is carried out, knew the organisational culture and at least knew the areas that are given greater priority in their organisations. This means that their responses were more accurate and a true representation of what actually takes place in their organisations. The table 2, below shows the demographic characteristics of the respondents in terms of their qualifications in information technology related training.
Frequen cy Valid None Certifica te level 12 19 11 6 48 Percenta ge 25.0 39.6 22.9 12.5 100.0 Valid Percenta ge 25.0 39.6 22.9 12.5 100.0 Cumulati ve Percentag e 25.0 64.6 87.5 100.0
77
International Journal of Computer Science and Network (IJCSN) Volume 1, Issue 6, December 2012 www.ijcsn.org ISSN 2277-5420 Diploma level Degree level Total Table 2: Information Technology related training
The rating or level of adherence to information security elements when implanting e-banking in Zimbabwe. As shown in the table 3 below, a t-test was performed using a test value of 8.0 at 5% level of significance. Any scores in the range of 8 and above were accepted as representing that the certain elements of information security were adhered to, whilst scores in the range of 7 and below were taken to mean that certain elements were not adhered to. The test value of 8.0 was therefore used as it is the minimum of the acceptable range of 8 to 10.
Table 3: e-banking security (One-Sample T-test)
According to the results from the table 2 above, at least 75% of the respondents had some qualification and training knowledge in information technology. However, only 12.5% had pursued it to degree level. Their expectations and their judgement when it comes to the security of information systems was therefore considered to be more reliable as they responded to things they at least have heard of or actually work with in their day to day activities.
Mean 8.5764 8.7639 7.0060 7.8194 7.2917 8.3083 8.2396 9.5521 8.0298 8.6458 8.0818
N = 48 Value = 8.0
df = 47
Test
Physical access controls Capital investment Logical access control Security of network services Behavioural security Security policy Human resources competence Organisational structure of IS department Compliance Risk management and disaster recovery plan Average security
Std. Deviation 1.57363 .90169 .89673 1.23763 1.81487 .64042 1.42821 .48641 .88258 1.00403 .62347
Std. Error Mean .22713 .13015 .12943 .17864 .26195 .09244 .20614 .07021 .12739 .14492 .08999
Sig. (2-tailed) .015 .000 .000 .317 .010 .002 .251 .000 .816 .000 .368
Mean Difference .5764 .7639 -.9940 -.1806 -.7083 .3083 .2396 1.5521 .0298 .6458 .0818
Any values deviating from the test value were checked if they were on the upper or lower end. The higher values were showing that respondents strongly or very strongly agreed that those particular elements of information security were adhered to. From the 10 dimensions that were tested, the following results were obtained: Physical access control has a high mean of 8.6 which meant that on average, the respondents strongly agreed that their banks have physical access controls put in place and the results show that the level of control is significantly higher than the minimum expected. Capital investment has a high mean of 8.8 and a low standard deviation of 0.9, meaning the respondents strongly agreed and all spoke with one voice that commercial banks are investing on the security of ebanking in Zimbabwe. There are backup communication networks, power sources and backup systems to ensure constant availability of information systems.
Logical access controls were significantly lower than expected with a mean of 7.0 and a standard deviation of 0.9 thus; respondents slightly agreed that logical access controls are put in place. There is a need for further analysis to find out which areas contributed to the low scores. Security of network services were not significantly different from, though lower than the mean, which meant that banks met the network security requirements as expected for e-banking services to be conducted. However, there is still a room for improvement as they stood at the mean of 7.8 with a standard deviation of 1.2 when the required mean was 8.0. Behavioural security stood at the mean of 7.3 and a high standard deviation of 1.8, which means most of the times banks fail to meet the requirements although on average, the respondents say often times the behavioural expectations are met. On this regard further analysis would need to be done in future.
78
International Journal of Computer Science and Network (IJCSN) Volume 1, Issue 6, December 2012 www.ijcsn.org ISSN 2277-5420
Security policy as revealed by statistics in the table 3 above, indicated that banks have security policies in place that are approved by management and are reviewed from time to time. This is supported by a mean of 8.3, which is significantly high and a very low standard deviation of 0.6, thus banks are generally meeting the expectations in that area. Human resources competence with a mean of 8.2 is though not significantly different, but higher than the test value of 8.0. Therefore, it can be concluded that the information technology staff at the banks are competent. Organisational structure of information systems department has a high mean score of 9.6 and a very low standard deviation of less than 0.5. The respondents all spoke with one voice to say that banks have directors responsible for information technology departments, and though working closely linked and working together, information technology department is clearly separated from information technology security department. Compliance is a very important element of information security. It is one thing to set policies and to have them followed is another. Banks as indicated by the results are managing to ensure that all sections of the organisation comply with the set standards and policies as shown by the mean of 8.01, which is not significantly different from the test value of 8.0. Thus, the respondents agree that compliance is ensured by the banks to make their ebanking services implementable.
