topic Re: Confidence interval with alpha = 0.01 in SAS Programming

Confidence interval with alpha = 0.01

Miah — Sun, 22 Apr 2018 01:03:37 GMT

I am trying to determine whether or not the mean of Potassium for the cereals whose manufacturer is General Mills or Kelloggs is different with a significance level of 0.01. I am having trouble getting the right observations selected, and therefore tusing PROC TTEST for the hypothesis test. Here it is my code and the data set:

DATA cereal;
INFILE "/folders/myfolders/cereal.txt"
DLM=',' FIRSTOBS=2 DSD MISSOVER;
INPUT Name :$50. Manufacturer $ Type $ Calories Protein Sodium Fiber Carbohydrates Sugars Potassium Vitamins Weight Cups;
RUN;

PROC PRINT DATA=cereal;
RUN; 

DATA cereal; 
SET Cereal;
IF Manufacturer='General Mills' | Manufacturer='Kelloggs'; 
RUN;

PROC PRINT DATA=Cereal;
RUN; 

PROC TTEST DATA=Cereal ALPHA=0.01;
CLASS Manufacturer;
VAR Potassium;
RUN;

PROC PRINT DATA=Cereal;
RUN;

Re: Confidence interval with alpha = 0.01

Reeza — Sun, 22 Apr 2018 01:21:36 GMT

What exactly is your question? Is there something wrong with the code?

@Miah wrote:

I am trying to determine whether or not the mean of Potassium for the cereals whose manufacturer is General Mills or Kelloggs is different with a significance level of 0.01. I am having trouble getting the right observations selected, and therefore tusing PROC TTEST for the hypothesis test. Here it is my code and the data set:
DATA cereal;
INFILE "/folders/myfolders/cereal.txt"
DLM=',' FIRSTOBS=2 DSD MISSOVER;
INPUT Name :$50. Manufacturer $ Type $ Calories Protein Sodium Fiber Carbohydrates Sugars Potassium Vitamins Weight Cups;
RUN;

PROC PRINT DATA=cereal;
RUN; 

DATA cereal; 
SET Cereal;
IF Manufacturer='General Mills' | Manufacturer='Kelloggs'; 
RUN;

PROC PRINT DATA=Cereal;
RUN; 

PROC TTEST DATA=Cereal ALPHA=0.01;
CLASS Manufacturer;
VAR Potassium;
RUN;

PROC PRINT DATA=Cereal;
RUN; 

Re: Confidence interval with alpha = 0.01

Miah — Sun, 22 Apr 2018 01:25:30 GMT

Yes. It is not generating the output that I need. I need to determine whether or not the mean of Potassium for the cereals whose manufacturer is General Mills or Kelloggs is different with a significance level of 0.01

Re: Confidence interval with alpha = 0.01

Reeza — Sun, 22 Apr 2018 02:01:41 GMT

What exactly is your hypothesis?

Your current code tests, is the Potassium level between General Mills and Kelloggs cereals different.

@Miah wrote:

Yes. It is not generating the output that I need. I need to determine whether or not the mean of Potassium for the cereals whose manufacturer is General Mills or Kelloggs is different with a significance level of 0.01

Re: Confidence interval with alpha = 0.01

Miah — Sun, 22 Apr 2018 02:06:35 GMT

The H0 is whether or not the means are different.

Re: Confidence interval with alpha = 0.01

Reeza — Sun, 22 Apr 2018 02:08:26 GMT

Between what?

@Miah wrote:

The H0 is whether or not the means are different.

Re: Confidence interval with alpha = 0.01

Miah — Sun, 22 Apr 2018 02:10:24 GMT

The mean difference of the variable Potassium between the cereals whose Manufacturer is General Mills and Kelloggs...

Re: Confidence interval with alpha = 0.01

Reeza — Sun, 22 Apr 2018 02:19:38 GMT

Ok, your code doesn't read the data in correctly. Check your data in Cereal data set, right after you import it.

Run a PROC FREQ on the Manufacturer column.

And read your log - does it show any errors for the code?

Also, never ever code like this:

data cereal;
set cereal;

At this point you've destroyed your original data set and no longer have access to the original data, so you need to re-read it in from scratch which is fine in this case because your data is small. If it was a larger data set that would slow you down a lot each time.

Once you make those changes, it works as expected.

@Miah wrote:

The mean difference of the variable Potassium between the cereals whose Manufacturer is General Mills and Kelloggs...