Interpretation of the ordinary regression model - the most popular technique for making predictions of a single continuous variable - focuses on the model's coefficients with the aid of three concepts: the statistical p-value, variables "held constant," and the standardized regression coefficient. The purpose of this article is to take a closer look at these widely used, yet often misinterpreted concepts. This article demonstrates that the statistical p-value as a sole measure for declaring Xi an important predictor is sometimes problematic; secondly, that the concept of variables "held constant" is critical for reliable assessment of how Xi affects the prediction of Y; and lastly, that the standardized regression coefficient provides the correct ranking of variables in order of predictive importance under special circumstances.