Dean's Office Official Model Question Paper

Answer: Empirical Research: Research based on direct observed and measured phenomena and verifiable empirical data rather than mere speculative theory or personal belief. Hallmark: Conclusions are derived from concrete evidence gathered through systematic observation or experimentation that can be independently verified and replicated.

Answer:

• Independent Variable ($X$): The variable that is manipulated, measured, or selected by the researcher to determine its relationship to an observed phenomenon (the presumed cause).
• Dependent Variable ($Y$): The variable that is observed and measured to determine the effect of the independent variable (the presumed outcome or effect).

Answer: Snowball Sampling: A non-probability sampling technique where existing study participants recruit future participants from among their acquaintances. When Employed: Used when studying rare, hidden, or hard-to-reach populations where no formal sampling frame exists (e.g., undocumented migrant workers, victims of corporate harassment, elite angel investors).

Answer: Cronbach’s Alpha ($\alpha$): A widely used statistical measure of internal consistency reliability among a set of survey Likert-scale items measuring a single underlying latent construct. Acceptable Threshold: A coefficient of $\alpha \ge 0.70$ is universally regarded as acceptable for academic research in business and social sciences.

Answer: The p-value is the probability of obtaining test results at least as extreme as the actual observed results, assuming the null hypothesis ($H_0$) is true:

• If $\mathbf{\text{p-value} \le \alpha}$ (e.g., $0.05$): Reject $H_0$ (Statistically significant).
• If $\mathbf{\text{p-value} > \alpha}$: Fail to reject $H_0$ (Insufficient empirical evidence).

1. Concept of Research Design

A Research Design is the master blueprint or comprehensive architectural framework specifying the methods and procedures for collecting, measuring, and analyzing data to answer the central research questions.

---

2. Comparative Matrix of the Three Major Research Designs

1. Hierarchy of Measurement Scales

                        Measurement Scale Hierarchy
                                     |
    [Nominal: Identity]  -->  [Ordinal: Order]  -->  [Interval: Distance]  -->  [Ratio: Absolute Zero]

1. Nominal Scale: Categorizes data into discrete, mutually exclusive qualitative groups with no quantitative meaning (e.g., Gender: $1 = \text{Male}, 2 = \text{Female}$; Bank type: Commercial, Development). Permissible statistics: Frequencies, mode, Chi-square.
2. Ordinal Scale: Ranks items in order of magnitude, but intervals between ranks are unknown and unequal (e.g., Brand rankings: $1^{\text{st}}, 2^{\text{nd}}, 3^{\text{rd}}$; Customer satisfaction: Low, Medium, High). Permissible statistics: Median, percentile, Spearman rank correlation.
3. Interval Scale: Possesses order and equal intervals between scale points, but lacks a true absolute zero (e.g., 5-point Likert agreement scale, temperature in Celsius). Permissible statistics: Mean, standard deviation, Pearson correlation, t-test, ANOVA.
4. Ratio Scale: Has all properties of interval scale PLUS a true, non-arbitrary absolute zero point (e.g., Annual revenue in Rs., age in years, inventory units). Permissible statistics: Geometric mean, regression analysis, all advanced parametric tests.

---

2. Why Scale Selection is Critical

Scale selection dictates the mathematical operations permissible on the data. Applying parametric tests (t-tests, ANOVA, linear regression) to nominal or ordinal data produces invalid statistical conclusions.

1. Probability vs. Non-Probability Sampling

---

2. Operational Procedures

A. Stratified Random Sampling

1. Divide the heterogeneous population into mutually exclusive, homogeneous subgroups (strata) based on a relevant criterion (e.g., bank size: Class A, B, C).
2. Establish a complete sampling frame within each stratum.
3. Draw an independent simple random sample from every stratum, either proportionally or disproportionately.
4. Benefit: Guarantees representation of minority subgroups and slashes sampling variance.

B. Systematic Sampling

1. Determine population size ($N$) and desired sample size ($n$).
2. Compute the sampling interval: $k = \frac{N}{n}$.
3. Select a random starting integer ($r$) between 1 and $k$.
4. Select every $k^{\text{th}}$ element thereafter: $r, r + k, r + 2k, \dots$
5. Caution: The sampling frame must not exhibit cyclical periodicity matching interval $k$, which introduces systematic bias.

