The p-value has a long and storied use in expert testimony and litigation. However, researchers and academics in science and academia are questioning the use and reliance of p-values for the determination of statistical significance of findings. It is a natural extension to explore the concerns of science and academia to how p-values are used in litigation. This paper explores the history behind the use of the p-value in expert testimony and litigation, the potential issues with triers of fact and experts relying on p-values to make decisions based on statistical significance. Using a gender pay equity model as an example, this paper will develop several simulated datasets and models to show how p-values can be influenced to show statistically significant results or non-statistically significant results using simple modifications to the model specifications or the amount of data being used. This paper will also explore potential methods for analyzing the power of the test and considerations for regression analysis. The analysis suggests that more consideration is necessary for the use of a p-value “bright line” for gender pay analyses, and that a simple “bright line” consideration for litigation could be easily manipulated.