Scatter Plot: A graph that displays the relationship between two quantitative variables by plotting ordered pairs as points on a coordinate plane.
- Identify the independent variable (x-axis) and dependent variable (y-axis)
- Set up coordinate axes with appropriate scales
- Plot each ordered pair as a point
- Analyze the pattern of points for association
Independent variable (x-axis): Hours studied. Dependent variable (y-axis): Test score.
X-axis: Hours (0 to 15), Y-axis: Score (60 to 95). Choose appropriate scale and label axes.
Plot: (2,65), (4,70), (6,75), (8,80), (10,85), (12,90)
Points form a nearly straight line going upward from left to right, indicating a positive association.
There is a strong positive linear association between hours studied and test scores. As study time increases, test scores tend to increase.
The scatter plot shows a strong positive linear association between hours studied and test scores. As the number of hours spent studying increases, the test scores also increase in a consistent manner.
• Variable Selection: Independent variable on x-axis, dependent on y-axis
• Association Direction: Positive (upward) or negative (downward) trend
• Strength Assessment: How closely points follow a linear pattern
Negative Association: A relationship where one variable increases as the other decreases. Points on scatter plot show a downward trend from left to right.
Independent variable (x-axis): Temperature (°F). Dependent variable (y-axis): Heating bill ($).
Plot: (40,150), (45,135), (50,120), (55,105), (60,90), (65,75)
Points form a straight line going downward from left to right, indicating a negative association.
All points lie exactly on a straight line, indicating a perfect negative linear association.
As temperature increases, heating bills decrease. This makes practical sense as warmer weather requires less heating.
The scatter plot shows a perfect negative linear association between temperature and heating bills. As temperature increases by 5°F, heating bills decrease by $15, demonstrating an inverse relationship.
• Negative Association: High values of x correspond to low values of y
• Perfect Linear: All points fall exactly on a straight line
• Real-World Context: Relationship should make practical sense
No Association: A relationship where there is no discernible pattern between variables. Points appear randomly scattered without a clear trend.
Independent variable (x-axis): Shoe size. Dependent variable (y-axis): Test score.
(7,85), (8,92), (9,78), (10,88), (8,91), (7,83), (9,80), (10,87), (8,89), (9,84)
Place each ordered pair on the coordinate plane
Points appear randomly distributed with no clear trend upward or downward
There is no apparent association between shoe size and test scores. The variables appear to be independent.
The scatter plot shows no apparent association between shoe size and test scores. The points appear randomly scattered with no clear pattern, suggesting these variables are independent of each other.
• Random Scatter: No clear pattern indicates no association
• Independence: Variables that don't influence each other
• Visual Pattern Recognition: Look for trends in point distribution
Scatter Plot: A graph showing the relationship between two quantitative variables by plotting ordered pairs (x, y) as points.
Association: A relationship between two variables where knowing the value of one variable gives information about the value of the other.
Positive Association: A relationship where higher values of one variable tend to correspond with higher values of the other variable.
Negative Association: A relationship where higher values of one variable tend to correspond with lower values of the other variable.
Strength of Association: How closely the points follow a particular pattern (strong, moderate, weak).
Form of Association: The shape of the relationship (linear, curved, clustered).
Outlier: A data point that deviates significantly from the overall pattern.
- Variable Identification: Determine which variable is independent (x) and dependent (y)
- Data Preparation: Organize data into ordered pairs
- Graph Construction: Create coordinate system and plot points
- Pattern Recognition: Look for direction, form, and strength
- Association Description: Describe the relationship in context
- Outlier Identification: Note any unusual data points
• Variable Assignment: Independent variable on x-axis, dependent on y-axis
• Association Direction: Positive (upward) or negative (downward) trend
• Strength Assessment: How closely points cluster around a line
• Form Recognition: Linear, quadratic, exponential, or no clear pattern
• Contextual Interpretation: Always interpret in the context of the problem
Strength of Association: How closely the data points follow a particular pattern. Form: The shape of the relationship (linear, curved, clustered).
(20,200), (25,210), (30,215), (35,225), (40,230), (45,240), (50,250), (55,260), (60,275), (65,290), (70,310), (75,330)
Age on x-axis, Reaction time on y-axis
As age increases, reaction time increases, indicating a positive association
Points generally follow a straight line pattern, indicating a linear relationship
Points follow the linear pattern fairly consistently but with some variation, indicating moderate strength
The data suggests that as people age, their reaction times tend to increase, which makes biological sense due to natural aging processes.
There is a positive, linear, moderate association between age and reaction time. As age increases, reaction time tends to increase in a relatively consistent linear fashion, though with some variation among individuals.
• Three Characteristics: Direction, form, and strength
• Linear Pattern: Points roughly follow a straight line
• Moderate Strength: Clear pattern but with some scatter
Outlier: A data point that significantly deviates from the overall pattern of the data. Outliers can affect the interpretation of associations.
Most points: (0,85), (15,82), (20,80), (30,78), (35,76), (40,75), (45,74), (50,72), (55,71), (60,70), (70,68), (80,67), (90,65), (100,64), (120,62)
Outlier: (100,95)
Exercise time on x-axis, Heart rate on y-axis
Most points show a clear negative linear association - as exercise increases, heart rate decreases
The point (100,95) is an outlier - it has 100 minutes of exercise but a high heart rate of 95 bpm, which contradicts the general pattern
Excluding the outlier, there is a strong negative linear association between exercise and heart rate
The general pattern makes sense - regular exercise typically lowers resting heart rate. The outlier may represent an individual with health conditions affecting heart rate despite exercise.
