Reading a Regression Equation

A least-squares regression line is just an equation, y-hat = a + bx, and it works like any line you've plugged numbers into before. The hat on y is a reminder: this is a predicted value, not an observed one. You get it by taking an x-value, multiplying by the slope b, and adding the intercept a. That's the whole mechanism — the statistics is in how a and b got chosen, but reading the line is pure arithmetic.

The slope b tells you how y changes, on average, for every one-unit increase in x. Always state it in context and in the original units: 'for every additional hour studied, predicted score increases by 1.7 points,' not just '1.7.' The intercept a is the predicted y when x = 0. Sometimes that's a real, sensible value — other times x = 0 is nonsense in context, like a height of zero, and then the intercept is just a mathematical anchor for the line, not a claim about the world.

This reading skill is what makes the rest of the unit matter: once you can state slope and intercept in words, you can check whether a prediction is trustworthy. Predicting inside the range of your x-data, called interpolation, is reasonable. Predicting far outside it, extrapolation, is risky, because you're assuming the same straight-line pattern keeps going where you have no evidence it does. That risk is exactly why the next ideas — residuals and checking linearity — matter before you trust any prediction.