[statistics] A probability resulting from a statistical test of the coefficient associated with each independent variable in a regression model. The null hypothesis for this statistical test states that the coefficient is not significantly different from zero. Small p-values reflect small probabilities. They suggest that the coefficient is significantly different from zero, and consequently, that the associated explanatory variable is helping to model or predict the dependent variable. Variables with coefficients near zero do not help predict or model the dependent variable; they are almost always removed from the regression equation (unless there are strong theoretical reasons to keep them).