# Thread: Comparing two %, what test can I use?

1. ## Comparing two %, what test can I use?

Good afternoon forum fellows,

I am currently conducting a prospective research study at my hospital. I haven't used my stats skills in awhile and was wondering if anyone can offer some help to me. I have two % I want to compare. The first % is the actual % determined from raw data while the second % is a sample generated by a computer(e.g., 30/500 vs 20/80. Is it possible to compare these two percentages with any test that could show that there is/isn't a significant difference between the actual % vs sample %?

Please let me know if you have more questions.

Thanks,

Jason

2. ## Re: Comparing two %, what test can I use?

while the second % is a sample generated by a computer(e.g., 30/500 vs 20/80)
Does the software actually generate 100 pseudo-subjects,
or do you mean, the software calculates the expected percentage
as 20%? In the former case, fourfold table with Chi² could be
an option; in the latter case, binomial test, or one-sample Chi².

With kind regards

K.

3. ## The Following User Says Thank You to Karabiner For This Useful Post:

jyw003 (03-16-2015)

4. ## Re: Comparing two %, what test can I use?

Can you describe the second group in more detail?

5. ## Re: Comparing two %, what test can I use?

Originally Posted by hlsmith
Can you describe the second group in more detail?
I guess I'll try to explain the study design in more depth.

The first group is me identifying all patients that have A and B within a given population. The second group is a computer that only picks a sample of the given population and identifies all patients that have A and B. I understand that the second group could miss out on potential patients that have A and B and that is why I would like to compare the true % compared to the sample %.

 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