Training Hours: 10, 20, 30, 40, 50, 60
Productivity Scores: 65, 70, 78, 85, 90, 95
a) Create a scatter plot
b) Describe the correlation
c) Predict the productivity score for 70 hours of training
Bivariate data: Data that involves two different variables measured on the same set of individuals or objects. Scatter plots are used to visualize relationships between these variables.
- Organize data into ordered pairs (x, y)
- Create a scatter plot with appropriate scales
- Identify the pattern and direction of the relationship
- Assess the strength of correlation
- Make predictions based on the trend
Independent variable (x-axis): Training Hours = [10, 20, 30, 40, 50, 60]
Dependent variable (y-axis): Productivity Scores = [65, 70, 78, 85, 90, 95]
Plot each ordered pair (Training Hours, Productivity Score)
Label axes appropriately with units
As training hours increase, productivity scores also increase
Points form an upward-sloping pattern
Indicates positive correlation
From 10 to 60 hours: Increase of 50 hours
Productivity increase: From 65 to 95 = 30 points
Rate: 30/50 = 0.6 points per hour
Additional 10 hours beyond 60 hours
Expected increase: 10 × 0.6 = 6 points
Predicted score: 95 + 6 = 101 (capped at 100) = 100
a) Scatter plot created with positive upward trend
b) Strong positive correlation exists
c) Predicted productivity score for 70 hours is approximately 100
• Variable identification: Correctly assign independent and dependent variables
• Pattern recognition: Identify trends in the data
• Rate calculation: Determine consistent rate of change
Temperature (°F): 80, 85, 90, 95, 100, 105
Ice Cream Sales ($): $120, $140, $160, $180, $200, $220
a) Determine the correlation coefficient
b) Interpret the strength and direction
c) Predict sales at 110°F
Correlation coefficient: A numerical measure that quantifies the strength and direction of the linear relationship between two variables. Ranges from -1 to +1.
Temperature (x): [80, 85, 90, 95, 100, 105]
Ice Cream Sales (y): [120, 140, 160, 180, 200, 220]
For every 5°F increase, sales increase by $20
Constant rate of change: $4 per °F
Perfect linear relationship
Since data points fall exactly on a straight line
Correlation coefficient r = +1 (perfect positive correlation)
Rate of change: $4 per °F
From 105°F to 110°F: 5°F increase
Expected increase: 5 × $4 = $20
Predicted sales: $220 + $20 = $240
a) Correlation coefficient r = +1 (perfect positive)
b) Strongest possible positive correlation
c) Predicted sales at 110°F is $240
• Constant rate: Equal changes in both variables indicate perfect correlation
• Linear relationship: Points form a straight line
• Prediction accuracy: Perfect correlation allows precise predictions
Study Hours: 2, 4, 6, 8, 10, 12
Exam Scores: 60, 68, 75, 82, 88, 93
The relationship can be modeled by y = 0.75x + 58
a) Interpret the slope and y-intercept
b) Predict the score for 15 hours of study
c) How many hours of study are needed to achieve a score of 95?
Linear model: An equation of the form y = mx + b, where m is the slope (rate of change) and b is the y-intercept (initial value when x = 0).
Slope (m = 0.75): For each additional hour of study, exam score increases by 0.75 points
Y-intercept (b = 58): Predicted score with 0 hours of study is 58 points
Substitute x = 15 into the equation:
y = 0.75(15) + 58 = 11.25 + 58 = 69.25
Predicted score: approximately 69 points
Set y = 95 and solve for x:
95 = 0.75x + 58
95 - 58 = 0.75x
37 = 0.75x
x = 37 ÷ 0.75 = 49.33
Approximately 49.3 hours of study needed
Check if predictions make sense in context
Score of 69 for 15 hours seems reasonable
Almost 50 hours for 95 points seems excessive but mathematically correct
a) Slope = 0.75 (0.75 points per hour), Y-intercept = 58 (base score)
b) Predicted score for 15 hours is approximately 69 points
c) Approximately 49.3 hours needed for 95 points
• Model interpretation: Understand real-world meaning of slope and intercept
• Substitution method: Replace known variable to find unknown
• Algebraic manipulation: Solve equations for desired variable
Bivariate Data: Data that involves two different variables measured on the same subjects
Independent Variable: The variable that is manipulated or controlled (x-axis)
Dependent Variable: The variable that responds to changes in the independent variable (y-axis)
- Read carefully: Identify the two variables involved
- Organize data: Create ordered pairs (x, y)
- Visualize: Create a scatter plot to see the relationship
- Analyze: Determine the type and strength of correlation
- Model: Find the equation that represents the relationship
- Apply: Use the model to make predictions or answer questions
- Validate: Check if the answer makes sense in context
• Variable assignment: Independent variable goes on x-axis, dependent on y-axis
• Correlation range: Values between -1 and +1 indicate strength and direction
• Linearity assumption: Models work best when relationship is approximately linear
• Context relevance: Always interpret results in real-world terms
Questions & Answers
Question: How do I know which variable should be on the x-axis and which should be on the y-axis when creating a scatter plot?
