Data Collection Methods: Techniques, Tools & Applications

TAKEAWAYS:

• Data collection methods are classified into primary (original data) and secondary (existing data), each with distinct tools and applications.
• Primary methods include surveys, interviews, observations, and experiments—each suited for different research goals (e.g., surveys for large-scale opinions, observations for behavioral insights).
• Secondary methods rely on published sources (government reports, academic papers, company databases) but require careful evaluation for credibility and relevance.
• Ethical considerations (informed consent, anonymity, confidentiality) are critical in data collection, especially in sensitive research (e.g., customer privacy in e-commerce).
• Real-world tools: eSewa uses surveys to gauge user satisfaction, while Daraz employs observational data (click patterns) to optimize product placement.
• Worked example: A Nabil Bank loan approval process traces how secondary data (credit scores) and primary data (customer interviews) combine to assess risk.

1. Introduction to Data Collection Methods

Data collection is the systematic gathering of information to answer research questions or test hypotheses. Methods are broadly categorized into primary (firsthand data) and secondary (pre-existing data). The choice depends on:

• Research objectives (exploratory vs. confirmatory).
• Budget and time constraints.
• Data availability (e.g., NEPSE stock data vs. custom surveys).

Primary vs. Secondary Data: Key Differences

mindmap
  root((Data Collection Methods))
    Primary Data
      "Collected firsthand by researcher"
      "Highly relevant but costly/time-consuming"
      Types["Surveys", "Interviews", "Observations", "Experiments"]
    Secondary Data
      "Existing data from other sources"
      "Cost-effective but may lack specificity"
      Sources["Government reports", "Academic journals", "Company databases", "News archives"]

2. Primary Data Collection Methods

Primary data is original, tailored to the research, and collected directly from respondents or sources. Common techniques:

A. Surveys

Definition: Structured questionnaires administered via mail, phone, online, or face-to-face to gather quantitative or qualitative data. Tools:

• Closed-ended questions (e.g., "Rate eSewa’s app: 1-5").
• Open-ended questions (e.g., "What challenges do you face using Khalti?").
• Likert scales (e.g., "Strongly disagree" to "Strongly agree").

Worked Example: NTC Customer Satisfaction Survey

1. Objective: Measure NTC’s service quality in Kathmandu.
2. Method: Online survey (Google Forms) with 500 respondents.
3. Questions:
   • Closed: "How often do you experience call drops? (Never/Rarely/Often/Always)."
   • Open: "Suggest improvements for NTC’s customer service."
4. Analysis: 60% reported "Rarely" for call drops; open responses highlighted slow complaint resolution.

Advantages/Disadvantages:

B. Interviews

Definition: One-on-one or group discussions to gather in-depth qualitative data. Types:

• Structured: Fixed questions (e.g., bank loan officer interviews).
• Unstructured: Open-ended (e.g., exploring why customers abandon Daraz carts).
• Focus groups: 6–10 participants discussing a topic (e.g., Himalayan Java’s new product ideas).

Worked Example: Daraz Seller Interview

1. Objective: Understand challenges faced by small sellers on Daraz.
2. Method: Semi-structured interview with 10 sellers in Pokhara.
3. Key Questions:
   • "What are your top 3 difficulties selling on Daraz?"
   • "How does Daraz’s logistics support compare to local competitors?"
4. Findings: 80% cited high logistics costs as a barrier; led to Daraz’s "Seller Protection Fund."

Advantages/Disadvantages:

C. Observations

Definition: Systematically recording behavior without direct interaction. Used when:

• Respondents may lie (e.g., self-reported vs. actual Khalti app usage).
• Studying natural behavior (e.g., customer traffic in a Nabil Bank branch).

Types:

• Participant observation: Researcher joins the group (e.g., shadowing a Daraz delivery agent).
• Non-participant observation: Researcher remains detached (e.g., recording checkout times at a supermarket).

Worked Example: Kathmandu Traffic Flow Study

1. Objective: Identify bottlenecks in Thapathali traffic.
2. Method: Non-participant observation at peak hours (7–9 AM).
3. Data Collected:
   • Vehicle types (bikes, cars, buses).
   • Average wait times at signals.
   • Pedestrian crossing patterns.
4. Findings: 40% delay caused by unregulated bike lanes; led to NTC’s pilot "smart traffic lights."

Advantages/Disadvantages:

D. Experiments

Definition: Manipulating one variable to measure its effect on another (e.g., A/B testing in digital marketing). Example in Nepal:

• Nepal Telecom (NTC): Tested a discounted night-time data plan in Bhaktapur vs. regular pricing. Result: 30% higher usage during night hours → rolled out nationally.

Key Steps:

1. Hypothesis: "Offering a 50% discount on night data will increase usage."
2. Control group: Bhaktapur (no discount).
3. Experimental group: Lalitpur (discount applied).
4. Measure: Data usage before/after.

Advantages/Disadvantages:

3. Secondary Data Collection Methods

Secondary data is pre-existing and sourced from internal or external records. Common sources:

• Internal: Company databases (e.g., Daraz’s sales records), employee reports.
• External: Government (NPC census data), academic (Nepal Rastra Bank reports), commercial (Nielsen consumer trends).

