Solving Anova Table? (stats problem)

[zoiberg137]

Here is the problem:

The data referred to in this question were collected on 41 employees of a large company. The company is trying to predict the current salary of its employees from their starting salary (both expressed in thousands of dollars). The following is ANOVA table for simple linear regression:

Its not letting me copy and paste the table, but I am given three numbers here. The df, the model ss, and the total ss. I need the f statistic and probability. How can I possibly solve this?

Every video I look up shows how to find the f and p statistics show how to insert numbers into three or four lists, which makes sense, except that I don't have three or four lists worth of numbers. I only have the three numbers to work with. I am using a ti 84. Please help! Thank you!

 

[zoiberg137]

ok, here is an update. I get how to fill in the anova table for the ss and ms column. I think i understand how to calculate the f statistic by dividing the first ms number by the second ms number. What I still don't understand is how to get the probability. What numbers exactly do I put into my "lists"?

 

[thelema418]

1. Take a screen shot of the table. Save the screen shot as a bmp or jpg and upload it into a post. This might help.

2. I don't know TI products specifically, but there should be an option under statistics for F distributions. I think it is going to Fcdf( or something like that. This requires the F-calculated value, the upper value (infinity), and the degrees of freedom). For infinity you can write in a very large number or E99.

 

[statdad]

IF you have two of the three sums of squares you can find the third:
SSTotal=SSFit + SSE

If you have one of the numbers of degrees of freedom you can find the others since you know the sample size
Once you have the sums of squares and the degrees of freedom you can find the required mean squares (mean square in each case = sum of squares divided by corresponding df)

Once you have the mean squares you can calculate the test statistic, and once you have that you can use it and the relevant dfs to find a p-value or required critical value

All you need to do is use the provided information and formulas to build the pieces of the complete table