ISSN: 1204-5357

All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Evaluation of Government Employees Bank Selection Criteria in Hargeisa, Somaliland by Using AHP

Najah Aden Ismail

College of Economics and Management, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China

*Corresponding Author:
Najah Aden Ismail
College of Economics and Management, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
Tel: +252 634416750
Email: [email protected]

Visit for more related articles at Journal of Internet Banking and Commerce


This paper investigates bank selection criteria in Hargeisa, Somaliland. The motivation to realize this work comes from the fact that no many research works have been done in this area, to investigate this topic. This study aimed at exploring and understanding government employees’ relative preference of bank selection based on the four selected criteria, i.e. what criteria does the target group take into consideration to determine their bank choice. Therefore, the researcher opted for the use of Analytic Hierarchy Process (AHP) as a research methodology for collecting and analyzing the data because it is the most objective method to get mathematical way to measure data of a subjective nature. Results indicate that Bank’s Location is the most significant criterion that government employees give importance to. Thus, results got from this paper are reliable are valuable on bank preference and bank selection criteria. Hence, this paper’s results can be used in future research works.


Bank Selection Criteria; AHP; Saaty’s Nine Scale; Somaliland


Recently, almost all the bank managers give a great importance to customers’ needs. They try to understand customers’ preferences to select their bank, based on different criteria. This created an atmosphere of a growing competitiveness and hard work among the various banks trying to better their services to attract more clients and keep the old ones. Consequently, many researchers, like [1], are involved in trying to investigate and understand what criteria customers take into consideration in order to choose their best bank. However, through our literature review the researcher noticed that a great deal of research work, in the field, has been conducted in the US and some countries in Europe. But only few pieces of research refer to bank selection criteria in African countries. Although the previous work’s findings are of a great importance, it would be irrelevant to apply their results on other countries. This is, obviously, due to the different cultural, economic, political and social different backgrounds between those countries. However, to some extent, people who share the same work experience like government employees, might share some viewpoints about the bank selection criteria that can help them choose the bank they are going to deal with for example, trust, distance and speed of response.

Accordingly, the researcher thought of investigating the bank selection criteria in Hargeisa, Somali-land: The self-declared but internationally unrecognized republic of Somaliland in the horn of Africa. Hargeisa is the capital and the largest city in Somaliland with about 1,100,000 total population in 2017. Hargeisa is a city where business has started to flourish, private and governmental companies began to appear and entrepreneurship started to gain an important position especially in the field of food processing, import and export. A town with such an interesting economic growth and significant population rate must be a field of lots of economical projects and business companies. Consequently, a lot of money is being used on daily basis, hence, the necessity of good bank services will be needed.

Hargeisa, as explained previously, is a space where business started to gain importance, but it still contains a modest number of banks. This current study examined the case of the largest three commercial banks; Dahabshiil Bank, Premier Bank and Darasalaam Bank. First, Dahabshiil Bank, founded in 1970. Started as a small company that helps people to send money to friends and family but that has been developed through time to become one of the biggest African Banks. It is known for its national and international high-level services. Second, Premier Bank, a privately-owned commercial bank, founded in 2014. Offering multiple services such as, Retail Banking, Mobile and Internet Banking, smooth money transfer and so on. Third, Dara salaam which has been established in 2010, with its stable and dynamic banking system.

The study focused on the selection of criteria of government employees. The sample included one hundred government employees having bank accounts each in one of the previously stated private banks. This is due to the low quality of functionality of Somaliland’s Central Bank. Those employees prefer to use private banks because they offer better quality services and sometimes services that do not exist in the Central Bank. The following criteria have been investigated in our research; Location of the Bank., Service Provision., Bank Charges, Bank’s Reputation.

Throughout this study the researcher aimed at understanding and stating the characteristics that government employees take into consideration to select the bank they think is the best for them. By the best bank it means, a bank that can meet its clients’ needs in terms of the four criteria that have mentioned above.

Furthermore, this research made a comparison between the three banks selected as a sample study. The findings of this study give precise information about the services offered by those three banks.

Literature Review

This review briefly examines general factors that determine bank selection and specifically considers factors that attract employees to a bank.

