what interpretation?

I have attached the calculations... However, what's the interpretation of these? :/

Attachment missing...

Two questions of my assignment:

1. "Is the average salary different for persons with different categories of education?"
2. "Are the conclusions from the previous question changed if gender is included in the analysis?"

I have attached the calculations... However, what's the interpretation of these? :/

Taking question 1 first, then answer is "yes". The p-value attached to the F-statistic in the Analysis of variance table is very small. This is the p-value assciated with the covariate that represents the category of schooling. If you then look at the parameter estimates, you can see a monotonic increasing value of the average salary as the level of education increases. Also note that the value of Adjusted $R^2$ is small (around the 10% mark), so the level of education explains only a little bit of the overall variation in salary levels.

For question 2, the answer is still "yes" but the details are more complex. Can you take the answer I gave you for the first part and see why I make these conclusions?

I think so... Because the AdjR^2 is only about the 10% wouldn't that make the difference insignificant?

Adjusted R^2 is around 20% in the second analysis. Roughly speaking, this means that adding the sex variable is doubling the amount of explanatory information.

However, take a look at the estimates of the mean effects of the education categories (that appear to be given relative to the top category).

There is something I don't understand ... In question 2 I'm asked whether the conclusion changes if I include gender... So it does change because the AdjR is is greater... But it seems that salary and education level are still more "linked" than salary and gender.
Is it that gender affects the salary, but that education is the variable that "affects" salary the most?

Given that adjusted R^2 is around 10% for the first model and 20% for the second, it looks as though they each have roughly the same explanatory power.

Now, I don't recognize the programme output (whcih programme is it?) and it looks as though the second analysis only has four degrees of freedom, which suggests that the model is not fitting the interaction term between gender and education. Also, the slight problem with model fit suggests that the interaction is missing. So this doesn't look like a standard two-way AoV. Can you force it to fit the interaction?

Given that adjusted R^2 is around 10% for the first model and 20% for the second, it looks as though they each have roughly the same explanatory power.

Now, I don't recognize the programme output (whcih programme is it?) and it looks as though the second analysis only has four degrees of freedom, which suggests that the model is not fitting the interaction term between gender and education. Also, the slight problem with model fit suggests that the interaction is missing. So this doesn't look like a standard two-way AoV. Can you force it to fit the interaction?

Perhaps this is what you mean?