Data Collection & Presentation: Methods, Tables, Graphs & Ethics

---

Core Concepts: Why Data Collection and Presentation Matter

Statistics begins with data—raw facts and figures that need structure before analysis. This unit covers:

1. Sources of data (primary vs. secondary)
2. Methods of collection (surveys, experiments, observation)
3. Data organization (tables, classification, tabulation)
4. Graphical presentation (bar charts, pie charts, histograms, line graphs)
5. Ethical considerations (bias, confidentiality, misrepresentation)

Why this matters for hotel management? Hotels rely on guest feedback (surveys), inventory data (secondary sources), and operational metrics (experiments) to improve service. Poorly presented data leads to wrong decisions—e.g., misinterpreting low occupancy rates as poor marketing when the issue is seasonal demand.

---

1. Sources of Data: Primary vs. Secondary

Data comes from two main sources, each with trade-offs in cost, time, and reliability.

Primary Data

Collected firsthand for a specific purpose. Examples:

• Surveys/Questionnaires (guest satisfaction forms)
• Experiments (testing new menu items on a sample group)
• Observation (counting foot traffic in a hotel lobby)

Advantages: ✔ Highly relevant to the study. ✔ Control over data quality (no outdated info). ✔ Can be tailored to specific needs.

Disadvantages: ✖ Expensive and time-consuming. ✖ Risk of bias (e.g., survey questions may lead respondents).

Secondary Data

Already exists and is reused. Examples:

• Government reports (Nepal Tourism Board statistics).
• Hotel chain databases (Marriott’s global occupancy rates).
• Academic research (studies on customer loyalty in hospitality).

Advantages: ✔ Saves time and money. ✔ Broader scope (e.g., national tourism trends).

Disadvantages: ✖ May not fit your exact needs (e.g., old data on Kathmandu traffic). ✖ Quality depends on the source (e.g., a biased blog vs. a peer-reviewed study).

---

Worked Example 1: Choosing Data Sources for a Hotel A 5-star hotel in Pokhara wants to analyze guest satisfaction.

• Primary Data: Conduct a survey (500 guests) on cleanliness, service, and food.
• Secondary Data: Use past reservation data (from the hotel’s PMS) to compare satisfaction scores with booking trends.
• Why both? Primary data answers why satisfaction is low (e.g., slow room service), while secondary data shows when it drops (e.g., during peak season).

---

2. Methods of Data Collection

How you collect data affects its accuracy, completeness, and bias.

A. Survey Method

• Tools: Questionnaires, interviews, online forms (Google Forms, Typeform).
• Types:
  • Structured: Fixed questions (e.g., "Rate our breakfast on a scale of 1–5").
  • Unstructured: Open-ended (e.g., "What did you dislike about your stay?").
• Pros: Flexible, large sample sizes possible.
• Cons: Response bias (e.g., guests may lie to appear polite).

Example for Hotels:

• eSewa’s "Guest Feedback" system uses structured surveys to rank hotels on cleanliness, Wi-Fi, and staff friendliness.
• Worked Example 2: Designing a Survey for a Guesthouse A small guesthouse in Kathmandu wants to improve breakfast service. Design a 5-question survey:
  1. Structured: "How would you rate the variety of breakfast options?" (1–5 scale).
  2. Structured: "Did you finish your meal within 30 minutes?" (Yes/No).
  3. Unstructured: "What’s one thing we could improve about breakfast?"
  4. Structured: "Would you recommend our breakfast to others?" (Likert scale: Strongly Disagree → Strongly Agree).
  5. Structured: "How likely are you to return?" (1–10 scale, used by hotels globally).

---

B. Experimental Method

• How it works: Introduce a variable and measure its effect.
• Example for Hotels:
  • A/B Testing: A hotel tests two breakfast menus (Menu A vs. Menu B) on two identical floors and compares guest satisfaction scores.
  • Result: If Menu B gets higher ratings, it’s adopted hotel-wide.

Worked Example 3: Experiment to Reduce Check-in Time A hotel notices long queues at check-in. They test:

• Control Group: Traditional counter check-in (average wait: 12 minutes).
• Experimental Group: Self-check-in kiosks (average wait: 4 minutes).
• Conclusion: Kiosks reduce wait time by 66%, justifying investment.

