Hi @LocalTables505,
Welcome in the Community !
The Lack of Fit test is helpful to evaluate if your model fits the data well. A statistically significant p-value for this test indicates that the model error is largely bigger than the pure error, and that your model may not be adequate : https://www.jmp.com/support/help/en/17.2/#page/jmp/lack-of-fit.shtml
This situation may happen in your example if your first order model is "too simple" for the results/data you have collected, for example if there are some missing terms to model curvature in your data (through 2nd order terms, interaction and/or quadratic effects).
You can also check the adequacy of your regression model through residuals analysis, to check if regression assumptions are met and if any patterns are detected in the residuals that could indicate model's inadequacy : https://www.jmp.com/en_us/statistics-knowledge-portal/what-is-regression/simple-linear-regression-assumptions.html
Hope this answer will help your understanding,
To add just a bit to everything @Victor_G wrote, which I agree with, is have you plotted your responses vs. predictors in a simple scatterplot fashion? I'd start there, even before modeling. This gives you a fighting chance at proposing an initial model that will most appropriately fit the data. It will help you identify a general pattern in the responses (linear, curvilinear, etc.) outliers, odd looking data points, and, if you have replicates of the predictor's values, a chance to look at their variability in a graphical fashion. As I always told my engineers, scientists and others, the three steps to successful data analysis are. 1. plot the data. 2. Plot The Data. 3. PLOT THE DATA!
