# Normal and un-normal distribution confusion

#1
So one set of data showed to be normal and another showed to be un-normally distributed. I've gone back and forth with the data but I don't know which statistic test to use. I eventually decided to use a 2 sample t-test but still no I don't know if I have chosen right :/

Can someone shed a little light int the hell of statistics?
0
4 years ago
#2
(Original post by JaffaNeko)
So one set of data showed to be normal and another showed to be un-normally distributed. I've gone back and forth with the data but I don't know which statistic test to use. I eventually decided to use a 2 sample t-test but still no I don't know if I have chosen right :/

Can someone shed a little light int the hell of statistics?
Before we think about what statistic to use, we need to know what scientific question is being asked of the data! If I guess that you've got two random samples and you've been asked whether they could have come from the same underlying population distribution, or whether they have come from two distinct populations with the same means, would I be close to right?
0
#3
(Original post by Gregorius)
Before we think about what statistic to use, we need to know what scientific question is being asked of the data! If I guess that you've got two random samples and you've been asked whether they could have come from the same underlying population distribution, or whether they have come from two distinct populations with the same means, would I be close to right?
The question relates to the height difference between males and females to try and see if there is a statistically significant difference or not.
0
4 years ago
#4
(Original post by JaffaNeko)
The question relates to the height difference between males and females to try and see if there is a statistically significant difference or not.
OK, that's nice and definite! The t-test is actually fairly resilient to departures from normality, if the sample sizes are reasonably big. Just make sure you use the version of the t-test that does not assume equal variances in the underlying populations.
0
#5
(Original post by Gregorius)
OK, that's nice and definite! The t-test is actually fairly resilient to departures from normality, if the sample sizes are reasonably big. Just make sure you use the version of the t-test that does not assume equal variances in the underlying populations.
Well I used a 2 sample test to start with but now i think its wrong and Im assuming it was a Mann-Whitney U test. It just doesn't make any sense at all.
0
4 years ago
#6
(Original post by JaffaNeko)
Well I used a 2 sample test to start with but now i think its wrong and Im assuming it was a Mann-Whitney U test. It just doesn't make any sense at all.
If you tell me a little more about where you're stuck conceptually, I might be able to help. But with what you've indicated so far, the problem you presented would quite reasonably be solved with an appropriate t-test or with a Mann-Whitney test.
0
#7
(Original post by Gregorius)
If you tell me a little more about where you're stuck conceptually, I might be able to help. But with what you've indicated so far, the problem you presented would quite reasonably be solved with an appropriate t-test or with a Mann-Whitney test.
I changed the test I used to Mann-Whitney but now i just need help figuring if my study is a between or within subject design.

I assume because I'm using two IV groups to one DV measure that it would be a between design.
0
4 years ago
#8
(Original post by JaffaNeko)
I changed the test I used to Mann-Whitney but now i just need help figuring if my study is a between or within subject design.

I assume because I'm using two IV groups to one DV measure that it would be a between design.
As this doesn't appear to be an experimental design, it's a bit of a stretch to apply this sort of terminology to it - but I suppose that by analogy we could call it a between group design.
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

#### Y13's - If you haven't confirmed your firm and insurance choices yet, why is that?

I am waiting until the deadline in case anything in my life changes (5)
17.24%
I am waiting until the deadline in case something else changes (e.g. exams/pandemic related concerns) (3)
10.34%
I am waiting until I can see the unis in person (4)
13.79%
I still have more questions before I make my decision (3)
10.34%
No reason, just haven't entered it yet (7)
24.14%
Something else (let us know in the thread!) (7)
24.14%