---

C. Observation Method

• How it works: Watch and record behavior without interfering.
• Example for Hotels:
  • Foot Traffic Analysis: Count how many guests use the hotel’s gym vs. pool at different times.
  • Staff Behavior: Observe if waiters spend too much time at tables (slowing service).

Worked Example 4: Observing Guest Behavior in a Café A café notices low sales at 3 PM. An observer records:

• Insight: The café could offer discounts at 3 PM to attract more guests.

---

3. Organizing Data: Tables and Tabulation

Raw data is useless until organized. Tabulation means arranging data into rows and columns for clarity.

Key Terms:

• Variable: What’s being measured (e.g., guest age, room rate).
• Class Interval: Groups for continuous data (e.g., "20–30 years," "30–40 years").
• Frequency: Number of times a value appears.

Steps to Tabulate Data:

1. Identify variables (e.g., guest nationality, room type booked).
2. Classify data into logical groups (e.g., "Domestic," "International").
3. Count frequencies (e.g., 50 domestic, 30 international).
4. Calculate percentages (e.g., 62.5% domestic).

Worked Example 5: Tabulating Guest Nationalities Data collected from 100 guests: Raw Data: IND, NEP, USA, FRA, NEP, IND, AUS, NEP, IND, USA, ... Tabulated Data:

---

Types of Tables:

1. Simple Table: One variable (e.g., room occupancy by day).
2. Frequency Distribution: Shows how often values occur (e.g., guest ages in 10-year groups).
3. Cross-Tabulation: Two variables (e.g., "Guests who booked via Daraz vs. those who didn’t, by nationality").

Example for Hotels: A hotel cross-tabulates:

• Variable 1: Booking method (Online vs. Walk-in).
• Variable 2: Guest satisfaction score (1–5). Result:

  Booking Method	Satisfaction Score 1–2	3–4	5	Total
  Online	5	20	75	100
  Walk-in	15	30	55	100
  Insight: Online bookings have higher satisfaction (75% score 5) vs. walk-ins (55%).

---

4. Graphical Presentation of Data

A picture is worth 1,000 numbers. Hotels use graphs to:

• Spot trends (e.g., peak booking months).
• Compare performance (e.g., revenue by restaurant vs. bar).
• Present reports to management quickly.

A. Bar Charts

• Best for: Comparing discrete categories (e.g., sales by month).
• Rules:
  • Categories on the x-axis, values on the y-axis.
  • Equal spacing between bars.
  • Avoid 3D effects (distorts perception).

Worked Example 6: Monthly Revenue for a Hotel Restaurant Data: [500K, 600K, 750K, 800K, 900K, 1M, 1.1M, 1.2M, 1M, 800K, 600K, 550K] (Jan–Dec). Bar Chart: Insight:

• Peak: July–August (summer season).
• Low: January (post-holiday slump).
• Action: Offer summer packages and January discounts.

B. Pie Charts

• Best for: Showing parts of a whole (e.g., revenue sources).
• Rules:
  • Total must be 100%.
  • Slice sizes proportional to percentages.
  • Avoid too many slices (>6 is cluttered).

Worked Example 7: Revenue Sources for a Hotel Data:

• Room sales: 60%
• F&B (Food & Beverage): 25%
• Events: 10%
• Spa: 5% Pie Chart: Insight:
• Room sales dominate (60%) → Focus on upselling premium rooms.
• F&B is 25% → Could be increased with lunch specials.

C. Histograms

• Best for: Showing distribution of continuous data (e.g., guest ages, room rates).
• Key Difference from Bar Charts:
  • X-axis shows ranges (bins), not categories.
  • No gaps between bars.

Worked Example 8: Guest Ages at a Resort Data: [22, 25, 30, 35, 40, 45, 50, 55, 60, 65] (sample ages). Histogram: Insight:

• Most guests are 20–30 or 50–60 → Tailor marketing to young travelers and retirees.

D. Line Graphs

• Best for: Showing trends over time (e.g., daily occupancy rates).
• Rules:
  • X-axis: Time (days, months).
  • Y-axis: Value (e.g., number of guests).
  • Connect points with lines.

Worked Example 9: Daily Occupancy for a Week Data: [40, 55, 70, 85, 90, 75, 60] (guests per day). Line Graph: Insight:

• Peak: Friday (90 guests).
• Dip: Sunday (60 guests).
• Action: Offer weekend packages to fill Sunday gaps.

