Statistics Basics: Data Types, Collection & Presentation

TAKEAWAYS:

• Statistics is the science of collecting, analyzing, and interpreting data to make informed decisions.
• Data can be classified into four types: nominal, ordinal, interval, and ratio, each with unique properties.
• Primary vs. secondary data differ in collection methods and reliability.
• Presentation tools (tables, graphs) transform raw data into meaningful insights.
• Ethical considerations in data collection ensure fairness and accuracy.
• Real-world applications span from eSewa’s transaction analysis to Ncell’s customer segmentation.

1. What is Statistics?

Statistics is the science of collecting, organizing, analyzing, interpreting, and presenting data to draw meaningful conclusions. It helps businesses, governments, and researchers make data-driven decisions.

Why Study Statistics?

• Decision-making: Banks use loan default rates to approve loans.
• Quality control: Factories like Nepal Electricity Authority (NEA) monitor energy output consistency.
• Market research: Companies like Daraz analyze customer purchase patterns to optimize inventory.

2. Types of Data

Data can be classified based on measurement scale and source.

A. Classification by Measurement Scale

Worked Example 1: Classify the following data types:

1. Blood group (A, B, AB, O) → Nominal
2. Movie ratings (1-5 stars) → Ordinal
3. IQ scores (80, 100, 120) → Interval
4. Salary (Rs. 25,000, Rs. 50,000) → Ratio

B. Classification by Source

Worked Example 2: A business wants to study student spending habits in Pokhara.

• Primary Data: Conducting a survey among students.
• Secondary Data: Using Daraz’s sales reports.

Which is better?

• Primary is more accurate but costly.
• Secondary is faster but may lack specificity.

3. Data Collection Methods

A. Methods of Collecting Primary Data

graph TD
    A["Primary Data Collection"] --> B["Survey Method"]
    A --> C["Observation Method"]
    A --> D["Experimental Method"]
    B --> B1["Questionnaire"]
    B --> B2["Interview"]
    C --> C1["Direct Observation"]
    C --> C2["Indirect Observation"]
    D --> D1["Controlled Experiment"]

Worked Example 3: Nepal Rastra Bank (NRB) wants to study inflation.

• Best method: Survey (asking households about spending).
• Why not observation? People may hide purchases.

B. Methods of Collecting Secondary Data

graph TD
    A["Secondary Data Sources"] --> B["Published Sources"]
    A --> C["Unpublished Sources"]
    B --> B1["Government Reports"]
    B --> B2["Books/Journals"]
    C --> C1["Internal Company Data"]
    C --> C2["Databases"]

Worked Example 4: A startup wants to analyze Pokhara’s real estate market.

• Best secondary sources:
  • Nepal Rastra Bank (NRB) reports on property prices.
  • Daraz/Khalti transaction data.

4. Presentation of Data

Raw data is meaningless without organization and visualization.

A. Tabular Presentation (Frequency Tables)

A frequency table organizes data into categories with counts/frequencies.

Example: Age distribution of a class (from past exam):

Worked Example 5: Convert the following data into a frequency table: Heights (cm): 160, 165, 170, 160, 175, 165, 180, 160, 170, 175

B. Graphical Presentation

Graphs make data easy to understand at a glance.

1. Bar Graph

• Best for categorical data (nominal/ordinal).
• Example: Ncell vs. NTC customer complaints per month.

2. Pie Chart

• Shows proportions of a whole.
• Example: eSewa transaction types (electricity, phone, internet).

3. Histogram

• Shows frequency distribution of continuous data.
• Example: Age distribution of a class (from Worked Example 5).

4. Line Graph

• Shows trends over time.
• Example: Nepal’s GDP growth (2010-2023).

5. Real-World Applications

A. eSewa: Transaction Analysis

• Data Type: Ratio (transaction amounts in Rs.).
• Presentation: Bar graphs showing most/least used services.
• Decision: Allocate more resources to electricity payments (highest volume).

B. Ncell: Customer Segmentation

• Data Type: Ordinal (customer satisfaction: Poor/Fair/Good/Excellent).
• Presentation: Pie chart of satisfaction levels.
• Decision: Improve network speed in areas with "Poor" ratings.

C. Daraz: Inventory Management

• Data Type: Ratio (sales volume per product).
• Presentation: Histogram of best-selling items.
• Decision: Stock more electronics (highest demand).

D. NTC: Traffic Flow Analysis

• Data Type: Interval (time taken to travel routes).
• Presentation: Line graph of congestion over time.
• Decision: Optimize signal timings during peak hours.

6. Ethical Considerations in Data Collection

• Bias: Avoid leading questions in surveys.
• Privacy: Anonymize data (e.g., Khalti transaction records).
• Accuracy: Use reliable sources (e.g., CBS Nepal for population data).

Worked Example 6: A company wants to survey Pokhara’s middle-class income.

• Unethical: Asking "How much do you earn?" directly (may cause bias).
• Ethical: Using proxy data (e.g., "What’s your monthly rent?").

7. Common Mistakes to Avoid

1. Misclassifying data types (e.g., treating temperature as ratio).
2. Using the wrong graph (e.g., pie chart for trends).
3. Ignoring outliers (e.g., a single Rs. 10 million transaction skewing eSewa’s average).
4. Not labeling axes properly (leads to misinterpretation).

Exam Tip

What to Expect in TU Exams

1. Definitions:

   • Expect 2-3 marks for defining nominal vs. ordinal data or primary vs. secondary data.
   • Example:

     "Define interval data with an example." → Temperature in °C (no true zero).

2. Classification:

   • 3-5 marks for classifying given data types.
   • Example:

     "Classify the following: (a) Blood group (b) Movie ratings (c) Height (cm)."

3. Frequency Tables & Graphs:

   • 5-7 marks for converting raw data into tables/graphs.
   • Example:

     "Prepare a frequency table for the following data: 10, 12, 10, 15, 12, 10, 18."

4. Real-World Applications:

   • 3-5 marks for linking concepts to businesses (e.g., eSewa, Ncell, Daraz).
   • Example:

     "How does Ncell use secondary data to improve services?"

5. Ethical Questions:

   • 2-3 marks for discussing bias/privacy in data collection.
   • Example:

     "Why is anonymizing customer data important for Khalti?"

How to Score Full Marks

✅ Memorize definitions (nominal, ordinal, primary, secondary). ✅ Practice classifying data types (use past exam papers). ✅ Master frequency tables & graphs (bar, pie, histogram). ✅ Relate to Nepalese companies (eSewa, Ncell, Daraz, NTC). ✅ Avoid calculation errors (double-check frequencies).

Final Note: Statistics is not just numbers—it’s about telling stories with data. Whether it’s eSewa’s transaction trends or Ncell’s network performance, understanding data helps businesses (and students!) make smart decisions.

Good luck! 🚀