Research Problem Formulation: Questions, Objectives & Proposal Components

Core Concepts

1. What is a Research Problem?

A research problem is a clear, focused, and significant gap in knowledge or practice that requires investigation. It must be:

• Specific: Narrow enough to investigate within a reasonable timeframe.
• Feasible: Achievable with available resources (time, budget, data).
• Relevant: Addresses a real-world need (e.g., improving traffic efficiency in Kathmandu).
• Novel: Contributes new insights (not just repeating existing studies).

Why is it crucial?

• Guides the entire research process.
• Ensures the study is purposeful (not random).
• Helps secure funding (e.g., for a PU-funded engineering project).

2. Formulating Research Questions

A research question (RQ) is a precise, answerable query that drives the study. It must be:

• Clear: Avoids ambiguity (e.g., "How does AI improve Daraz’s delivery efficiency?" vs. "Can AI help?").
• Researchable: Data or methods exist to answer it.
• Significant: Matters to stakeholders (e.g., NTC, Pathao, or government).

Types of Research Questions

Worked Example: Ncell’s Network Optimization Problem: High latency in Pokhara during peak hours (5–8 PM). Research Question: "What factors contribute to increased latency in Ncell’s 4G network in Pokhara between 5 PM and 8 PM, and how can they be mitigated?" Why this works:

• Specific: Time (5–8 PM), location (Pokhara), technology (4G).
• Feasible: Ncell can provide network logs.
• Relevant: Directly impacts user experience.

3. Setting Research Objectives

Objectives are SMART (Specific, Measurable, Achievable, Relevant, Time-bound) goals derived from the research question.

How to Write Objectives

1. Start with an action verb: "To analyze," "To compare," "To evaluate."
2. Be precise: Avoid vague terms like "study" or "examine."
3. Include criteria for success: Quantify if possible.

Example for Daraz’s Delivery Route Optimization Research Question: "How can Daraz reduce delivery delays in Lalitpur by optimizing route planning?" Objectives:

1. To map the current delivery routes in Lalitpur using GIS tools.
2. To compare delivery times before and after implementing a dynamic routing algorithm.
3. To evaluate the cost savings from reduced fuel consumption and vehicle usage.

In the Real World

1. Pathao’s Ride-Hailing Algorithm

   • Idea Used: Causal research question + optimization objectives.
   • How: Pathao’s engineers asked: "How does real-time traffic data affect ride-matching success rates?" They set objectives to:
     • Reduce wait times by 30% using AI-driven route suggestions.
     • Compare performance in Kathmandu vs. Pokhara.
   • Outcome: Faster matches and higher driver satisfaction.

2. NTC’s Internet Speed Monitoring

   • Idea Used: Descriptive research question + data-driven objectives.
   • How: NTC’s study: "What are the average internet speeds in Nepal’s top 10 districts?" Objectives:
     • Measure speeds at 3-hour intervals for 3 months.
     • Identify districts with <50 Mbps speeds (e.g., Doti, Achham).
   • Outcome: Targeted infrastructure upgrades in low-speed areas.

3. Khalti’s Fraud Detection System

   • Idea Used: Comparative research question + hypothesis testing.
   • How: Khalti’s team asked: "Does two-factor authentication reduce fraudulent transactions by >50%?" Objectives:
     • Test fraud rates before/after implementing SMS + OTP verification.
     • Compare fraud rates across Khalti, eSewa, and IME Pay.
   • Outcome: Reduced fraud by 60%, saving millions in losses.

