Statistical Analysis: Pass/Fail Correlation

J

Jim J.

I have an Excel spreadsheet (.xlsx) that I want to use to determine
correlation. There are 30 columns and 690 rows (1 row of “Labelsâ€, and 689
rows of data). The 2nd column is Pass (“1â€) or Fail (“0â€) data. My goal is
to determine what factors most strongly influence the part to Pass, and which
most strongly influence the part to Fail.
I am no wizard at statistical analysis, so I would appreciate any help I can
get. I can make the spreadsheet available to you so you can play with it.
There is no proprietary data in the spreadsheet (everything has been coded
with a numerical value). My email is (e-mail address removed)
Thanks in advance!
 
M

Mike Middleton

Jim -

One method is called "discriminant analysis," often appropriate for a
categorical dependent variable and categorical/numerical explanatory
variables, usually requiring advanced statistical software.

For the special case (your case) of a binary dependent variable, you can
perform _approximate_ discriminant analysis using Excel's Regression tool. I
gave very detailed instructions in the "Regression Using Categorical
Variables" chapter of my book "Data Analysis Using Microsoft Excel."

However, the Regression tool is limited to 16 explanatory variables, and you
have 30. So, you may want to find someone with SPSS or similar software to
help you with your analysis.

- Mike
http://www.MikeMiddleton.com


I have an Excel spreadsheet (.xlsx) that I want to use to determine
correlation. There are 30 columns and 690 rows (1 row of “Labelsâ€, and 689
rows of data). The 2nd column is Pass (“1â€) or Fail (“0â€) data. My goal is
to determine what factors most strongly influence the part to Pass, and
which
most strongly influence the part to Fail.
I am no wizard at statistical analysis, so I would appreciate any help I can
get. I can make the spreadsheet available to you so you can play with it.
There is no proprietary data in the spreadsheet (everything has been coded
with a numerical value). My email is (e-mail address removed)
Thanks in advance!
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top