Answer: The key is to identify cause and effect:
- Independent variable (x-axis): The variable you control or manipulate, or the one you think influences the other
- Dependent variable (y-axis): The variable that responds to changes in the independent variable
Ask yourself: "Which variable am I using to predict or explain the other?" The predictor goes on the x-axis.
Examples: Study hours (x) predicts test scores (y), Temperature (x) affects ice cream sales (y), Age (x) influences height (y).
If it's unclear which variable influences the other, you can place either on the x-axis, but the interpretation changes.
Question: Why do we say correlation doesn't imply causation? Can't a strong correlation mean one thing causes another?
Answer: This is a critical concept! A correlation only shows that two variables move together - it doesn't prove that one causes the other. Reasons include:
- Third variable: A hidden factor might cause both variables to change (like ice cream sales and drowning incidents both rising in summer due to heat)
- Coincidence: Variables might correlate by chance
- Reverse causation: The direction of causality might be opposite to what we assume
Example: There might be a correlation between ice cream sales and crime rates, but eating ice cream doesn't cause crime. Instead, hot weather increases both ice cream consumption and outdoor activities where crimes occur.
Always consider whether a causal relationship makes logical sense in the real world.
Question: When is it appropriate to make predictions using a linear model, and when should I be careful?
Answer: Prediction accuracy depends on the range of your data:
- Safe predictions (interpolation): Making estimates within the range of your data points
- Risky predictions (extrapolation): Making estimates far outside your data range
Example: If your data shows study time from 1-12 hours, predicting scores for 8 hours is safe. Predicting for 20 hours would be risky because the relationship might not continue linearly.
Always consider whether the linear relationship makes sense beyond your observed data. Look for physical or logical constraints that might change the relationship.
Bivariate Data: Data that involves two different variables measured on the same set of individuals or objects.
Independent Variable: The variable that is manipulated or controlled, typically plotted on the x-axis.
Dependent Variable: The variable that responds to changes in the independent variable, typically plotted on the y-axis.
Scatter Plot: A graph showing the relationship between two variables as a collection of points.
Correlation: A measure of the strength and direction of the linear relationship between two variables.
Linear Model: An equation of the form y = mx + b that represents the relationship between variables.
• Variable Assignment: Independent variable on x-axis, dependent variable on y-axis
• Correlation Direction: Positive (same direction), negative (opposite direction), or none
• Correlation Strength: How closely points follow a linear pattern
• Causation Warning: Correlation does not prove causation
- Problem Reading: Carefully read the problem to identify the two variables involved
- Data Organization: Create ordered pairs (x, y) from the given information
- Visualization: Create a scatter plot with properly labeled axes
- Pattern Recognition: Identify the direction and strength of the relationship
- Model Creation: Find the equation that best represents the relationship
- Application: Use the model to make predictions or answer questions
- Validation: Check if the results make sense in the real-world context
- Communication: Clearly state the findings with proper interpretation
Examples - Advanced: Economic indicators, scientific experiments with multiple variables
Key Tips: Always identify the relationship direction, check reasonableness of predictions
Students should be able to organize bivariate data, create scatter plots, identify correlation patterns, interpret linear models, make predictions, and understand the limitations of correlation-based inferences.