A continuous variable is one that can take on infinitely many values within a given range. Between any two values, no matter how close together, there’s always another possible value. Height, weight, blood pressure, temperature, and time are all continuous variables. If you’re learning statistics or trying to make sense of a dataset, understanding this concept is one of the first steps.
How Continuous Variables Work
The defining feature of a continuous variable is that you can always find an intermediate value between any two measurements. Consider a length measurement between 5 and 10 centimeters. You could record 5.1, then 5.12, then 5.123, and keep going with more decimal places indefinitely. The only thing limiting your precision is the tool you’re using to measure.
This is what makes continuous variables fundamentally different from variables that jump between fixed values. An animal’s mass could theoretically be 123.75921 kilograms, and even that might not capture the exact value. You could always get more precise. Common examples in everyday life and research include body mass, height, blood pressure, cholesterol levels, temperature, and time. In finance, the price of gold, rates of return on investments, and energy consumption per unit of production all function as continuous variables.
Continuous vs. Discrete Variables
The simplest way to tell the two apart: if you can list or count every possible value, it’s discrete. If you can’t, it’s continuous.
A die roll is discrete. It lands on 1, 2, 3, 4, 5, or 6. That’s the complete list. The year a student was born is discrete: 1992, 1993, 2001. There’s no value between 2000 and 2001 that makes sense. You can count these possibilities on your fingers, or at least on a spreadsheet.
Now think about the exact finishing time of a 100-meter sprint. Not rounded, but the true, precise time. It could be 9.571 seconds, or 9.5713, or 9.57134. There’s no way to list every possible value because there are infinitely many. That’s a continuous variable. Interestingly, if you round that same finishing time to the nearest hundredth of a second, it becomes discrete. It could be 9.56, 9.57, 9.58, and so on. Now you can list the possibilities.
The Measurement Problem
Here’s something that trips people up: in practice, continuous variables are almost always recorded as rounded numbers. A bathroom scale gives you weight to the nearest tenth of a pound. A thermometer reads to the nearest degree. Netflix logs viewing time to the nearest second. Does that make these variables discrete?
Not really. The underlying phenomenon is still continuous. Your actual weight isn’t locked to neat decimal points. It’s some infinitely precise number that your scale rounds for you. The discrete appearance comes from measurement limitations, not from the true nature of what’s being measured. This is why statistics treats these variables as continuous even though the recorded data has a finite number of values. The same logic applies to prices. A gold price of $1,206.90 is technically part of a discrete list of penny values, but the divisions are so small that treating it as continuous makes far more sense for analysis.
Interval and Ratio Scales
Continuous variables typically fall on one of two measurement scales: interval or ratio. The difference comes down to whether zero actually means “none.”
On a ratio scale, zero means the complete absence of the thing being measured. Zero meters is zero meters, whether you’re using feet or kilometers. This means ratios are meaningful: 20 kilometers is genuinely twice as far as 10 kilometers. Height, weight, distance, and time elapsed all sit on ratio scales.
On an interval scale, zero is just an arbitrary point. Zero degrees Fahrenheit doesn’t mean “no temperature.” And because the zero point is arbitrary, ratios don’t work: 20°F is not twice as warm as 10°F. Calendar years work the same way. The distinction matters when you’re choosing how to analyze or interpret your data, because certain calculations (like percentages or growth rates) only make sense with ratio-scale data.
Why It Matters for Analysis
Whether your variable is continuous or discrete determines which statistical tools you should use and how you should visualize your data.
For statistical testing, continuous data that follows a bell-shaped (normal) distribution can be analyzed with tools like t-tests and analysis of variance (ANOVA), which compare group averages. These methods assume the data is spread in a roughly symmetrical pattern. When continuous data doesn’t meet that assumption, alternative tests like the Mann-Whitney test perform nearly as well, with about 95.5% of the efficiency of standard methods even when the normal distribution assumption holds.
Visualization choices also depend on variable type. Histograms are one of the most common ways to see how continuous data is distributed. They divide your data into bins (say, 10-pound increments for weight) and show how many observations fall into each bin. This helps you spot patterns like whether your data clusters around a central value or skews to one side. Box plots serve a different purpose: they give you a compact summary showing the median, the middle 50% of values, and any extreme outliers. Scatterplots work well when you’re looking at the relationship between two continuous variables, like height and weight.
In medicine and clinical research, continuous variables like BMI, blood pressure, and body temperature are central to analysis. Researchers sometimes convert continuous variables into categories for practical reasons. For example, a lab value like creatinine might be split into “above 1.8” and “below 1.8” to classify patients into risk groups. This trades away some of the richness of continuous data in exchange for simpler interpretation.
Probability and Continuous Variables
One subtle but important concept: with a continuous variable, the probability of getting any single exact value is essentially zero. The chance that someone weighs exactly 150.000000 pounds, carried out to infinite decimal places, is vanishingly small. Instead, probabilities for continuous variables are defined over ranges. You can ask “what’s the probability that someone weighs between 145 and 155 pounds?” and get a meaningful answer.
This is why statisticians use different mathematical tools for continuous and discrete data. Discrete variables use probability mass functions, which assign a specific probability to each possible value (like a 1-in-6 chance for each face of a die). Continuous variables use probability density functions, which describe the relative likelihood of values across a range. The probability of falling within any specific interval is calculated as the area under the curve between those two points. The total area under the entire curve always equals 1, representing 100% certainty that the variable takes some value.