How is the analyze multi variance factor data and what is the correct sas code?

[Chathu]

Hi dear sir/ madam,

I want to know about suitable analyzing method and sas code to 3 factors. Here it is those factor,

1). Packaging method- Vacuum and Non Vacuum

2). Storage condition - Room and Cold temperature

3). Varieties- V1, V2, V3, V4, and V5

It is confusing to analyzing me I try to several way by using sas 9.1. Here it is data I attach to below (It is name "Fat Content").

I kindly request you to help me to analyze the data.

Thank you,

Chathu

[PaigeMiller]

We can't advise on a "suitable analyzing method" because you have not told us what questions are to be answered by this data and by this analysis.

[Chathu]

Dear sir, I want to know proc glm with manova. because It contain 3 factor I think MANOVA suitable to it. And I want standard error with proc glm. I try to several time with several sas code to proc glm with manova. But It is complicate to me. Thank you, Chathu

[PaigeMiller]

Show us the code you have written.

And you still haven't really told us what question you are trying to answer. Saying you want to use PROC GLM doesn't really tell us what question you are trying to answer.

[Chathu]

Dear Sir, Here it is my code, and I want express all single factors and their interaction by analyzing, also I want express their significant difference between single factors and significant interaction among factors combination. I am not professional for data analyzing by sas 9.1 I kindly request to help me to solve these matter. Data pag; input variety packaging storecondition rep fc; cards; 1 1 1 1 2.32 1 1 1 2 2.52 1 1 2 1 2.36 1 1 2 2 2.56 1 2 1 1 2.23 1 2 1 2 2.53 1 2 2 1 2.60 1 2 2 2 2.50 2 1 1 1 4.40 2 1 1 2 4.50 2 1 2 1 4.00 2 1 2 2 4.50 2 2 1 1 4.77 2 2 1 2 4.57 2 2 2 1 4.11 2 2 2 2 4.15 3 1 1 1 1.25 3 1 1 2 1.45 3 1 2 1 1.16 3 1 2 2 1.56 3 2 1 1 1.02 3 2 1 2 1.62 3 2 2 1 1.59 3 2 2 2 1.69 4 1 1 1 3.13 4 1 1 2 3.53 4 1 2 1 3.40 4 1 2 2 3.30 4 2 1 1 3.42 4 2 1 2 3.62 4 2 2 1 3.20 4 2 2 2 3.68 5 1 1 1 2.42 5 1 1 2 2.62 5 1 2 1 2.15 5 1 2 2 2.45 5 2 1 1 2.97 5 2 1 2 2.67 5 2 2 1 2.31 5 2 2 2 2.51 run; proc glm; class variety packaging storecondition; model fc = variety|packaging|storecondition; manova h= _All_; means variety/duncan; means packaging/duncan; means storecondition/duncan; means variety packaging storecondition; lsmeans variety/stderr; lsmeans packaging/stderr; lsmeans storecondition/stderr; run;

[PaigeMiller]

To enter your code, click on the {i} icon and paste it into the dialog box that appears.

[Chathu]

Data pag;
input variety packaging storecondition rep fc;
cards;
1 1 1 1 2.32
1 1 1 2 2.52
1 1 2 1 2.36
1 1 2 2 2.56
1 2 1 1 2.23
1 2 1 2 2.53
1 2 2 1 2.60
1 2 2 2 2.50
2 1 1 1 4.40
2 1 1 2 4.50
2 1 2 1 4.00
2 1 2 2 4.50
2 2 1 1 4.77
2 2 1 2 4.57
2 2 2 1 4.11
2 2 2 2 4.15
3 1 1 1 1.25
3 1 1 2 1.45
3 1 2 1 1.16
3 1 2 2 1.56
3 2 1 1 1.02
3 2 1 2 1.62
3 2 2 1 1.59
3 2 2 2 1.69
4 1 1 1 3.13
4 1 1 2 3.53
4 1 2 1 3.40
4 1 2 2 3.30
4 2 1 1 3.42
4 2 1 2 3.62
4 2 2 1 3.20
4 2 2 2 3.68
5 1 1 1 2.42
5 1 1 2 2.62
5 1 2 1 2.15
5 1 2 2 2.45
5 2 1 1 2.97
5 2 1 2 2.67
5 2 2 1 2.31
5 2 2 2 2.51
run;
proc glm;
class variety packaging storecondition;
model fc = variety|packaging|storecondition;
manova h= _All_;
means variety/duncan;
means packaging/duncan;
means storecondition/duncan;
means variety packaging storecondition;
lsmeans variety/stderr;
lsmeans packaging/stderr;
lsmeans storecondition/stderr;
run;

[PaigeMiller]

Since you have only one Y variable, you cannot do MANOVA here.

Other than that, your code seems to work properly.

[Chathu]

Dear sir, Here you mention about only one y variable. please kindly tell me what is that? I think it is "fc" And I want to really need to know if not suitable MANOVA what is suitable method. That is the question sir. And I think 3 way factorial Complete Randomize Design suitable for that. But unfortunately I don't know sas code for that.

[PaigeMiller]

It is impossible to do MANOVA when the only response variable is FC. MANOVA requires two or more response variables. The suitable analysis method is what you have programmed, without the MANOVA statement.

[Chathu]

Dear Sir,

I understood. what about ANOVA and 3 way factorial CRD can I use for that? 

[PaigeMiller]

---

@Chathu wrote:

Dear Sir,

I understood. what about ANOVA and 3 way factorial CRD can I use for that? 

---

That's what your code does, it analyses the 3 way factorial completely randomized design via ANOVA.

[Chathu]

Dear Sir, I thought 3 way factorial NAOVA for that. That' mean PROC ANOVA suitable for that. And I want to know clearly SAS code to CLASS and MODEL statement. I want to take Duncan Multiple Rang Test and Standard error with PROC ANOVA. But I don't know SAS code for that. Thank you Chathu

[PaigeMiller]

PROC GLM in this case will produce the same results as PROC ANOVA. The DUNCAN option in the MEANS statement gives you the Duncan test. See https://documentation.sas.com/?docsetId=statug&docsetVersion=14.3&docsetTarget=statug_glm_syntax15.h...

[Chathu]

Dear Sir,

Actually it is work. But I don't know whether correct or not. And I don't know possibility to use MANOVA to that data set.