A linear coefficient is a number that describes the relationship between two variables in a straight-line equation. In its simplest form, it’s the slope in the equation y = mx + b, where the coefficient m tells you how much y changes every time x increases by one unit. The term also appears in statistics, where the “linear correlation coefficient” measures how strongly two variables move together in a straight line, and in physics, where a “linear coefficient of thermal expansion” describes how much a material stretches when heated.
Because the phrase shows up across several fields, the meaning depends on context. Here’s how it works in each one.
The Slope in a Linear Equation
The most basic linear coefficient is the slope in the standard equation of a line: y = mx + b. In this equation, m is the coefficient (the number multiplied by x), and b is the y-intercept, the point where the line crosses the vertical axis. The slope describes the rate of change between the two variables. If m = 3, then every time x goes up by 1, y goes up by 3. If m is negative, y decreases as x increases.
You may also see this written as y = a + bx, where b is the slope and a is the intercept. The letters change depending on the textbook, but the idea is identical: the linear coefficient is whatever number sits in front of the variable x. It controls how steep the line is and whether it tilts upward or downward.
Coefficients in Linear Regression
In statistics, linear regression uses the same straight-line structure to predict one variable from one or more others. The coefficient for each predictor variable (often called beta, or b) estimates the exact change in the outcome when that predictor increases by one unit, while holding everything else constant.
Some concrete examples make this easier to grasp:
- Education and income: A coefficient of 0.21 means that for every additional year of education, income increases by 0.21 units on whatever scale the researchers used.
- Work hours and job prestige: A coefficient of 0.84 means that for every extra hour worked per week, a person’s job prestige score rises by 0.84 points on a 0-to-100 scale, after controlling for age and family background.
- Homeownership and state spending: A coefficient of -3.44 means that for every 1% increase in a state’s homeownership rate, the state spends $3.44 less per person on corrections, after accounting for income, population, and other factors.
When a coefficient equals zero (or is statistically indistinguishable from zero), it means that predictor has no detectable linear effect on the outcome. Researchers test this with a p-value. If the p-value falls below a chosen threshold, typically 0.05, the coefficient is considered statistically significant, meaning the relationship is unlikely to be due to chance alone.
The Linear Correlation Coefficient (r)
Another common use of “linear coefficient” refers to the Pearson correlation coefficient, written as r. This single number summarizes how tightly two variables follow a straight-line pattern. It always falls between -1 and +1.
An r of +1 means a perfect positive linear relationship: as one variable rises, the other rises in lockstep. An r of -1 means a perfect negative relationship: one goes up, the other goes down in exact proportion. An r of 0 means no linear relationship at all. Most real-world data lands somewhere in between.
One widely used scale for interpreting the strength of r:
- 0 to 0.19: Very weak
- 0.20 to 0.39: Weak
- 0.40 to 0.59: Moderate
- 0.60 to 0.79: Strong
- 0.80 to 1.0: Very strong
These thresholds apply to the absolute value of r, so -0.75 is just as strong as +0.75. The BMJ notes these cutoffs are “rather arbitrary,” and context matters. A correlation of 0.30 might be meaningful in social science research but trivial in a physics experiment.
One useful property of r is that it’s unitless. Whether you’re measuring height in centimeters or income in dollars, the correlation coefficient strips away the units and gives you a pure number you can compare across different datasets. The sign of r always matches the sign of the regression slope: if the slope tilts upward, r is positive; if the slope tilts downward, r is negative.
Linear Coefficient of Thermal Expansion
In physics and engineering, “linear coefficient” usually refers to the coefficient of linear thermal expansion. This number tells you how much a material’s length changes per degree of temperature change, relative to its original length. It’s typically expressed in parts per million per degree Celsius (ppm/°C).
The principle is straightforward: when you heat a material, it gets longer. When you cool it, it shrinks. The linear expansion coefficient tells you the rate at which this happens. Materials with high coefficients expand a lot; materials with low coefficients barely budge.
Some common values:
- Pure aluminum: 23.6 ppm/°C
- Pure iron: 11.7 ppm/°C
- 304 stainless steel: 17.3 ppm/°C
- 316 stainless steel: 16.0 ppm/°C
Aluminum expands roughly twice as much as iron for the same temperature change. This matters in engineering design. If you bolt aluminum and steel together and the assembly heats up, the two metals expand at different rates, which can create stress or warping. Engineers choose materials with compatible expansion coefficients, or they design joints that allow for movement.
How to Tell Which Meaning Applies
If you’re in an algebra or precalculus class, “linear coefficient” almost certainly means the slope, the number multiplying x in a straight-line equation. If you’re reading a research paper or working through a statistics course, it likely refers to either a regression coefficient (how much the outcome changes per unit of the predictor) or the Pearson correlation coefficient r (how strong the linear relationship is). In a physics or materials science context, it points to thermal expansion.
All three share the same core idea: a single number that captures how one quantity changes in proportion to another along a straight line. The slope tells you the steepness. The regression coefficient tells you the predicted effect. The correlation coefficient tells you how closely the data follows the pattern. And the thermal expansion coefficient tells you how much a material stretches. Different fields, same underlying concept of a linear, proportional relationship.