Hello,
I am looking for help on how to break down the "Model" or "Regression" source in the ANOVA table. I am trying to get the Seq SS, Seq MS, F-value, and P-Value. In Proc Reg you can inlcude the "SS1" option in the model statement, but this gives me t-values and p-values. But if I use Proc Mixed with the "method=type1" option, I get all that I want, except this ANOVA table does not contain the total column (this is not too big of a deal, but it makes comparing SS for full and reduced models a bit nicer). Just curious if one can get the best of both. Is it possible to get Proc Reg to break down the "Model" source in the ANOVA table? Or is there a better procedure to use?
I am using SAS 9.4.
Thank you,
Jeff S. O.
Example Code:
data ACL;
input Subj ACL Vocab Abstract SDMT;
datalines;
1 6 28 36 70
2 5.4 34 32 49
3 4.7 19 8 28
4 4.8 32 28 47
5 4.9 22 4 29
6 4.5 24 24 23
7 6.3 24 12 40
8 5.9 18 14 50
9 4.1 31 20 32
10 4.8 14 8 27
11 4 24 8 33
12 4.5 34 36 40
13 5.8 29 20 66
14 6 27 34 46
15 4.5 15 10 26
16 4.7 31 24 42
17 5.9 30 32 42
18 4.7 17 26 52
19 4.7 26 26 35
20 3.8 18 28 41
21 6 32 26 58
22 5.6 19 16 41
23 4.8 14 10 13
24 5.8 27 36 62
25 4.5 21 20 46
26 4.8 26 28 52
27 4.7 22 14 63
28 4.5 22 26 42
29 6 30 26 66
30 5.6 26 26 55
31 6.3 22 28 55
32 5.2 22 28 43
33 4.8 16 10 48
34 5.8 32 36 47
35 4.8 26 30 50
36 3.7 11 16 29
37 4.5 18 8 17
38 4.9 14 2 39
39 5 30 32 31
40 3.9 30 36 61
41 3.7 31 18 45
42 5.6 23 18 56
43 4.8 28 20 53
44 5.6 17 8 29
45 6.6 31 30 63
46 4.3 12 6 19
47 4 18 6 23
48 4.2 23 8 40
49 5.6 22 6 20
50 3.4 13 8 2
51 4 30 22 41
52 4.5 29 24 44
53 4.9 28 36 51
54 4.2 20 8 37
55 4.5 32 36 56
56 4.8 33 36 37
57 6 26 20 76
58 4 26 8 42
59 4.5 13 10 20
60 4 27 16 48
61 4.7 40 40 54
62 6 25 32 53
63 4.5 20 8 39
64 4.8 26 10 35
65 6.6 26 30 63
66 4.1 16 16 17
67 4.5 31 24 44
68 6.6 30 36 47
69 4.9 10 19 35
;
run;
proc reg data=ACL;
model ACL = Vocab SDMT / lackfit SS1;
run;
proc mixed data=ACL method=type1;
model ACL = Vocab SDMT;
run;
PROC GLM should give you the ANOVA table you are looking for. PROC REG also has the requisite information, but you have to manually back compute the pieces.
PROC GLM should give you the ANOVA table you are looking for. PROC REG also has the requisite information, but you have to manually back compute the pieces.
Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 25. Read more here about why you should contribute and what is in it for you!
Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.
Find more tutorials on the SAS Users YouTube channel.