Histogram: A graphical representation of quantitative data using bars to show frequency distribution.
Bin Width: The range of values included in each bar (interval size).
Frequency: The number of data points falling within each bin.
- Determine the range of the data (max - min)
- Choose an appropriate number of bins (typically 5-15)
- Calculate bin width: Range ÷ Number of bins
- Define bin boundaries ensuring no gaps or overlaps
- Count data points in each bin
- Draw bars with heights proportional to frequencies
Minimum: 65, Maximum: 100, Range: 100 - 65 = 35
For 50 data points, 8 bins of width 5 seem appropriate
Starting at 65, with bin width of 5: [65-70), [70-75), [75-80), [80-85), [85-90), [90-95), [95-100), [100-105)
Each bin contains approximately 6-7 data points
The distribution appears roughly symmetric with a slight right skew due to the last bin.
Most scores cluster in the 80-99 range.
Median: Approximately 85 (middle value of 50 data points)
Mode: Appears to be in the 80-94 range (highest frequencies)
Spread: Scores range from 65 to 100
The histogram shows a roughly symmetric distribution of test scores.
Center: Around 85, Spread: 65-100
Most students scored between 80 and 99.
• Bin Width Selection: Balance between detail and clarity (typically 5-15 bins)
• Continuous Data: Histograms for quantitative variables
• Shape Analysis: Symmetric, skewed left/right, uniform
Continuous Data: Data that can take any value within a range.
Skewed Distribution: Asymmetric with a longer tail on one side.
Right Skewed: Tail extends to the right (higher values).
Range: 1.2 to 9.3, Total data points: 40
Using 10 bins of width 1.0 pound
The distribution appears slightly right-skewed.
Lower weight bins have fewer packages, while higher weight bins have more.
The last bin (10.0-11.0) has the highest frequency.
With 40 values, median is average of 20th and 21st values when ordered.
Looking at the ordered data: median ≈ (5.1 + 5.3) ÷ 2 = 5.2 pounds
Range: 9.3 - 1.2 = 8.1 pounds
Most packages weigh between 3.0 and 10.0 pounds
More packages in higher weight categories, suggesting a right-skewed distribution.
Lightest packages: 1.2-2.0 lbs (3 packages)
Heaviest packages: 9.0-10.0 lbs (4 packages) and 10.0-11.0 lbs (6 packages)
The weight distribution is right-skewed with a median of 5.2 pounds.
Range: 8.1 pounds, Mode: 10.0-11.0 pound range
More packages are in the higher weight categories.
• Skewness Detection: Longer tail indicates direction of skew
• Median Calculation: Middle value(s) when data is ordered
• Range: Maximum - Minimum
Bin Count: The number of intervals used to group the data.
Frequency Density: Frequency divided by bin width (for unequal bins).
Range: 90 - 15 = 75 minutes
With 7 bins: Bin width ≈ 75 ÷ 7 ≈ 10.7, rounded to 10 for simplicity
The distribution appears left-skewed (negatively skewed).
Most commuters have shorter commute times (15-35 minutes).
Fewer people have longer commutes, with a tail extending to the right.
With 60 values, median is average of 30th and 31st values.
Median ≈ 35-40 minutes (in the 35-45 bin)
Range: 90 - 15 = 75 minutes
IQR: Interquartile range would span the middle 50% of data
Majority of workers have commutes of 15-35 minutes
Commute times decrease as duration increases
Long commutes (75+ minutes) are rare
The commute time distribution is left-skewed.
Center: Approximately 35-40 minutes, Range: 75 minutes
Most workers have shorter commutes with fewer having long commutes.
• Skewness Direction: Tail direction determines skew name
• Bin Width Selection: Balance detail with clarity
• Mode Identification: Highest frequency bin
Histogram: A graphical display of quantitative data using adjacent bars to show the distribution of values.
Bin: An interval of values that groups the data.
Frequency: The number of data points in each bin.
Relative Frequency: The proportion of data points in each bin.
• Bin Width = (Max - Min) ÷ Number of Bins
• Relative Frequency = Frequency ÷ Total Count
• Density = Frequency ÷ Bin Width (for unequal bins)
• Center: Use median or mean depending on distribution shape
Normal Distribution: Bell-shaped, symmetric distribution with specific properties.
Mode: The value with the highest frequency.
The distribution is unimodal with the peak in the 25-30 mpg range.
It appears slightly right-skewed as the tail extends to the right.
No, this distribution is not normal:
• It's right-skewed, not symmetric
• Normal distributions have specific proportions in each standard deviation range
The mode is in the 25-30 mpg range with 25 cars.
With 80 values, median is between 40th and 41st values.
Cumulative frequencies: 5, 17, 37, 62, 77, 80
40th and 41st values fall in the 25-30 mpg range.
Estimated median: 25-30 mpg (approximately 27.5 mpg)
Range: 40 - 10 = 30 mpg
Most cars (75%) have mileage between 15-35 mpg
The distribution is right-skewed, not approximately normal.
