No, the range is not a resistant measure of spread. It is one of the most sensitive statistics to outliers because it depends entirely on the two most extreme values in a dataset. A single unusually large or small observation changes the range dramatically, which is the opposite of what “resistant” means in statistics.
What “Resistant” Means in Statistics
A resistant measure is one that doesn’t change much when a few data points are extreme or unusual. The median is a classic example: if you have the dataset 2, 4, 6, 8, 10 and you change the 10 to 1,000, the median stays at 6. A non-resistant measure, by contrast, shifts substantially when even one value becomes extreme. The mean would jump from 6 to about 204 in that same example.
The range fails the resistance test completely. It’s calculated as the maximum value minus the minimum value, so it literally uses only the two data points most likely to be outliers. Every other value in the dataset is ignored. If you change any interior value, the range doesn’t budge. But if you move either endpoint, the range changes by exactly that amount.
Why the Range Is So Sensitive to Outliers
Consider a simple example: test scores of 70, 85, 90, and 95. The range is 95 minus 70, which equals 25. Now suppose one student scored 20 instead of 70. The range jumps to 75, tripling in size because of a single data point. No other values in the dataset matter to this calculation.
This makes the range unstable in practice. A dataset of 10,000 carefully measured values can have its range completely redefined by one recording error, one unusual observation, or one legitimate but rare extreme. The range tells you about the boundaries of your data, not about how the bulk of your data is distributed.
How It Compares to Standard Deviation
Standard deviation measures the average distance of each data point from the mean, incorporating every value in the dataset. This makes it a more comprehensive picture of spread. While standard deviation is also affected by outliers (it’s not resistant either), the impact is diluted because all data points contribute to the calculation. One extreme value pulls the standard deviation, but it doesn’t single-handedly define it the way it does with the range.
For a dataset of 100 values, the range uses exactly 2 of them. Standard deviation uses all 100. That fundamental difference is why standard deviation is generally considered more reliable and informative as a measure of dispersion.
If you need a measure of spread that truly is resistant, the interquartile range (IQR) is the standard choice. The IQR measures the spread of the middle 50% of your data, ignoring extreme values on both ends. Adding an outlier to your dataset typically doesn’t change the IQR at all.
When the Range Is Still Useful
Despite its lack of resistance, the range is far from useless. Its biggest advantage is simplicity. You can calculate it instantly, explain it to anyone, and it gives you a quick sense of how spread out your data is. For small datasets where outliers are unlikely or easy to spot, the range works well as a first look. If you’re a teacher glancing at exam scores to see how wide the gap is between your strongest and weakest students, the range answers that question in seconds.
In manufacturing, range charts (called R-charts) are a core part of quality control. Workers track the range of product measurements like weight or length across small production samples, typically 4 to 6 items at a time. At that scale, the range is efficient and effective for detecting when a process is drifting out of specification. Meteorologists report daily temperature ranges. Investors track the range of stock prices over a period to gauge volatility. Sales teams monitor range in daily figures to flag unusual demand swings.
The common thread is that these applications either involve small samples, use the range as a quick screening tool, or specifically care about extremes. When the goal is to know how far apart the boundaries are, the range answers exactly the right question. The problem only arises when you want a stable summary of typical spread and your data might contain outliers.
Quick Summary of Resistance
- Range: Not resistant. Uses only the maximum and minimum, making it highly sensitive to any extreme value.
- Standard deviation: Not resistant, but more stable than the range because it incorporates all data points.
- Interquartile range (IQR): Resistant. Focuses on the middle 50% of data and ignores extremes.
- Median absolute deviation: Resistant. Measures typical distance from the median, largely unaffected by outliers.
If a statistics question asks whether the range is resistant, the answer is a clear no. It’s the least resistant common measure of spread, precisely because it depends on the values most likely to be unusual.