Row C: Tests sep. intercep, sep slope vs. sep intercep, com slope

The row C test is to see if the user can assume that the batch does not have significant impact on the slope of the model. The null hypothesis is that these batches may be pooled together with a common slope unless evidence is present to reject this hypothesis.

If Batch and Months are included in the model as additive terms, then the intercept value differs depending on the batch. However, a common slope still exists.

The Batch * Time interaction term indicates the additional variance explained by having separate slopes instead of common slopes. The p-value of this term (.473259) is greater than the chosen value of .25. Hence, this hypothesis is not rejected.

In the event that the p-value is significant (below the chosen value of .25), you cannot pool the batches together as the slopes are different. In this case, select the Sep. Intercept, Sep Slope model.

The Error row also matches with Row D in the Model Test spreadsheet. Row D is not used in the selection of the model. It represents the unexplained variance that still exists when Time, Batch, and Batch*Time interaction are all in the model.