Mode: 25-30 mpg (25 cars), Estimated median: ~27.5 mpg
Most cars have gas mileage between 15-35 mpg.
• Normal Distribution: Symmetric, bell-shaped with specific proportions
• Skewness Detection: Longer tail indicates direction
• Median Estimation: Find middle value using cumulative frequencies
Midpoint: The center value of each bin, used for approximation.
Weighted Average: Mean calculated using frequencies as weights.
The distribution is unimodal with the peak in the 60-70k range.
It appears slightly right-skewed as the tail extends to higher salaries.
The mode is in the 60-70k range with 25 employees.
With 100 employees, median is between 50th and 51st values.
Cumulative frequencies: 8, 23, 45, 70, 88, 98, 100
50th and 51st values fall in the 60-70k range.
Median: 60-70k (approximately 65k)
Mean = Σ(Midpoint × Frequency) ÷ Total
= (280 + 675 + 1210 + 1625 + 1350 + 850 + 190) ÷ 100
= 6180 ÷ 100 = 61.8 thousand dollars
Range: 100k - 30k = 70k
Most employees (90%) earn between 40k-90k
The salary distribution is right-skewed with a mode of 60-70k.
Median: ~65k, Mean: 61.8k
Most employees earn between 40k-90k annually.
• Weighted Mean: Mean = Σ(midpoint × frequency) ÷ total
• Median Estimation: Find middle value using cumulative frequencies
• Skewness Impact: Mean > Median in right-skewed distribution
Complete Summary: Analyzing Association in Two-Way Tables
Two-Way Table: Displays the frequency distribution of two categorical variables.
Marginal Frequency: Totals for each row and column.
Joint Frequency: Frequency in each individual cell.
Conditional Frequency: Frequency of one variable given a specific value of another.
Association Detection: Compare conditional relative frequencies across categories.
Independence: Conditional frequencies are similar across categories.
Dependence: Conditional frequencies differ significantly across categories.
Chi-Square Test: Statistical test for independence (advanced).
Step 1: Construct the two-way table with observed frequencies.
Step 2: Calculate marginal totals.
Step 3: Compute conditional relative frequencies for each category.
Step 4: Compare conditional frequencies to assess association.
Step 5: Draw conclusions about the relationship.
Strong Association: Large differences in conditional frequencies.
Moderate Association: Noticeable but not dramatic differences.
Weak Association: Small differences in conditional frequencies.
Independence: Similar conditional frequencies across categories.
Association: A relationship between two variables where knowing the value of one variable provides information about the value of the other.
Conditional Relative Frequency: The relative frequency of one variable given a specific value of another variable.
Independence: No relationship exists between variables; the conditional relative frequencies are the same across all categories.
Contingency Table: Another name for a two-way table.
Marginal Distribution: The distribution of one variable ignoring the other.
- Construct Table: Organize data into a two-way table format.
- Calculate Marginals: Find row and column totals.
- Compute Conditional Frequencies: Calculate P(A|B) for each category of B.
- Compare Frequencies: Look for differences in conditional relative frequencies.
- Determine Association: Decide if variables are associated or independent.
• Conditional Relative Frequency = Cell Value ÷ Row/Column Total
• Marginal Relative Frequency = Row/Column Total ÷ Grand Total
• Association exists when conditional frequencies differ significantly
• Independence exists when conditional frequencies are similar
Questions & Answers
Question: How do I know if the differences in conditional relative frequencies are significant enough to claim association?
Answer: At the high school level, use these guidelines:
Practical Significance:
- Differences > 15-20% suggest strong association
- Differences 5-15% suggest moderate association
- Differences < 5% suggest weak or no association
Context Matters:
What matters depends on the situation. A 5% difference in drug effectiveness could be clinically significant, while a 20% difference in survey responses might not be practically meaningful.
Advanced Statistics:
Later, you'll learn the chi-square test of independence, which provides a statistical measure of significance based on sample size and observed differences.
Question: Can two variables be associated but not causally related? What's the difference?
Answer: Yes, association does not imply causation!
Association vs. Causation:
- Association: Variables tend to occur together (correlation)
- Causation: One variable directly causes changes in another
Examples of Association Without Causation:
- Ice cream sales and drowning incidents (both increase in summer)
- Shoe size and reading ability in children (both related to age)
- Number of firefighters and fire damage (more severe fires require more firefighters)
Association indicates a relationship that may warrant further investigation, but establishing causation requires controlled experiments and ruling out confounding variables.
Question: When calculating conditional relative frequencies, should I condition on rows or columns?
Answer: The choice depends on what question you're trying to answer:
Condition on Rows: When you want to know the distribution of the column variable given each row category.
Example: P(Like Product | Gender) - probability of liking the product given the person's gender.
Condition on Columns: When you want to know the distribution of the row variable given each column category.
Example: P(Gender | Like Product) - probability of being a certain gender given that the person likes the product.
Both approaches are valid and can reveal different insights. The key is to be consistent in your analysis and clearly state what you're conditioning on.