Worked Example: Nabil Bank Loan Approval

1. Objective: Predict loan default risk for SMEs.
2. Secondary Data Used:
   • Credit bureau reports (past loan repayment history).
   • NPC economic indicators (GDP growth, inflation).
   • Internal data (applicant’s savings, business plan).
3. Analysis: Combined with primary data (interviews with loan officers), Nabil Bank developed a risk-scoring model reducing defaults by 20%.

Advantages/Disadvantages:

4. Ethical Considerations in Data Collection

Ethics ensures respect for participants and data integrity. Key principles:

1. Informed Consent: Participants must know the purpose, risks, and voluntary nature of data collection.
   • Example: eSewa’s user surveys include a checkbox: "I consent to participate."
2. Anonymity/Confidentiality:
   • Anonymity: Data cannot be linked to individuals (e.g., NTC’s call drop surveys).
   • Confidentiality: Data is secure (e.g., Nabil Bank’s customer financial records).
3. Avoiding Harm: No deception or coercion (e.g., Daraz’s seller interviews avoid pressuring responses).
4. Data Privacy Laws: Compliance with Nepal’s Data Privacy Act (2018) and GDPR (for global studies).

Worked Example: Khalti’s User Data Policy

• Ethical Issue: Khalti collects transaction data for fraud detection.
• Solution:
  • Users opt-in during registration.
  • Data encrypted and stored securely.
  • Transparent privacy policy on their website.

5. Choosing the Right Method

Selecting a method depends on:

Decision Flowchart:

flowchart TD
    A["Research Objective"] --> B{"Is data specific to your study?"}
    B -->|"Yes"| C["Use Primary Data"]
    B -->|"No"| D["Use Secondary Data"]
    C --> E{"What’s your budget?"}
    E -->|"High"| F["Surveys/Interviews"]
    E -->|"Low"| G["Observations"]
    D --> H{"Is data reliable?"}
    H -->|"Yes"| I["Proceed"]
    H -->|"No"| J["Collect Primary Data"]

6. Real-World Applications in Nepal

A. eSewa: Survey-Driven Improvements

• Method: Annual online surveys (primary) + transaction data (secondary).
• Example: After 2022 surveys revealed 50% of users struggled with QR payments, eSewa introduced biometric authentication and simplified the app interface.

B. Daraz: Observational Data for Logistics

• Method: Clickstream data (primary) + warehouse movement tracking (secondary).
• Example: Observed that 70% of abandoned carts occurred at checkout → Daraz added one-click payments and guest checkout options.

C. Nabil Bank: Hybrid Approach for Loan Risk

• Primary: Customer interviews (qualitative).
• Secondary: Credit bureau reports (quantitative).
• Outcome: Reduced loan default rates by 15% through predictive modeling.

D. NTC: Experimental A/B Testing

• Method: Tested two pricing models for data plans in different regions.
• Result: Night-time discounts in Lalitpur increased usage by 35%, leading to a nationwide rollout.

7. Common Pitfalls and How to Avoid Them

8. Case Study: Himalayan Java’s Market Research

Objective: Launch a new organic instant coffee in Nepal. Methods Used:

1. Primary:
   • Surveys: 1,000 consumers on taste preferences (closed-ended + open-ended).
   • Focus groups: Discussions with urban vs. rural consumers.
2. Secondary:
   • NPC import-export data (competitor coffee imports).
   • Nepal Rastra Bank reports (disposable income trends). Findings:

• Urban consumers preferred sweetened, creamy coffee.
• Rural areas favored strong, bitter brews. Action: Developed two variants (Urban Blend, Rural Roast) with local distribution partnerships.

Exam Tip

1. Define Clearly: Always start with definitions (e.g., "Primary data refers to...").
2. Compare Methods: Exams often ask to contrast surveys vs. interviews or primary vs. secondary data. Use tables for clarity.
3. Real-World Links: Relate answers to Nepali companies (e.g., "Like Daraz, a business could use observational data to...").
4. Ethics: Expect short-answer questions on informed consent or confidentiality.
5. Worked Examples: Practice one full example (e.g., designing a survey for NTC or an interview for Nabil Bank).
6. Diagrams: Draw flowcharts for research processes or tables for method comparisons.

Common Exam Questions:

• "Differentiate between structured and unstructured interviews with examples from Nepali businesses."
• "How would you collect data to study ‘customer satisfaction with Pathao’s delivery delays’? Justify your choice."
• "Discuss ethical issues in secondary data collection, using a case from eSewa or Khalti."

Visual Summary:

mindmap
  root((Data Collection Methods in Business Research))
    Primary Data
      Surveys["eSewa feedback forms"]
      Interviews["Nabil Bank loan officer talks"]
      Observations["Daraz warehouse traffic"]
      Experiments["NTC’s night-data discount test"]
    Secondary Data
      Internal["Daraz sales databases"]
      External["NPC census reports"]
    Ethical Considerations
      Consent["eSewa’s opt-in surveys"]
      Confidentiality["Nabil Bank’s encrypted records"]
    Choosing Methods
      Budget["Low? Use secondary data."]
      Time["Fast? Use existing reports."]
      Relevance["High specificity? Primary data."]