There is a strong negative linear association between daily exercise minutes and resting heart rate. Most individuals show that increased exercise correlates with lower heart rates. However, there is one clear outlier at (100, 95), which represents someone who exercises 100 minutes daily but has an unexpectedly high heart rate.
• Outlier Detection: Points that deviate significantly from the overall pattern
• Pattern Assessment: Analyze the general trend, considering outliers separately
• Contextual Explanation: Consider possible reasons for outliers in real-world scenarios
Scatter Plot: A graphical representation showing the relationship between two quantitative variables by plotting ordered pairs as points on a coordinate plane.
Association: A relationship between two variables where knowing the value of one variable provides information about the value of the other variable.
Positive Association: A relationship where higher values of one variable tend to correspond with higher values of the other variable (points trend upward from left to right).
Negative Association: A relationship where higher values of one variable tend to correspond with lower values of the other variable (points trend downward from left to right).
Strength of Association: How closely the points follow a particular pattern, ranging from strong (tight clustering) to weak (loose clustering).
Form of Association: The shape of the relationship, which can be linear, curved, clustered, or random.
Outlier: A data point that significantly deviates from the overall pattern of the data and may affect the interpretation of the relationship.
Correlation: A measure of the strength and direction of a linear relationship between two variables.
- Data Preparation: Organize paired data into ordered pairs (x, y)
- Variable Identification: Determine independent (x) and dependent (y) variables
- Graph Construction: Create coordinate system with appropriate scales and labels
- Point Plotting: Plot each ordered pair as a point on the coordinate plane
- Pattern Recognition: Identify direction, form, and strength of association
- Outlier Identification: Note any points that deviate significantly from the pattern
- Interpretation: Describe the relationship in the context of the problem
Association Characteristics:
• Direction: Positive (upward trend) or negative (downward trend)
• Form: Linear, quadratic, exponential, or no clear pattern
• Strength: Strong (tight clustering), moderate (some scatter), or weak (large scatter)
Interpretation Guidelines:
• Context Matters: Always interpret results in the context of the real-world problem
• Outlier Impact: Consider how outliers affect the overall pattern
• Sample Size: Larger samples generally provide more reliable patterns
Perfect Positive: All points on upward-sloping line (strength = 1)
Strong Positive: Points closely follow upward trend
Moderate Positive: Clear upward trend with some scatter
Weak Positive: Slight upward tendency with considerable scatter
No Association: Random scatter with no discernible pattern
Weak Negative: Slight downward tendency with considerable scatter
Moderate Negative: Clear downward trend with some scatter
Strong Negative: Points closely follow downward trend
Perfect Negative: All points on downward-sloping line (strength = -1)
Correlation Coefficient (r): A numerical measure of the strength and direction of a linear relationship, ranging from -1 to +1.
Line of Best Fit: A straight line that best represents the trend of the data points in a scatter plot.
Curved Associations: Relationships that follow non-linear patterns (quadratic, exponential, logarithmic).
- Data Visualization: Create scatter plot to visualize the relationship
- Pattern Recognition: Identify direction, form, and strength
- Statistical Measures: Calculate correlation coefficient if needed
- Outlier Analysis: Assess impact of unusual data points
- Model Selection: Choose appropriate mathematical model for the relationship
- Prediction Making: Use the model to make informed predictions
• Accurate Data: Ensure all data points are correctly plotted and recorded
• Proper Scaling: Choose axis scales that clearly show the data patterns
• Thorough Analysis: Consider all aspects of association (direction, form, strength)
• Contextual Relevance: Connect statistical findings to real-world meaning
• Appropriate Conclusions: Avoid overstating the strength of relationships
• Clear Communication: Present findings in accessible language
Questions & Answers
Question: How do I determine if an association is strong, moderate, or weak just by looking at a scatter plot?
Answer: Use these visual guidelines:
- Strong Association: Points are tightly clustered around a line or curve with little scatter
- Moderate Association: Points show a clear pattern but with noticeable scatter around the trend
- Weak Association: Points show a general tendency but with substantial scatter
- No Association: Points appear randomly scattered with no discernible pattern
Think about how easily you could draw a line through the points. If you can draw a line where most points are close to it, the association is strong. If points are widely scattered, it's weak.
The tighter the clustering around a line, the stronger the association.
Question: What's the difference between association and causation? Why does it matter?
Answer: The key differences are:
- Association: A statistical relationship where variables tend to vary together
- Causation: A relationship where one variable directly causes changes in another
For example, ice cream sales and drowning incidents are associated (both increase in summer), but one doesn't cause the other. They're both caused by hot weather.
It matters because observing an association doesn't mean changing one variable will definitely change the other. Always consider other possible explanations and conduct controlled experiments to establish causation.
Remember: "Correlation does not imply causation!"
Question: What should I do when I see an outlier in my scatter plot? Should I remove it?
Answer: Don't automatically remove outliers! Instead:
- Investigate: Check if it's a data entry error
- Understand: Consider if there's a legitimate reason for the unusual value
- Document: Note the outlier in your analysis
- Analyze both ways: Describe the association with and without the outlier
Outliers might represent interesting exceptions worth investigating or indicate special circumstances. They could be the most important part of your data!
Only remove outliers if you're certain they're errors. Otherwise, report how they affect your analysis.
The presence of outliers often leads to the most interesting insights in data analysis.