Mainstream studies in bank selection are replete with studies on bank selection criteria employed by the general public. Due to the rapid changes in the economic, environment and the banking sector in particular that impact customers bank selection and patronage, how consumers select their banks, referred to in the literature as bank selection [2-5].

Few African studies have also been reported, which corroborate most of the mainstream findings. For instance, in a Nigerian study, safety of funds, efficient service, speed of transaction, and recommendation by relatives/friends were reported as factors that influence customers’ selection of banks in Nigeria [6], Ghanaian studies report proximity and accessibility [7] and technological factors, services factors, and bank image factors [8] as bank selection factors.

These and other more recent studies in Europe; see, e.g. [9] corroborate bank selection factors reported in most US studies. Among the factors reported are: convenient location [9], availability of ATM [9], friends/family/relatives or personal influence [9,10].

The provision of high-quality service positively affects the customer’s choice of a bank [11-13] Hence a bank with inferior service quality is unlikely to attract customers [11].

Lam and Burton [14] identified perceived service quality as a major selection criterion by SMEs in Hong Kong. Others have also found service quality to be an important determinant of building bank loyalty [15], Owing to its strong impact on business performance, lower costs and customer satisfaction, service quality has become a very important issue to researchers and business managers a like [16].

In the research, the bank selection criteria that was mostly considered by employees was centered on the bank’s location to either home or work.

Data and Methodology

In order to meet the purpose of this study, which is measuring the performance of the selected bank criteria. The study adopted the Analytical Hierarchy Process (AHP). AHP is a structured technique for organizing and analyzing complex decisions. This model was developed by Thomas L. Saaty. It has been massively studied and developed by many researchers after this.

As a first step, the researcher has met ten government employees from different departments as a focused group, discussed together the bank selection criteria relying on their own subjective judgments and personal points of view. After a long fruitful discussion, many criteria have been mentioned by the employees.

For more reliability, the researcher has carefully chosen the most recurrent and significant ones. Afterwards, four criteria have selected and designed a questionnaire that circles all.

In the questionnaire the respondents were making pairwise comparisons of each bank to every single criterion. The nine-point fundamental scale was used for the pair wise and comparisons. Detailed instruction on how to use the scale to complete a pairwise comparison of the criteria with respect to the objective of selecting the bank were given. Similar instructions were given to obtain the relative preference of a bank over another with respect to each criterion.

After collecting the data, this study has employed AHP model to analyze it because it is the most relevant model for making selecting different criteria subjective decisions in a mathematical way. Furthermore, the data analysis outcome of AHP model is quite accurate and confidential. In addition, the researcher has modeled the problem as a hierarchy, selected the criteria and alternatives and then drawn links between them in order to start the analysis.

Using the Scale mentioned earlier, the respondents’ judgements were first obtained. For example, if a respondent compares two attributes, Location (A1) and service provision coverage (A2) and indicates that A1 is strongly more important than A2, the value of A12 = 5 is assigned to this pairwise comparison. By definition, comparison of an attribute with itself results in a relative importance value of 1. In addition, the value of reciprocal comparison is the reciprocal of the relative importance value. Thus, with respect to the decision of selecting a bank, the pairwise comparison A matrix has two rows and two columns, and its elements are a11 = 1, a12 = 5, a21 = 1/5, a22 = 1.

According to Saaty (1990) Consistency index (CI) and consistency ration (CR) are used to check for consistency associated with the comparison matrix. When a positive reciprocal matrix of order n is consistent, the principal eigenvalue possesses the value n. Conversely, when it is inconsistent, the principal eigenvalue is greater than n and its difference will serve as a measure of CI. Therefore, to ascertain that the priority of elements is consistent, the maximum eigenvector or relative weights/λmax can be determined.

CI=(λmax – n)/n – 1. (1)

Where n is the matrix size or the number of items that are being compared in the matrix.

The consistency ratio (CR) can be determined as:

CR=CI/RI= [(λmax – n)/n – 1]/RI. (2)

Where RI represents average consistency index over a number of random entries of same order reciprocal matrices The CR is acceptable, if its value is less than or equal to 0.10. If it is greater than 0.10, the judgment matrix will be considered inconsistent.

To rectify a judgment matrix that is inconsistent, decision-makers‟ judgments should be reviewed and improved.