1. Standard Structure of a Business Research Report

                             Structure of Research Report
                                           |
     +--------------------------+----------+--------------------------+
     |                          |                                     |
Preliminary Section        Body of the Report                    Supplementary Section
- Title Page               - Chapter 1: Introduction             - References (APA 7th)
- Declaration & Approval   - Chapter 2: Literature Review        - Appendices (Questionnaire)
- Table of Contents        - Chapter 3: Research Methodology     - Statistical Output
- Executive Summary        - Chapter 4: Results & Discussion     - Glossary
                           - Chapter 5: Conclusions & Recommendations

---

2. Essential Guidelines to Prevent Plagiarism

1. Explicit In-Text Citation: Every paraphrased concept, statistic, framework, or empirical finding from external authors must include immediate parenthetical citation (e.g., Kotler & Armstrong, 2021).
2. Proper Quoting: Enclose verbatim text in quotation marks and provide the exact page number.
3. Thorough Paraphrasing: Completely rewrite source ideas in one’s own voice and structure rather than superficially swapping a few synonyms (patchwriting).
4. Automated Similarity Screening: Run manuscripts through certified anti-plagiarism software (Turnitin, iThenticate) prior to submission, ensuring similarity indices remain safely below 15%.

Solution: Quantitative Econometric Interpretation

---

Part (a): Problem Statement, Conceptual Framework & Hypotheses (6 Marks)

1. Problem Statement: Despite explosive growth in digital payment transactions in Nepal, fintech applications face widespread user apprehensions regarding transaction failures and cyber fraud. Digital wallet developers lack empirical clarity regarding the relative importance of technical security features versus front-end visual simplicity in establishing long-term customer trust.

2. Conceptual Framework:

   [ System Security Features (X1) ] ----+
                                         |---> [ Consumer Trust (Y) ]
   [ UI Simplicity (X2) ] --------------+
   

3. Hypotheses:

   • $H_1$: System security features have a significant positive impact on consumer trust in digital wallets.
   • $H_2$: User interface simplicity has a significant positive impact on consumer trust in digital wallets.

---

Part (b): Multiple Regression Equation & Interpretations (4 Marks)

$$\hat{Y} = 1.80 + 1.15 X_1 + 0.65 X_2$$

• Intercept ($b_0 = 1.80$): When both security features ($X_1$) and UI simplicity ($X_2$) are rated zero, the baseline customer trust score is 1.80 points.
• Partial Slope $\beta_1 = 1.15$: Holding UI simplicity constant, every 1-unit increase in system security rating leads to an average increase of 1.15 points in consumer trust.
• Partial Slope $\beta_2 = 0.65$: Holding security features constant, every 1-unit increase in UI simplicity rating leads to an average increase of 0.65 points in consumer trust.

---

Part (c): Interpretation of $R^2$ (3 Marks)

$R^2 = 0.7744$ indicates that 77.44% of the total variance in consumer trust in digital wallets is explained by the combined linear effects of system security features and user interface simplicity. The remaining 22.56% of variance is attributed to other factors (brand reputation, customer support) or random error.

---

Part (d): Overall Model Significance ($F$-test) (3 Marks)

• $H_0: \beta_1 = \beta_2 = 0$ (Model has no explanatory power).
• $H_1:$ At least one $\beta_j \neq 0$.
• Calculated $F = 37.76$; Critical $F_{0.05, 2, 22} = 3.44$.
• Decision: Since $F_{\text{calc}} (37.76) > F_{\text{crit}} (3.44)$, we reject $H_0$ at the 5% significance level. Conclusion: The overall regression model is statistically highly significant.

---

Part (e): Individual Significance Tests & Fintech Strategy (4 Marks)

1. Test for Security Features ($X_1$):
   • $t_{\text{calc}} = 5.23, \quad t_{\text{crit}} = 2.074$
   • Since $|5.23| > 2.074$, reject $H_0$. Security features significantly influence trust.
2. Test for UI Simplicity ($X_2$):
   • $t_{\text{calc}} = 3.61, \quad t_{\text{crit}} = 2.074$
   • Since $|3.61| > 2.074$, reject $H_0$. UI simplicity significantly influences trust.

Strategic Recommendations for Fintech Firms:

• Security is the Dominant Anchor: Because $\beta_1 (1.15)$ is almost twice as influential as $\beta_2 (0.65)$, fintech executives must prioritize biometric authentication, 2-factor OTP encryption, and transparent real-time fraud monitoring above superficial visual animations.
• Streamlined Usability: Keep the user checkout journey clean and intuitive, eliminating cluttered pop-ups to reassure first-time digital users.