Statistics Basics: Data, Types, Scales & Uses

What is Statistics?

Statistics is the science of collecting, analyzing, interpreting, and presenting data to make informed decisions. It helps us understand patterns, trends, and relationships in data, whether in business, healthcare, or hospitality.

Why Study Statistics in Hotel Management?

In hospitality, statistics helps with:

• Guest satisfaction surveys (analyzing feedback).
• Inventory management (predicting demand for food/beverages).
• Revenue forecasting (seasonal trends in bookings).
• Quality control (tracking food waste or service delays).

Types of Statistics

1. Descriptive Statistics

• Purpose: Organizes, summarizes, and presents data in a meaningful way (e.g., tables, graphs, averages).
• Example: A hotel’s monthly sales report showing average room occupancy rates.
• Tools: Measures of central tendency (mean, median, mode), dispersion (range, standard deviation), and visualizations (bar charts, pie charts).

2. Inferential Statistics

• Purpose: Uses sample data to make generalizations about a larger population (e.g., predicting customer satisfaction across all hotels based on a survey of 100 guests).
• Example: Ncell analyzing call-drop rates in Kathmandu to improve network reliability nationwide.
• Tools: Probability, hypothesis testing, confidence intervals.

Data: The Raw Material

Data is the raw facts and figures collected for analysis. It can be classified in multiple ways:

1. By Source

2. By Nature

Measurement Scales: How Data is Measured

The level of measurement determines what statistical operations you can perform. There are four scales, ranked from lowest to highest precision:

Worked Example 1: Classifying Data Scales

Question: Classify the following data types by their measurement scale:

1. Hotel star ratings (1* to 5*).
2. Guest ages (25, 30, 45 years).
3. Customer feedback: "Delicious," "Average," "Poor."
4. Monthly revenue (NPR 500,000; 750,000).

Solution:

flowchart TD
  A["1. Hotel star ratings"] -->|"Ordinal"| B["Categories with order but no equal intervals."]
  C["2. Guest ages"] -->|"Ratio"| D["Numeric with true zero (can be divided)."]
  E["3. Customer feedback"] -->|"Ordinal"| F["Ordered categories (Delicious > Average > Poor)."]
  G["4. Monthly revenue"] -->|"Ratio"| H["Numeric with true zero (NPR 0 is meaningful)."]

In the Real World

1. eSewa (Nepal):

   • Idea Used: Descriptive Statistics (mean transaction value, peak usage hours).
   • How: eSewa analyzes monthly transaction data to identify trends (e.g., higher bill payments during salary days). This helps optimize server capacity and marketing.

2. Ncell (Nepal):

   • Idea Used: Inferential Statistics (sample surveys for network quality).
   • How: Ncell conducts call-drop rate surveys in 5 districts and uses the data to predict and improve coverage across Nepal. Example: If 15% of calls drop in Bhaktapur, they may upgrade towers there.

3. Daraz (Nepal):

   • Idea Used: Measurement Scales (nominal data for product categories).
   • How: Daraz classifies products into nominal categories (Electronics, Grocery, Fashion) to organize inventory. For promotions, they might analyze ordinal data (customer ratings: 1-star to 5-star) to identify best-selling items.

Worked Example 2: Real-World Application (Nepal Traffic Routes)

Scenario: The Kathmandu Metropolitan City (KMC) wants to reduce traffic congestion by analyzing data from key routes. They collect:

• Primary Data: GPS coordinates of vehicles at peak hours (8 AM–10 AM).
• Secondary Data: Historical accident reports from the Metropolitan Police.

Tasks:

1. Classify the data types.
2. Suggest a measurement scale for each.
3. Propose a visualization to present findings.

Solution:

1. Data Types:
   • GPS coordinates: Quantitative (Continuous).
   • Accident reports: Qualitative (Categorical).
2. Measurement Scales:
   • GPS coordinates: Ratio (distance has a true zero).
   • Accident reports: Nominal (types: collision, hit-and-run, etc.).
3. Visualization:

Key Terms and Definitions

Common Mistakes to Avoid

1. Confusing Primary and Secondary Data:

   • ❌ Using old sales reports (secondary) as if they were fresh surveys (primary).
   • ✅ Always check if secondary data matches your current needs.

2. Misclassifying Measurement Scales:

   • ❌ Labeling hotel star ratings as interval (they lack equal intervals between 1* and 2*).
   • ✅ Correct: Ordinal (ordered categories).

3. Ignoring the Context of Data:

   • ❌ Analyzing guest ages (ratio) with nominal data (e.g., "prefers breakfast/lunch/dinner").
   • ✅ Combine scales meaningfully (e.g., "Guests aged 25–35 prefer breakfast").

Exam Tip

How This Unit is Tested (TU Pattern)

1. Short Questions (5–10 marks):

   • Define statistics, population vs. sample, or types of data.
   • Example Question: "Differentiate between descriptive and inferential statistics with examples from the hospitality industry." Answer:

     | **Descriptive**               | **Inferential**               |
     |-------------------------------|-------------------------------|
     | **Purpose**: Summarizes data. | **Purpose**: Makes predictions. |
     | **Example**: Mean room rate at a hotel = NPR 8,000. | **Example**: Predicting 15% increase in bookings during Dashain. |
     | **Tools**: Tables, graphs.     | **Tools**: Hypothesis testing. |
     

2. Problem-Solving (10–15 marks):

   • Classify data types/scales or match real-world scenarios to statistical concepts.
   • Example Question: *"Classify the following into primary/secondary and qualitative/quantitative data, and suggest a measurement scale:
     1. Feedback forms from guests at the Himalayan Hotel.
     2. NTC’s monthly internet speed reports for 2023."* Solution:

     | **Data**                          | **Type**       | **Scale**       |
     |-----------------------------------|----------------|-----------------|
     | Guest feedback forms              | Primary, Qualitative | Ordinal        |
     | NTC internet speed reports        | Secondary, Quantitative | Ratio       |
     

3. Application-Based (10–20 marks):

   • Relate statistics to real hospitality scenarios (e.g., revenue analysis, customer feedback).
   • Example Question: *"A 5-star hotel in Pokhara collects the following data:
     • Guest nationalities (Nepali, Indian, Chinese).
     • Room tariffs (NPR 5,000, 7,500, 10,000).
     • Customer satisfaction ratings (1–5 stars). Classify each into the appropriate measurement scale and suggest one statistical tool to analyze each."* Solution:

     | **Data**               | **Scale** | **Statistical Tool**               |
     |------------------------|-----------|------------------------------------|
     | Guest nationalities    | Nominal   | Mode (most common nationality).    |
     | Room tariffs           | Ratio     | Mean/median tariff.                |
     | Satisfaction ratings   | Ordinal   | Median rating (avoids outliers).   |
     

Pro Tips for Full Marks

1. Use Real Examples:

   • Always tie definitions to Nepalese contexts (e.g., "Like how eSewa analyzes transaction data...").

2. Draw Diagrams:

   • For measurement scales, sketch a hierarchy (Nominal → Ordinal → Interval → Ratio) with arrows.

3. Memorize Key Differences:

   • Primary vs. Secondary Data: Think "firsthand" vs. "pre-existing."
   • Qualitative vs. Quantitative: "Quality" (words) vs. "Quantity" (numbers).

4. Practice Classification:

   • Create a table like the one above and fill it with 5–10 examples from your daily life (e.g., "Your phone’s battery percentage" = Ratio).