T-test

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

Midna
Posts: 3
Joined: Thu Feb 04, 2016 2:14 pm

T-test

Postby Midna » Thu Feb 04, 2016 2:26 pm

Hi, I'm having trouble using a T-test.

I asked participants about Big Five characteristics, once when in a social situation and again (same characteristics) in a personal situation.
So for example I want to know if there is a significant difference between a person's answer on how extravert they are in a social situation vs their answer on how extravert they are in a personal situation.

First: I'm not sure if I should use a one-sample t-test (I think so ...) or a paired t-test seeing as the data comes from the same person.
Second: I have done the one-sample and the paired one and both show ,000 sign for all my data. Which to me seems very strange. As some of the data (Agreeableness) only has 0,3 difference in mean if you compare the social against the personal situation (if you put it in a graph the means are nearly identical). I'm thinking it's maybe because of my data. It's a 7 point scale (from "very like me" to "totally not like me") because as I understand data should be continuos for this test (and mine is not). But my professor told me to use a t-test, so not sure what I should do ... Normally we get a whole course about this but I was allowed to do a shorter program because of a pervious degree, but now I'm having a lot of trouble with SPSS since I've never used it before until a few months ago :).

Any advice would be great! (hope I used al the right terms since English isn't my native language :))
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: T-test

Postby GerineL » Tue Feb 09, 2016 3:36 pm

Midna wrote: First: I'm not sure if I should use a one-sample t-test (I think so ...) or a paired t-test seeing as the data comes from the same person.
paired sample. one sample t test is if you want to compare a group mean to a specific mean (e.g., see if the mean of a variable is significantly different from 100).
It is paired because observations are not independent (they come from the same person).
Second: I have done the one-sample and the paired one and both show ,000 sign for all my data. Which to me seems very strange.
I really wonder what you have done with the one sample t-test, because it does not seem possible to do this with your design. What have you clicked on?? Maybe you compared both variables means with 0 by accident.
As some of the data (Agreeableness) only has 0,3 difference in mean if you compare the social against the personal situation (if you put it in a graph the means are nearly identical). I'm thinking it's maybe because of my data. It's a 7 point scale (from "very like me" to "totally not like me") because as I understand data should be continuos for this test (and mine is not). But my professor told me to use a t-test, so not sure what I should do ... Normally we get a whole course about this but I was allowed to do a shorter program because of a pervious degree, but now I'm having a lot of trouble with SPSS since I've never used it before until a few months ago :).
people tend to use likert scales in these types of situations, but indeed it is not correct. if it is ordinal, use Wilcoxon signed ranks test.

also, consider that your different measures of big five may be interrelated, in which case you may want to test them together instead of one by one.
Midna
Posts: 3
Joined: Thu Feb 04, 2016 2:14 pm

Re: T-test

Postby Midna » Thu Feb 11, 2016 8:26 am

I did the one sample by entering one of the two groups and calculating the mean of the other (and entering that one as specific mean).

I just tried the Wilcoxon signed ranks test and it also says that every one is significant ... (all .000), I had already tested them together but same effect. Don't know what I'm doing wrong :(. Or is it possible the have a significant effect when the means hardly vary?

Thanks for the help!
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: T-test

Postby GerineL » Thu Feb 11, 2016 9:10 am

I have not seen your data so I can't say for sure whether you did something wrong or not, but sure, it is possible. If you have a large sample size, or if there is small variation in both groups, it is possible for a difference to be significant even if it seems small.
Midna
Posts: 3
Joined: Thu Feb 04, 2016 2:14 pm

Re: T-test

Postby Midna » Thu Feb 11, 2016 3:15 pm

I think I used SPSS correctly and I haven't edited any of my data, they just range between 1-7 as in the original Likert scale.
Also my sample size is 220 which I know enhances the changes of a significant result. But it still seems really counter intuitive and maybe I'm missing something in the interpretation? Here's my output if that helps?
You do not have the required permissions to view the files attached to this post.
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: T-test

Postby GerineL » Mon Feb 15, 2016 8:36 am

It looks okay to me!

Who is online

Users browsing this forum: No registered users and 1 guest

cron