The most important statistic in the analysis of variance table is the p-value (P). There is a p-value for each term in the model (except for the error term). The p-value for a term tells you whether the effect for that term is significant:

·    If P is less than or equal to the a-level you have selected, then the effect for the term is significant.

·    If P is larger than the a-level you have selected, the effect is not significant.

If the effect of a fixed factor is significant, then the level means for the factor are significantly different from each other.

If the effect of a random factor is significant, then the variance of the factor is not zero.

If the effect of an interaction term is significant, then the effects of each factor are different at different levels of the other factor(s).For this reason, it does not make sense to try and interpret the individual effects of terms which are involved in significant higher-order interactions.

In the salary analysis, of the effects of subject, degree, and the degree by subject interaction were assessed. Assuming the commonly chosen a-level of 0.05 was chosen, the results indicate the following:

·    The p-value for the interaction term (0.910) is greater than 0.05. Thus, the interaction of the subject and degree factors is not significant, and you are free to consider the effects of the individual factors separately.

·    The p-value for the subject factor is given as 0.000. This means that the actual p-value is less than 0.0005. Since this is less than the chosen a-level of 0.05, it means the effect of subject on salary is significant. In other words, the mean salary is different for the different subject areas.

·    The p-value for degree is also less than 0.05, indicating that there is also a significant effect of degree on salary. In other words, the mean salary of the professors is different for the different degrees.