Risk management and disaster recovery plan scored a significantly high mean of 8.6 with a very low standard deviation of 0.6 which meant that the respondents strongly agreed that commercial banks have disaster recovery plans and risk mitigation measures put in place to ensure critical information and information systems are available even in case of a disaster. Generally, the average security level that was computed yielded a high mean of 8.08 which is not significantly different from the desired test value of 8.0, and a low standard deviation of 0.62 which means that on average elements of information security are adhered to ensure that e-banking is implantable. Research Question: Is there a difference between the perceptions of management and non management on the adherence to elements of information security in implementing e-banking in Zimbabwe, as developing country? There are basically ten elements of information security that were measured to determine if the perceptions of management and non management differ with respect to elements of information security. Table 4, below shows the mean scores of management versus mean scores of non management for each element that was measured.
Table 4: Group statistics for perceptions
Current Position Physical access controls Capital investment Logical access control Security of network services Behavioural security Security policy Human resource competence Organisational structure of IS department Compliance Risk management and disaster recovery plan Average security Non management Management Non management Management Non management Management Non management Management Non management Management Non management Management Non management Management
Mean
30 18 30 18 30 18 30 18 30 18 30 18 30 18 30 18 30 18 30 18 30
8.2667 9.0926 8.5222 9.1667 6.7286 7.4683 7.4889 8.3704 6.8444 8.0370 8.4600 8.0556 7.9000 8.8056 9.6833 9.3333 8.1857 7.7698 8.8000 8.3889 7.9770
Std. Deviatio n 1.62452 1.37582 .93335 .69780 .84253 .80635 1.28872 .94204 1.86258 1.49897 .54114 .72536 1.44079 1.24722 .40436 .54233 1.02478 .49763 .92027 1.10926 .63089
.29660 .32428 .17041 .16447 .15382 .19006 .23529 .22204 .34006 .35331 .09880 .17097 .26305 .29397 .07383 .12783 .18710 .11729 .16802 .26146 .11518
74
International Journal of Computer Science and Network (IJCSN) Volume 1, Issue 6, December 2012 www.ijcsn.org ISSN 2277-5420 Non management Management Non management Management Non management Management Non management Management 18 8.2566 .58649 .13824
The table 4 above shows the average scores for all items computed and the values obtained. These were summarised in variable, average security. This average security was used to determine the overall responses of the respondents as to whether or not they perceived that the elements of information security are adhered to. According to the overall average score there is no significant difference between the perceptions of management and perceptions of non management.
[3] C. E. Lucier, and J. D Torsilieri, Analysing Requirements and Defining Microsoft.net Solution Architectures: E-Business Microsoft Corporation, Microsoft Press 2002. [4] K. E. Kendal, and J. E. Kendal, Systems Analysis and Design. 5th ed. India: Prentice Hall.2004. [5] Microsoft Corporation, Analysing Requirements and Defining Microsoft.net Solutions Architectures, Microsoft Press 2005. [6] J. A. O`Brien, Introduction to Information Systems. 12th ed. Irwin: McGraw-Hill 2005. [7] M. Stamp, Information Security: Principles and Practice. USA: John Wiley &Sons Inc 2005. [8] F. Warwick, Computer Communications Security: Principles, Standard Protocols and Techniques. USA: Prentice Hall. 1994. [9] T. R. Peltier, Information Security Risk Analysis. USA: Auerbach Publications 2001. [10] S. Petterson, Database and network, An International Journal of Database and network practice, vol.38, 2008, pp.9-23. [11] J. Forristal, Physical and Logical Security. Retrieved May 24, 2009, from Web site: http//www.networkcomputing.com/show Article.jhtml? Articled=194200006, 2006. [12] B. Hurley, Physical and Logical Security. Retrieved May 21, 2009, from Web site: http//searchsecurity.techtarget.com/news/article/0,28914 2,sid14-gci1241956,00.html, 2007. [13] T. P Layton, Information Security: Design, Implementation, Measurement, and Compliance. USA: Auerbach Publications 2006.
References
[1] K. C Laudon, and J. P Laudon, Management Information Systems-Managing the digital firm. 8th ed. India: Prentice Hall 2004. [2] Research and markets.com, The E-payments and Ebanking Market Outlook. Retrieved September 8, 2008, from Web site: http//www.researchandmarkets.com/theepayments-and-ebanking-market-outlook.htm.
74
International Journal of Computer Science and Network (IJCSN) Volume 1, Issue 6, December 2012 www.ijcsn.org ISSN 2277-5420
[14] S. Lomash, and P. A Mishra, Business Policy and Strategic Management. New Delhi, India: Vikas Publishing House (Pvt) Ltd 2005. [15] Federal Financial Institution Examination Council, IT Handbook InfoBase Booklet: E-Banking. Retrieved May 18, 2009, from W eb site: http://www.ffiec.gov/ffiecinfobase/booklets/e-banking/ebanking-03a-exam-points.htm 2009.
74