Solved

# T-Test vs Chi-Squre in Excel PAQ

Posted on 2000-04-06
220 Views
Tess, in a previous discussion, asked the following question:

I am hoping someone can help me out. I want to run a TTEST in excel. And I am not sure how to do it. I understand where to get my Array 1 and 2, but I am unsure what to put in for tails and type. Does the array have to be numbers? I will give you an example of what I am trying to do. I have data on patients who are take either Drug 1 or Drug 2, some of these patients develop a reaction, and we want to see if there is a statistical significance between the number of patients taking Drug 1 who have a reation to the number of patients taking Drug 2 who have a reaction? What numbers would be included in the array, the total count of patients on Drug 1 & 2, or do I select all records (do these have to be numbers). As well, we want to calaculate the statistical significane of the the number of patients who were African-AMerican and Caucasian. ANy info or help will be greatly appreciated.

This appeared in: http://www1.experts-exchange.com/bin/Q.10319556

However, since that is now a PAQ, it will cost points to have a look.

During that discussion it was evident that Tess required a more comprehensive solution to the problem.  I personally felt that the type of analysis suggested may have been unsuitable - that of a Student's Distribution T-Test.  Further, I was desirous of making the results available to others who were involved in the discussion.

To that end I state the following:

Tess communicated further details of the problem privately.
I then constructed a WorkBook that provided/included:
The data was set out in a contingency table
The relevant chi-square test applied
Along with several little 'tricks' to enable further analysis etc.

It is virtually impossible to detail how this was achieved in EE.  Therefore, if any would like to obtain a copy of the outcome, which was accepted by Tess as meeting the requirements of the question, then please feel free to request a copy by emailing me:

d64471023a@hotmail.com

Also, so that this lot doesn't hang around in current questions I will have to accept an answer from someone - it will be graded A, but of course no points.  The first to respond in any way will get the A.

Dave

0
Question by:TigerMan
[X]
###### Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

• Help others & share knowledge
• Earn cash & points
• 2
• 2

LVL 9

Accepted Solution

antrat earned 0 total points
ID: 2692137
Hey I'll take the "A" I could do with at least one "Good Answer" e-mail this month.

antrat
0

LVL 5

Author Comment

ID: 2692145
antrat,

I will leave this out for the paint to dry for a couple of days - then accept your response.

Dave
0

LVL 5

Author Comment

ID: 2695300
Thanks antrat.  That workbook was a real hit - it is now being incorporated as the model for presenting a case that there is no difference between the two drugs in terms of causation etc.

Chow for now.

Dave
0

LVL 9

Expert Comment

ID: 2695340
It's a nice piece of work and should prove the case.

antrat
0

## Featured Post

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

Access developers frequently have requirements to interact with Excel (import from or output to) in their applications.  You might be able to accomplish this with the TransferSpreadsheet and OutputTo methods, but in this series of articles I will di…