2.3 Measures of the Location of the Data

Measures of the Location of the Data

Quartiles and percentiles tell you where a specific data value sits relative to the rest of the dataset. Instead of summarizing data with a single number (like the mean), these measures divide data into equal parts so you can say things like "this score is higher than 85% of all scores." They're essential for comparing individual data points across different distributions.

The median and interquartile range (IQR) are robust alternatives to the mean and standard deviation. "Robust" here means they resist the pull of outliers. When a distribution is heavily skewed, the median and IQR often give a more honest picture of center and spread than the mean and standard deviation do.

more resources to help you study

practice questions

Measures of the Location of the Data

Quartiles and percentiles calculation

Quartiles divide an ordered dataset into four equal parts. Each quartile marks a boundary where a certain fraction of the data falls below.

• First quartile (Q1): the 25th percentile. 25% of the data falls at or below this value.
• Second quartile (Q2): the 50th percentile, also known as the median. It splits the data in half.
• Third quartile (Q3): the 75th percentile. 75% of the data falls at or below this value.

Percentiles generalize this idea by dividing an ordered dataset into 100 equal parts. The $k$th percentile is the value below which $k$% of the data falls. So if you're at the 60th percentile on an exam, 60% of test-takers scored below you.

To calculate quartile positions:

1. Arrange the data in ascending order.
2. Find the position of each quartile using $\text{position} = \frac{k(n+1)}{4}$, where $n$ is the number of data points and $k = 1, 2, 3$ for Q1, Q2, Q3.

To calculate the position of any percentile:

1. Arrange the data in ascending order.
2. Use $\text{position} = \frac{i(n+1)}{100}$, where $i$ is the desired percentile.

If the position lands between two data points, you interpolate between them (more on this below).

Median and interquartile range interpretation

The median is the middle value of an ordered dataset. Half the data falls below it and half falls above.

• In a symmetric distribution, the median equals the mean.
• In a skewed distribution, the median stays near the center of the data while the mean gets pulled toward the tail. That's why the median is preferred as a measure of center when outliers or skew are present.

For example, if five home prices are $150K, $160K, $170K, $180K, and $900K, the mean is $312K but the median is $170K. The median better represents a "typical" home price here because the $900K outlier inflates the mean.

The interquartile range (IQR) measures the spread of the middle 50% of the data:

$IQR = Q3 - Q1$

• A larger IQR means more variability in that middle portion.
• A smaller IQR means the central data values are tightly clustered.
• Because the IQR ignores the lowest 25% and highest 25%, it isn't affected by extreme values. This makes it especially useful for comparing spread across skewed datasets or datasets with outliers.

Percentile formulas and applications

Finding the value at a specific percentile:

1. Arrange the data in ascending order.
2. Calculate the position: $L = \frac{i(n+1)}{100}$, where $i$ is the percentile and $n$ is the number of data points.
3. If $L$ is a whole number, the percentile value is the data point at position $L$.
4. If $L$ is not a whole number, interpolate between the two surrounding data points. Take the data value at the position below and add the fractional part times the difference to the next value.

For example, suppose you have 12 data points and want the 25th percentile. The position is $\frac{25(13)}{100} = 3.25$. You'd take the 3rd data value and add 0.25 times the difference between the 4th and 3rd values.

Finding the percentile of a specific value:

1. Arrange the data in ascending order.
2. Count how many data points fall below the value of interest.
3. Compute the percentile rank:

$\text{Percentile} = \frac{\text{Number of data points below the value}}{\text{Total number of data points}} \times 100$

Common applications of percentiles:

• Standardized testing: SAT and ACT scores are reported with percentile ranks so you can see how you performed relative to all other test-takers.
• Pediatric growth charts: Doctors track a child's height and weight percentile relative to their age group. A child at the 40th percentile for height is taller than 40% of children the same age.
• Income data: Economists use percentiles to describe income inequality (e.g., "the top 10% of earners" means those at or above the 90th percentile).

Measures of Central Tendency and Variability

These measures often appear alongside quartiles and percentiles, so it's worth knowing how they connect.

• Mean: The arithmetic average, calculated by summing all values and dividing by $n$. Sensitive to outliers.
• Mode: The most frequently occurring value. A dataset can have one mode, multiple modes, or no mode at all.
• Range: $\text{Range} = \text{max} - \text{min}$. Simple to compute, but since it depends on only the two most extreme values, a single outlier can make it misleading.
• Variance: Quantifies spread by averaging the squared deviations from the mean. The formula for sample variance is $s^2 = \frac{\sum (x_i - \bar{x})^2}{n-1}$. Squaring makes the units hard to interpret directly.
• Standard deviation: $s = \sqrt{s^2}$. It brings the units back to the same scale as the original data, making it more interpretable than variance.

The key distinction: the mean, variance, and standard deviation are all sensitive to outliers, while the median and IQR are resistant. Choosing between them depends on the shape of your distribution.