Close Home Forum Sign up / Log in

very easy stats / SPSS question

B

I need to test if the difference between two results of option 'A' is significant.

Person 1 - option A = 165 (68%) option B = 79 (32%)
Person 2 - option A = 139 (60%) option B = 91 (40%)

I know it's simple, but I'm new to this. Chi-square? T-Test?

Thanks :)

Non parametric = Mann-Whitney U test, parametric = student's t test. You will need to decide which fits your study. Ginga xx

C

on first glance this looks to me like you'd use a chi sqaured but it kinda depends on what type of data this is. What are the numbers? it reads like it's the number of times they''ve chosen an option (A or B)but are they measurements of some kind?
do you just have two participants or is there other unshown data?

sorry lots of questions!

B

thanks both. Yes, definitely chi squared, just looked back at something sup said.

Yes, the figures represent choices between A and B. There are other participants, but I need to see if each pair is significantly different from each other.

B

Just had an unsolicited email from sup saying 'thought you might like this' - a nice simple description of why I need chi squared for this. He mentioned it in a meeting but I hadn't asked for clarification cos I didn't want to look stupid.

Maybe there is no such thing as a stupid question :)

12065