III.3.2 Sampling definition
The survey will be conducted to sixteen (16) commercial banks
located in the capital city of the DRC (Kinshasa), the Central Bank of the DRC
and the three (3) main mobile operators offering m-payment services.
Commercial banks are those implementing e-payment systems and
offering e-payment services to customers while mobile operators are
implementing m-payment systems and offering m-payment services. On top of these
two organizations, the Central Bank is acting as the monetary authority in the
country and de facto regulator of the banking system in the country and
therefore in charge of defining e-payments regulation and rules.
Page | 39
For the commercial banks and the mobile operators sectors, the
sample will consist of:
> 1 C-level member (CEO or COO)
> 1 IT director or IT manager, the head of the IT
department
> 1 IT security officer within the IT department
> 1 e-payment or m-payment business system manager or sales
manager
> 1 Information system auditor
For the central bank, as the guarantor of the financial
regulation, answers will be provided by:
> 1 C-level or board member (Governor or Managing director)
> 1 Director in charge of regulation
> 1 Director in charge of operations
> 1 IT Director
> 1 IT Security officer
> 1 IS Auditor
This sampling definition ensures that all the existing
commercial banks in the country will participate in the survey and therefore
constitute the whole population of possible informants for this research. The
size of the sample for which we are expecting respondents for this survey is
101.
III.3.3 Data Collection method
For this research, the survey is the method of the data
collection.
Questionnaire uses closed-ended questions and checklists will use
likert rating scale which will provide all alternatives of the response.
The likert rating scale of 5 has been used for the alternative
responses in the way that 5 indicates the best option and 1 the worst option:
5=strongly agreed, 4=agree, 3=neutral, 2=disagree, 1=strongly disagree.
Surveys have been sent by email through the traditional mail
system (hands delivery to the reception) with an acknowledgement of the receipt
of the questionnaire.
For this reason, questions have been clearly explained to
allow self-explanation to the respondent. However, for the CEO and the IT
officer, some interviews have been requested to clarify particularity of
questions which have sensitive perception for them.
Page | 40
III.4. Data analysis method III.4.1 Introduction
Data collected through the survey have to be organized in such
way to be used by descriptive statistics tools because the research method used
is quantitative.
The first step consists of grouping data into tables;
therefore for this research the frequency distribution is the best method for
the organization of data collected.
The research question focuses on characteristics of e-payment
systems which are meaningful for its security. This limitation influences the
way collected data will be used to assess the security of e-payment systems.
In the literature review chapter, the research presented
elements that are likely to be considered in order to determine if the
information system on which e-payment system is built is compliant to the CIA
concept of this research.
Therefore, knowing the number of organizations involved in the
survey, the better way to represent the existence of specific element is the
frequency distribution. For example, the analysis will provide such information
like how many banks have firewall protecting their e-payment systems. This
information will be better represented in a frequency distribution.
The frequency distribution tables for some elements of the
survey will also be represented graphically using bars. Bar graphs will be used
instead of histogram because for this research responses collected through the
survey are «qualitative variable» which mean «categorical
variable for which each value represents a discrete category» (Sherril L.
Jackson, 2010, p218).
Page | 41
3.4.2 Data analysis technique
The descriptive statistics measures are most suitable to
analyze data for this research. Prior to mention which analysis technique is
used for this research, an overview of the three (3) main measures of central
tendency; mean, median and mode, is given in the table below.
Table 2: Types of central tendency measures
Source: Sherril
L. Jackson (2010, p225)
|
Mean
|
Median
|
Mode
|
|
Definition
|
The arithmetic average
|
The middle score in a distribution of scores organized from
highest to lowest or lowest to highest
|
The score occurring with greatest frequency
|
|
Use with
|
Interval and ration data
|
Ordinal, interval, and ratio data
|
Nominal, ordinal, interval or ratio
|
|
Cautions
|
Not for use with distribution with a few extreme score
|
|
Not q reliable measure of central tendency
|
For this research, data collected use the likert rating scale
with values ranging from 1 to 5; 5 being the best choice.
According to this rating, this research will use the mode as
data analysis technique because it is the appropriate method to assess the CIA
capability of the e-payment system by indicating the score occurring with the
highest frequency.
For example, if the survey returns that among sixteen (16)
banks 10 of them has a firewall in place, this information is more relevant
than the mean or the median.
However, to some extend, the mean can be used to define the
average of some capabilities according to the data collected and the element
which is assessed.
For example, to assess the existence of data backup capability
globally is the sample, it can be easier to say that this capability exists on
an average of 65%. This calculation uses the mean as data analysis technique to
assess the overall capability.
While the mode will be used to assess individual capability,
conclusion in the context of assessment of all the e-payments systems studied
will use the mean to state the average of organizations having such capability
implemented.
Page | 42
Chapter 4: Research findings
| 