  Dahabshiil: 0.1867
  Premier: 0.1578
Bank’s Reputation: 0.0698 0.6555
  Dahabshiil: 0.115
  Premier: 0.4055
  Dara-Salaam: 0.4796


The Figure 1 shows the AHP- Hierarchy of bank selection. While the first level of the Hierarchy is the objective, the second level is the four criteria that have chosen and the third level is the three banks that we have implemented in this study.


Figure 1: Decision hierarchy of a bank selection.

In order to understand the prioritization of the service, characteristics that government employees deem important in bank selection decisions, the respondents’ judgments on the four criteria were examined. Table 1 reports the priority vectors for level 2 and level 3 of the decision hierarchy. The priorities of level 2 provide the relative important of the four criteria when choosing a bank. The priorities level 3 gives the relative preference of three banks with respect to each of the four criteria in level 2.

Attributes Level 2 priorities Banks Level 3 priorities
Service Provision coverage 0.2291 Dahabshiil 0.0782
Premier 0.4353
Dara-Salaam 0.4866
Location of the bank 0.6216
Dahabshiil 0.4353
Premier 0.0782
Dara-Salaam 0.4866
Bank charges 0.0798

Table 1: Priority aspects for the decision hierarchy.

Interpretation of Level 2 Priorities-Selection Criteria

Based on the priorities level 2, the two most important criteria affecting government employees’ bank selection decisions are location of the bank and service provision coverage, they are followed by bank charges and bank reputation. Specifically, the respondents considered the location of the bank over bank charges (Weight=0.6216) more than important as bank charges (0.0794), and Service provision (0.2291) over bank’s reputation (0.0698). The consistency ratio for this level was 0.02, within the acceptable range of 0.10 (Saaty,1988)

Interpretation of Level 3 Priorities-Decision Alternatives

To report the second objective determining government employees’ preferences of the banks with respect to each criterion considered, the level 3 findings were examined. From the priorities of the banks in Table 1. Dara-salaam is the most preferred bank for all the four criteria. These are Service provision (weight=0.4866), location of the bank (0.4866), Bank charges (0.6555) and bank’s reputation (0.4796). Overall, the priorities in Level 3 show that government employee’s relative preferences for banks are same over the four attributes.

Banking Behavior of the Respondents

The one hundred percent (100%) government employees had a total of one hundred-twenty (120) accounts in the banks. Seventy (70%) in Darasalaam, twenty (20%) in Premier Bank, and ten (10%) in Dahabshiil. The researcher found that Darasalaam Bank has the largest percentage of accounts and this is reflected in findings of this study about bank selection criteria. This study also revealed in the second position as Premier Bank then Dahabshiil in the last position.

A majority of the employees expressed that they have no intention in having new accounts in other banks, this is because of their satisfaction with their bank’s services, charges, bank image and especially bank’s location. Results show that the clients are loyal to their chosen banks.

In Table 2 below shows the consistency ratio of each attribute, as well the consistency ratio of total attributes. Since all of them are less than 0.10 it means that the judgment matrix is consistent and reliable.

Attributes C. Ratio
Service provision coverage 0.0121
Location of the bank 0.0121
Bank charges 0.0281
Bank’s Reputation 0.0280
Overall Consistency ratio 0.0866

Table 2: Consistency Ratio.


The purpose of this study, as mentioned above, was to investigate bank selection criteria among government employees which are the target population of our study. The above findings will be a reference in giving deeper insights on which criteria government employees take into consideration when choosing their preferred bank in Hargeisa, Capital of Somaliland. The study can be a guide that helps bank managers understand more about both needs and preference of clients and also help improve their services to meet those needs.

The study also serves as an evident and objective analysis of bank selection criteria that affords a rich background for future researchers in this field in all African countries. Other African countries can refer to this paper’s findings since they all mostly share a great deal of historical, geographical, social and economic criteria.

The researcher ended up gathering very accurate data about the target group using AHP model. This helped to retrieve subjective data and interpret it into numbers and percentages.


I would like to thank Derrick Dela Dumenu, Katani Emmanuel Kwadzo, Bouakkaz Amal for their support and for sharing their pearls of knowledge during the course of this research. I am also immeasurably grateful to Abdiaziz Mahmud Ismail for his comments on an earlier version of the manuscript, as well as his technical help throughout this study.