A Statistical Analysis about Mobile Banking Applications

A study was conducted to understand the strengths and weaknesses of mobile banking applications in Sri Lanka through a survey and statistical analysis.

Background Tran Mau Tri Tam on Unsplash

The survey contained 32 questionnaires, and I received 210 responses from all around the country (Sri Lanka) to avoid bias. Survey responses were analyzed using R, and hypotheses testings and confidence intervals (CI) were carried out. Banking industries can use these results to fine-tune their mobile banking applications.

Mobile banking refers to the use of a mobile device to carry out financial transactions. It has become more accessible with the development of mobile applications. In Sri Lanka, mobile banking is still an emerging technology that hasn’t reached its pinnacle yet. Based on our knowledge from research conducted worldwide, I ran a study locally through a survey (An online survey due to the COVID-19 pandemic). The survey included demographic questions (gender, age, occupation, etc.) and mobile banking related questions.

Distribution of respondents in a heatmap

Pilot runs on the survey was conducted with 15 individuals. These people were from different backgrounds and genders. I had to modify a few questions and answers based on the feedback and responses to meet my study objective. Then the survey went live all around the island.

Descriptive Analysis

The below figure depicts the breakdown of respondents as per the demographic questions.

Respondents examination in different perspectives

• 75% of the respondents were male.
• 84% of the respondents were above 25 years of age.
• 68% of the respondents have at least one degree.
• 87% of the respondents are employed.

What banks were the respondents using primarily?

Findings from this study are primarily applicable to Commerical bank, Bank of Ceylon, HNB, Sampath Bank and People’s Bank. Most of the respondents were the account holders of the mentioned banks.

Why are respondents using the mobile bank?

Above 80% of the respondents use mobile banks for transactions, balance inquiries, and pay utility bills.

Statistical Inference

Statistical inference is the process of concluding an underlying population based on a sample or subset of the data. In most cases, it is not practical to obtain all the measurements in a given population.

The null (H0) and alternative (Ha) hypotheses are mutually exclusive statements about a population. The hypothesis test uses sample data to determine whether to reject the null hypothesis. Differences in proportion and confidence intervals are primarily used techniques to analyze the hypothesis.

Does age have an impact on mobile bank usage?

H0: People below age 35 and above 35 use the mobile bank app in the equal proportion
Ha: The proportion of people below age 35 who uses mobile bank is more than the proportion of people above 35 who uses the mobile bank

I have used differences in proportions for this test.

Z-value - 3.799
P-value - 0.000073

There is significant evidence with 95% confidence that the proportion of people under 35 years old using mobile banking applications is more than that of those above 35.

Does monthly income have an impact on mobile bank usage?

H0: All the people in different income bucket use the mobile bank app in the equal proportion
Ha: Proportion of people with income >LKR 70000 uses the mobile bank more than the people with income <= LKR 70000

I have used differences in proportions for this test.

Z-value - 1.699
P-value - 0.0446

There is evidence with 95% confidence that the proportion of people with income higher than LKR 70000 using mobile banking applications is more than those with an income lower than LKR 70000.

What proportion of account holders uses mobile banking applications?

Following are the 95% confidence intervals (CI):

• Commercial bank — between 77.25% and 91.75% of Commercial Bank users use mobile banking applications.
• Bank of Ceylon — between 68.78% and 86.82% of Bank of Ceylon users use mobile banking applications.
• HNB — between 69.42% and 90.58% of HNB users use mobile banking applications.

What makes users unhappy about mobile banking?

Following are the 95% confidence intervals:

• Between 16.13% and 28.67% of users are unhappy or neutral about the application’s transaction capability.
• Between 11.42% and 22.78% of users are unhappy or neutral about the application’s balance inquiry feature.
• Between 46.65% and 61.55% of users are unhappy or neutral about the sudden crashes of the mobile banking application.
  The percentage shows that at least half of the mobile banking users are experiencing sudden crashes in mobile banking applications, and they are unhappy or neutral about that.

What authentication methods are preferred in mobile banking?

Following are the 95% confidence intervals:

• Between 54.55% and 69.05% of users prefer passwords as the authentication method.
• Between 49.05% and 63.95% of users prefer passwords with 2FA.
• Between 34.35% and 49.25% of users prefer fingerprint as the authentication method.
• Between 9.41% and 19.99% of users prefer face unlock for authentication.

What are the significant challenges in using mobile banking applications?

Following are the 95% confidence intervals:

• Between 29.85% and 44.35% of users see a lack of cellular coverage as an obstacle to using the mobile banking app.
• Between 21.94% and 35.66% of users don’t feel secure when using the mobile banking app.
• Between 18.23% and 31.17% of users see the lack of required features in the app as a demotivating factor.
• Between 12.92% and 24.68% of the users doesn’t feel the mobile banking app is user-friendly.
• Between 9.81% and 20.79% of users feel that the instability of mobile banking app is a struggle.

Summary

• Banks in Sri Lanka need to spread awareness about mobile banking applications among people above 35 years old, non-private employees, and with a monthly income less than LKR 70000.
• Mobile banking application developers need to focus on application stability and reliability.
• The management needs to invest in UX engineers to revamp the mobile banking experience with the seamless transaction and balance inquiry capabilities.
• Almost half of the mobile banking application users population prefer fingerprint authentication methods. Therefore, banks that don’t support fingerprints yet can focus on it.
• Since 22–37% of mobile banking users feel insecure when using mobile banking applications (95% CI), the bank associates need to spread awareness and provide assurance (e.g. security certifications on the application) to use the mobile banking application without any concerns.

Happy Coding! ✌️
- Vithulan