My first time with statistics - Chi Square help needed.

#1
Hi guys, first post and first time EVER playing with statistics:eek: .

I am looking for ways to summarise two sets of data I have collected. They are usability characterises from a 7-point Likert based survey. I showed a user 1 interface (filled in the survey), then showed them another (filled in the survey again).

My maths isn't too hot so I have been taking a look at Excel and the CHITEST function. So far I have got thus far:

Q1 Q2 Q3 up to Q10...
5 3 2
5 3 1
1 4 3
onwards for all users tested...

I ran the range for each question through the CHITEST function with the expected range as 4 (the middle figure of the Likert 1 to 7 range). For each row of questions I have CHI figures such as:

0.3314 0.0489 0.9885

Are these numbers the p-value or something else?

My question really is how do I interpret these numbers which Excel has provided me with?

Also, I would like to group questions to form categories for easier analysis. For example Q1 and Q2 would be combined because both statements refer to 'interface style' for example. At the moment I see no way to do this as the CHITEST function only allows for 1 range, and I need 2 or 3.

Sorry to bombard you with questions but you do seem to be a knowledgeable bunch!

Many thanks,
DrZoidberg
 
#2
don't do CHITEST in Excel since it's a goodness-of-fit chi-square test. You're more likely to need chi-test of homogeneity.

What are you trying to prove? Tell more about your hypothesis, so I cam help you out.

btw, those are p values, and only the midle one is significant (and barely).

best wishes!