Watch
#1
So for my psych class we had to conduct an experiment and me and my friend decided to do a disney experiment to see the difference between boys and girls, it was a score out of 20, to see which gender got more correct. We have to do a statistical test and I dont know which one??? There are two groups (boys and girls) and they both experienced the same conditon so its independent design, i think? Its a test of difference. I dont know whether it is interval or ordinal data.

I've tried the Mann whitney test but its coming up with a negative U value and thats not supposed to happen. I think it might be unrelated T but im not sure? I need to do it for tommorow.
0
3 years ago
#2
(Original post by Katieee2017)
So for my psych class we had to conduct an experiment and me and my friend decided to do a disney experiment to see the difference between boys and girls, it was a score out of 20, to see which gender got more correct. We have to do a statistical test and I dont know which one??? There are two groups (boys and girls) and they both experienced the same conditon so its independent design, i think? Its a test of difference. I dont know whether it is interval or ordinal data.

I've tried the Mann whitney test but its coming up with a negative U value and thats not supposed to happen. I think it might be unrelated T but im not sure? I need to do it for tommorow.
girls and boys are doing the same conditions?
0
#3
Yeah
0
3 years ago
#4
Scratch that, I don't know what I'm talking about...
0
3 years ago
#5
So it's quasi-experimental, because what you're varying is the gender of the groups. What's the DV, performance on a Disney quiz?

What about a between subjects t-test?

I know there may be issues with normality of data, but the t-test is very robust.
0
3 years ago
#6
(Original post by Twinpeaks)
So it's quasi-experimental, because what you're varying is the gender of the groups. What's the DV, performance on a Disney quiz?

What about a between subjects t-test?

I know there may be issues with normality of data, but the t-test is very robust.
Isn't a between subjects design for when your groups are in different conditions (I.e. boys get a car quiz and girls get a Disney quiz). I think the Mann Whitney U is correct and I think that having a negative U is not a problem? Does that not just mean that the second group had lower ranks than the first - I'm not sure about that, which is why I deleted my first post.
0
3 years ago
#7
(Original post by _Sinnie_)
Isn't a between subjects design for when your groups are in different conditions (I.e. boys get a car quiz and girls get a Disney quiz). I think the Mann Whitney U is correct and I think that having a negative U is not a problem? Does that not just mean that the second group had lower ranks than the first - I'm not sure about that, which is why I deleted my first post.
There are different conditions, one group of participants is male, the other female. That's what the OP is manipulating, independent variable= gender, dependent variable= performance. Although like I said, it's quasi-experimental.

See this example from laerd statistics- https://statistics.laerd.com/spss-tu...statistics.php
They've used a t-test to test for difference between genders in graduate salaries. IV= male, female. DV= salary.

Like I said, the only possible issue is with normality of data, but generally for a large group of participants it's fine.

I can't really speak for a U, but I'm pretty sure for a t-test you can ignore the sign (negative/ minus), because you'd use a graph to illustrate the direction of the difference anyway.So I imagine it's the same with the Mann Whitney?
0
3 years ago
#8
(Original post by Twinpeaks)
There are different conditions, one group of participants is male, the other female. That's what the OP is manipulating, independent variable= gender, dependent variable= performance. Although like I said, it's quasi-experimental.

See this example from laerd statistics- https://statistics.laerd.com/spss-tu...statistics.php
They've used a t-test to test for difference between genders in graduate salaries. IV= male, female. DV= salary.

Like I said, the only possible issue is with normality of data, but generally for a large group of participants it's fine.

I can't really speak for a U, but I'm pretty sure for a t-test you can ignore the sign (negative/ minus), because you'd use a graph to illustrate the direction of the difference anyway.So I imagine it's the same with the Mann Whitney?
I think the example I found when I looked it up confused me... In any case, do the t-test and the Mann Whitney and if they give similar results, pick one and use it.
0
X

new posts
Back
to top
Latest
My Feed

### Oops, nobody has postedin the last few hours.

Why not re-start the conversation?

see more

### See more of what you like onThe Student Room

You can personalise what you see on TSR. Tell us a little about yourself to get started.

### Poll

Join the discussion

Yes (254)
34%
No (493)
66%