BookmarkSubscribeRSS Feed
Khaladdin
Quartz | Level 8

Hi all,

 

I want to compare two percentages. I think Chi-Square test is appropriate for this. So, I have the following datasets;

 

Group    Number of Success         Total number

1                   5                                     10

2                   4                                      30

 

I need to compute the percantages of number of success in the first and second group and then compare them whether the difference between this comparision is statistically significant by using Chi-Square test. So the should be one more column:

 

Group    Number of Success         Total number            Percentages

1                   5                                     10                              50%   (5/10*100)

2                   3                                      30                             10%   (3/30*100)

 

And then I want to see whether the difference between percantages (50% and 10%) is statistically significant by using Chi-Square test. 

 

 

3 REPLIES 3
Reeza
Super User

Chi Square would use the raw numbers from the 2x2 table, not the percentages. 

 

To compare percentages you can use PROC TTEST but you need to make sure to include the correct N and possibly the standard deviation. You also need to factor in a continuity correction.

 

I recommend PROC FREQ instead of trying to test the percentages. The data lends itself to that format naturally. 

Khaladdin
Quartz | Level 8
Dear Reeza,

Many thanks for reply. Could you please explain how to use PROC FREQ? could you please give me the sample codes?
PaigeMiller
Diamond | Level 26

An example showing how to use PROC FREQ to compare two proportions is:

http://support.sas.com/kb/22/561.html

--
Paige Miller

SAS Innovate 2025: Call for Content

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!

Submit your idea!

Discussion stats
  • 3 replies
  • 2925 views
  • 0 likes
  • 3 in conversation