---

---

5. Ethical Considerations in Data Collection

Poor ethics lead to biased, unreliable, or illegal data. Key principles:

1. Confidentiality: Never disclose guest names or personal details.
2. Bias Avoidance: Ensure survey questions are neutral (e.g., avoid leading questions like "Don’t you agree our service is terrible?").
3. Informed Consent: Guests must know they’re being surveyed.
4. Accuracy: No fudging numbers to meet targets.

Real-World Example: Ncell’s Data Ethics

• What they do right:
  • Anonymize data when analyzing call patterns.
  • Get consent before using customer data for marketing.
• What they avoid:
  • Selling customer lists to third parties.
  • Misrepresenting call drop rates.

Worked Example 10: Ethical Survey Design A hotel wants to survey staff satisfaction. Avoid: ❌ "You must love working here, right?" (Leading question). ✅ "On a scale of 1–5, how satisfied are you with your work environment?" (Neutral).

---

In the Real World

Hotels and travel apps use these concepts daily to improve services. Here’s how:

1. eSewa’s Guest Feedback System

   • Idea Used: Primary data collection (surveys) + tabulation.
   • How? After a guest books a hotel via eSewa, they get a post-stay email survey (structured questions on cleanliness, staff, value). Responses are tabulated to rank hotels, which influences future bookings.
   • Impact: Hotels with high ratings get more visibility, while low-rated ones must improve.

2. Pathao’s Ride Demand Prediction

   • Idea Used: Secondary data (historical ride data) + line graphs.
   • How? Pathao analyzes past ride volumes (e.g., more rides on Fridays) to predict demand and allocate drivers efficiently. They use line graphs to show trends like:
   • Impact: Reduces wait times and driver shortages during peak hours.

3. Nepal Tourism Board’s Marketing Decisions

   • Idea Used: Cross-tabulation + pie charts.
   • How? They collect data on:
     • Guest nationality (India, China, USA).
     • Purpose of visit (tourism, business, pilgrimage).
   • Example Table:

     Nationality	Tourism (%)	Business (%)	Pilgrimage (%)
     India	70	20	10
     China	60	30	10
     USA	80	15	5

   • Pie Chart Insight:
   • Action: They increase marketing in India and China (highest tourism percentages) and target business travelers with visa facilitation.

---

Exam Tip

This unit is heavily tested in TU exams with:

1. Short Questions (5 marks):
   • Define primary vs. secondary data.
   • Difference between bar charts and histograms.
   • When to use a pie chart vs. line graph.
2. Long Questions (10–15 marks):
   • Design a survey for a given scenario (e.g., "How would you collect data on guest complaints?").
   • Tabulate raw data and draw a graph (always label axes clearly).
   • Interpret a graph (e.g., "What trend does this histogram show about guest ages?").
3. Case Studies (15 marks):
   • Given hotel data, you must:
     1. Classify it (primary/secondary).
     2. Tabulate it.
     3. Choose the best graph and draw it.
     4. Interpret insights (e.g., "Should the hotel increase weekend promotions?").

Common Mistakes to Avoid: ❌ Mislabeling axes (e.g., putting time on the y-axis in a line graph). ❌ Using pie charts for trends (always use line graphs for time-series data). ❌ Ignoring ethical considerations (e.g., not mentioning confidentiality in surveys). ❌ Poor survey design (leading questions or bias).

Pro Tip:

• Always sketch graphs neatly in exams—even if not perfect, partial credit is given for effort.
• Memorize when to use each graph:
  • Categories? → Bar chart.
  • Parts of a whole? → Pie chart.
  • Trends over time? → Line graph.
  • Distributions? → Histogram.

---

Final Challenge: A 3-star hotel in Chitwan collects the following data on guest complaints over a month: Complaints: Noise (15), Cleanliness (20), Food (10), Staff (8), Others (7).

1. Tabulate the data.
2. Choose the best graph to present this.
3. Write one actionable insight for the hotel manager.

(Answer at the end of the note.)

---

Answer to Final Challenge:

1. Tabulated Data:

   Complaint Type	Frequency
   Cleanliness	20
   Noise	15
   Food	10
   Staff	8
   Others	7
   Total	60

2. Best Graph: Bar chart (comparing discrete categories).

3. Actionable Insight:

   • Cleanliness is the top complaint (33%) → Train staff on daily room checks and laundry standards.
   • Noise is second (25%) → Enforce quiet hours and provide earplugs for light sleepers.