Data Analysis and Interpretation

What is Data Analysis?

Data analysis is the process of examining data to find patterns, draw conclusions, and make decisions.

Steps:

1. Collect data
2. Organize and display data
3. Analyze patterns and trends
4. Draw conclusions
5. Make predictions or decisions

Choosing the Right Display

Different data types need different displays!

Categorical Data

Best for: Categories or groups

Bar Graph: Compare different categories

• Favorite colors, types of pets, sports teams

Pie Chart: Show parts of a whole

• Budget breakdown, survey results

Numerical Data

Best for: Measurements or counts

Histogram: Show frequency distribution

• Test score ranges, heights

Line Graph: Show change over time

• Temperature over days, stock prices

Box Plot: Compare distributions

• Test scores across classes

Scatter Plot: Show relationship between two variables

• Height vs. weight, study time vs. grades

Example: Choose the Display

Situation: Track daily high temperatures for a month

Best choice: Line graph Why? Shows change over time

Situation: Compare number of students in each grade

Best choice: Bar graph Why? Compares different categories

Analyzing Center and Spread

Center: Typical value (mean, median, mode)

Spread: How scattered the data is (range, IQR)

Example: Two Classes

Class A: Test scores 75, 78, 80, 82, 85

• Mean = 80
• Range = 10 (consistent scores)

Class B: Test scores 60, 70, 80, 90, 100

• Mean = 80
• Range = 40 (varied scores)

Analysis: Both have same mean, but Class B is more spread out!

Identifying Trends

Trend: General direction or pattern in data

Increasing: Values going up over time

• Sales growing each month

Decreasing: Values going down over time

• Temperature cooling in fall

Constant: Values staying about the same

• Steady heart rate

Cyclical: Pattern that repeats

• Seasonal sales (high in December)

Example: Analyze Trend

Monthly sales: Jan=100, Feb=110, Mar=115, Apr=125, May=130

Trend: Increasing Rate: About +7 to +10 per month Prediction: June might be around 135-140

Comparing Data Sets

Example: Compare Two Stores

Store A:

• Mean daily sales: $5,000
• Range: $500
• Consistent, predictable

Store B:

• Mean daily sales: $5,000
• Range: $3,000
• Variable, less predictable

Conclusion: Store A more stable, Store B more volatile

Identifying Patterns

Clusters: Groups of similar values

• Most students score 70-80

Gaps: Missing ranges

• No students scored 50-60

Outliers: Extreme values

• One student scored 100 (rest below 85)

Symmetry: Balanced distribution

• Equal spread on both sides of center

Drawing Conclusions

Based on data, not assumptions!

Example: Survey Results

Data: 80 out of 100 students prefer morning classes

Valid conclusion: "Most students prefer morning classes"

Invalid conclusion: "All students hate afternoon classes"

• Data doesn't say this!

Misleading Displays

Watch out for:

1. Scale Manipulation

Broken axis: Makes small differences look large

• Graph starts at 95 instead of 0

2. Missing Labels

No axis labels: Can't tell what's being measured

3. Cherry-Picking Data

Selective data: Only showing data that supports a conclusion

• Showing sales from best months only

4. Wrong Display Type

Pie chart for rankings: Doesn't show order well

Making Predictions

Use data to make informed guesses about future values.

Example: Plant Growth

Data: Week 1=2cm, Week 2=4cm, Week 3=6cm, Week 4=8cm

Pattern: Growing 2cm per week

Prediction: Week 5 will be about 10cm

Caution: Assumes pattern continues!

Statistical Questions

Good statistical question: Has variability, answered with data

Examples:

• ✓ "How tall are students in our class?" (varies)
• ✗ "How tall is Sam?" (single answer)

Characteristics:

• Anticipates variability
• Can be answered by collecting data
• Summarized with statistics

Real-World Applications

Business: Analyze sales trends

• Which products sell best?
• When do sales peak?

Science: Interpret experiment results

• Did the treatment work?
• Is there a significant difference?

Sports: Evaluate player performance

• Who scores most consistently?
• Are free throws improving?

Health: Track fitness progress

• Is heart rate decreasing?
• Are workouts effective?

Education: Assess learning

• Are students improving?
• Which teaching method works better?

Critical Thinking with Data

Always ask:

• Where did the data come from?
• Is the sample representative?
• Could there be bias?
• What's being measured?
• Are there other explanations?

Example: Toothpaste Study

Claim: "9 out of 10 dentists recommend our toothpaste"

Questions to ask:

• How many dentists were asked?
• Were they paid?
• What were the choices?
• Which 10 dentists?

Practice