Visualizing Research Problem Formulation

Step-by-Step Process

flowchart TD
    A["Identify a Real-World Issue"] --> B["Define the Gap in Knowledge"]
    B --> C["Formulate a Research Question"]
    C --> D["Develop SMART Objectives"]
    D --> E["Write the Research Proposal"]
    E --> F["Conduct Literature Review"]
    F --> G["Design Methodology"]
    G --> H["Collect & Analyze Data"]
    H --> I["Draw Conclusions"]
    I --> J["Report Findings"]

Components of a Research Proposal

A well-structured proposal includes these core sections (visualized below):

mindmap
  root((Research Proposal))
    Title["1. Title (Clear & Concise)"]
    Introduction["2. Background & Justification"]
    LiteratureReview["3. Literature Review"]
    ResearchQuestions["4. Research Questions & Objectives"]
    Methodology["5. Methodology (How?)"]
      Methodology1["Data Collection"]
      Methodology2["Tools/Software"]
      Methodology3["Sampling Technique"]
    ExpectedOutcomes["6. Expected Outcomes"]
    Timeline["7. Timeline & Budget"]
    References["8. References"]

Key Components Explained:

IMAGE: Research Proposal Structure Diagram

Worked Example: Bank Loan Default Prediction

Problem: Commercial banks in Nepal (e.g., NMB, Global IME) lose billions annually due to loan defaults. Research Question: "Which customer attributes (income, credit history, employment status) most strongly predict loan defaults in Nepal’s commercial banks?" Objectives:

1. To collect loan default data from 5 major banks (2018–2023).
2. To analyze the correlation between default rates and attributes like income level and credit score.
3. To develop a predictive model using machine learning (e.g., logistic regression).
4. To compare model accuracy across different banks.

Proposal Outline:

flowchart LR
    A["Problem: High Loan Defaults"] --> B["RQ: Predict Defaults Using Customer Data"]
    B --> C["Objective 1: Data Collection"]
    B --> D["Objective 2: Statistical Analysis"]
    B --> E["Objective 3: Model Development"]
    E --> F["Outcome: 85% Accuracy Model"]

Common Mistakes to Avoid

Exam Tip

How to Score Full Marks in PU Exams

1. For "Explain in Detail" Questions:

   • Use the problem → question → objectives → methodology framework.
   • Example Answer Structure:

     "The research problem in [context] is [define]. The research question is [state]. Objectives are [list 2–3 SMART goals]. The methodology involves [data collection] and [analysis], ensuring [feasibility]."

2. For Short Notes (e.g., "Research Objectives"):

   • Define in one line, then give 2–3 bullet points with examples.
   • Example:

     Research Objectives: Clear, measurable goals guiding a study.

     • Must be specific (e.g., "Reduce delivery time by 20%").
     • Derived from research questions.
     • Used in proposals to justify the study.

3. For Real-World Applications:

   • Always tie to Nepal: Use Daraz, Ncell, Khalti, or NTC as examples.
   • Example:

     "Like Pathao uses real-time traffic data to optimize ride matching, a research problem on [topic] could apply similar techniques to [solve X]."

4. Diagrams Save Marks:

   • Draw flowcharts for processes (e.g., proposal structure).
   • Use tables for comparisons (e.g., types of research questions).

5. Avoid Plagiarism:

   • Paraphrase definitions (e.g., don’t copy-paste "A research problem is...").
   • Cite sources for theories (e.g., "As per Saunders et al. (2019), objectives must be...").

Practice Question with Model Answer

Question: "A bank wants to conduct research on customer satisfaction with digital banking. Formulate a research problem, research question, and two objectives."

Model Answer: Research Problem: Nepal’s digital banking adoption (e.g., NMB Online, SBI Yono) is growing, but customer satisfaction scores remain below 70% due to poor app usability and slow transaction speeds.

Research Question: "What specific features of digital banking apps (e.g., transaction speed, customer support, UI design) most influence customer satisfaction in Nepal?"

Objectives:

1. To survey 500 digital banking users across Kathmandu, Pokhara, and Lalitpur to identify key pain points.
2. To compare satisfaction scores between users of NMB Online, SBI Yono, and Global IME Mobile Banking.

Why This Works:

• Problem: Real-world issue (low satisfaction).
• Question: Specific (features → satisfaction).
• Objectives: Measurable (survey 500 users) and comparative.