Data Visualization Fundamentals

1.1 Con số vs Hình ảnh

1Raw data:
2Q1: 150, Q2: 180, Q3: 210, Q4: 195
3 
4Observation từ số: "Q3 cao nhất"
5 
6Observation từ chart:
7- Q3 cao nhất
8- Uptrend Q1→Q3
9- Q4 giảm nhẹ (potential concern)
10- Growth rate giảm dần

1.2 Anscombe's Quartet

4 datasets có cùng thống kê (mean, variance, correlation) nhưng hoàn toàn khác khi visualize.

1Dataset 1: Linear relationship
2Dataset 2: Quadratic relationship
3Dataset 3: Perfect line + 1 outlier
4Dataset 4: Vertical line + 1 outlier
5 
6Statistics: Identical → Visualization: Completely different!

Lesson: Luôn visualize data, không chỉ dựa vào statistics.

1.3 Cognitive Load

Visualization giúp:

• ✅ Pattern recognition nhanh
• ✅ Outlier detection dễ dàng
• ✅ Comparison trực quan
• ✅ Trend identification rõ ràng

2.1 Decision Framework

2.2 Chart Selection Guide

2.3 Common Mistakes

❌ Pie chart cho 15 slices → Dùng horizontal bar chart

❌ 3D charts → Luôn dùng 2D

❌ Dual-axis abuse → Separate charts hoặc normalize

❌ Truncated Y-axis → Bắt đầu từ 0 hoặc clearly labeled

3.1 Color Purposes

3.2 Best Practices

1✅ Limit to 5-7 colors max
2✅ Use colorblind-friendly palettes
3✅ Consistent colors across dashboard
4✅ Gray for context, color for focus
5 
6❌ Rainbow colors (hard to distinguish)
7❌ Red = Good, Green = Bad (reverse!)
8❌ Same color, different meanings
9❌ Bright colors for background

3.3 Accessible Color Palettes

1Colorblind-safe:
2#1f77b4 (Blue) #ff7f0e (Orange) #2ca02c (Green)
3#d62728 (Red)  #9467bd (Purple) #8c564b (Brown)
4 
5Sequential: #deebf7 → #9ecae1 → #3182bd → #08519c
6Diverging:  #d7191c → #fdae61 → #ffffbf → #a6d96a → #1a9641

4.1 Data-Ink Ratio

Maximize data, minimize non-data ink. Bỏ gridlines nặng, 3D effects, decorative elements.

4.2 Gestalt Principles

Proximity: Nhóm related items gần nhau

Similarity: Same color/shape = Same category

Enclosure: Border groups related info

4.3 Visual Hierarchy

1Most Important → Large, bold, top position
2Supporting → Medium, prominent
3Detail → Smaller, bottom

5.1 Story Structure

11. SETUP (Context):     "Q3 revenue was $2.5M..."
22. CONFLICT (Problem):  "...but growth slowed to 5% vs 15% Q2"
33. RESOLUTION (Action): "Marketing spend dropped 20%
4                         → Recommend: Increase budget $50K"

5.2 Annotation Techniques

• Callouts: Highlight specific points
• Reference lines: Targets, averages
• Trend lines: Show direction
• Notes: Context, caveats

5.3 Dashboard Flow

Attributes Brain Processes Instantly

1Make important data POP:
2 
3Before: ■ ■ ■ ■ ■ ■ ■ ■ (all same)
4After:  ■ ■ ■ ■ █ ■ ■ ■ (target highlighted)
5               ↑
6           "This one!"

Exercise 1: Critique This Dashboard

1- 3D Pie Chart with 12 slices
2- Rainbow colors
3- Legend far from chart
4- No title
5- Y-axis starts at 1000
6- Dual axis with different scales
7 
8→ List 6 problems and suggest fixes.

Exercise 2: Choose the Right Chart

1. "Compare revenue across 10 regions" → ?
2. "Show how market share changed over 5 years" → ?
3. "Display correlation between ad spend and sales" → ?
4. "Break down expenses by category" → ?
5. "Show distribution of customer ages" → ?

Kiến thức đã học

Câu hỏi tự kiểm tra

1. Anscombe’s Quartet chứng minh điều gì về data visualization?
2. Pre-attentive attributes nào hiệu quả nhất cho so sánh?
3. Data-Ink Ratio là gì?
4. Data Storytelling gồm những bước nào?

Bài tiếp theo: Tableau Fundamentals — Bắt đầu với Tableau Desktop và tạo visualization đầu tiên!