# Thread: What statistical test should I use to check to ratio?

1. ## What statistical test should I use to check to ratio?

I have data for two sets of sample in the form of frequency of three category.Say A,B & C are three category and I have data for sample 1 as
Freqency of A=200,Frequency of B=300,frequency of C=100.For sample 2 Frequency of A=100,Frequency of B=200 & frequency of C=50.I want to check whether the ratio of A & B for sample 1 is same with the ratio of A & B for sample 2.Is there any test using which I can check my hypothesis?

2. ## Re: What statistical test should I use to check to ratio?

Please wait for responses from others too!

First of all, find out if there is a significant correlation between between Sample 1 and 2 on the category A and later on category B. It measures to some extent that there is some relationship between category A and B.

Finding out a ratio is a simple arithmetic. Just find out the ratio in sample 1 and and later the ratio in sample 2. Then, compare.

3. ## Re: What statistical test should I use to check to ratio?

Sorry,I did not get your answer.Why do I need to check correlation between two sample on Category A & Category B?

My Question is if I'm getting ratio A/B for sample 1 is X1=0.5 and for sample 2 X2=0.8 & I want to make a Null hypotheis X1=X2 & Altrenative hypothesis X1 Not equal to X2 then will be my test procedure & test statistic?

4. ## Re: What statistical test should I use to check to ratio?

Sorry,I did not get your answer.Why do I need to check correlation between two sample on Category A & Category B?

My Question is if I'm getting ratio A/B for sample 1 is X1=0.5 and for sample 2 X2=0.8 & I want to make a Null hypotheis X1=X2 & Altrenative hypothesis X1 Not equal to X2 then will be my test procedure & test statistic?
If I understand correctly you simply want to test for a (significant) difference in two proportions. You can use a t-test to do this.

5. ## Re: What statistical test should I use to check to ratio?

Originally Posted by duskstar
If I understand correctly you simply want to test for a (significant) difference in two proportions. You can use a t-test to do this.
There are two samples (each with sample size 1). You can't really do any inferential procedures here. The best you do is look at the sample ratios and compare them.

6. ## Re: What statistical test should I use to check to ratio?

Originally Posted by Dason
There are two samples (each with sample size 1). You can't really do any inferential procedures here. The best you do is look at the sample ratios and compare them.

You're very right, I didn't read the post correctly sorry. If there is only a sample size of 1 it doesnt make sense to compare statistically.

7. ## Re: What statistical test should I use to check to ratio?

Originally Posted by Dason
There are two samples (each with sample size 1). You can't really do any inferential procedures here. The best you do is look at the sample ratios and compare them.
No no.Sample size is not one.My sample size is 20

 Tweet

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts