/sci/ - Science & Math » Thread #11104917

23KiB, 460x454, a1Q4VyG_460s.jpg

View Same Google iqdb SauceNAO

I'm a fucking brainlet

Anonymous Wed 30 Oct 11:42:50 2019 No.11104917 View Reply Original Report

Quoted By: >>11104999

Please help me with this Basic Statistics question.

I have data from 52 participants (in procent) from year 1 and data from the same 52 participants (in procent) from year 2.

How do I calculate the mean change in proportions from year 1 to year 2? I need to answer with a 95% confidence interval. The T-test is two-sided.

Anonymous

Anonymous Wed 30 Oct 2019 12:37:20 No.11104999 Report

Quoted By: >>11105023

>>11104917
https://www.statisticshowto.datasciencecentral.com/probability-and-statistics/confidence-interval/

This will prob answer your question. Though I recommend spss / R for it since you just click a few buttons on spss for it or copy paste in R. But I dno if you need to meme and do it by hand (and then never do it again lel)

Anonymous

Anonymous Wed 30 Oct 2019 12:47:04 No.11105023 Report

Quoted By: >>11105347

>>11104999
I just need to give the confidence interval as my answer. What confuses me is that the data I have been given is in procent, if it wasn't for this I could have easily calculated the Confidence intervals for the difference between two proportions.

I just need to know what my p-hat is here and also my p-hat1 and p-hat2.

Anonymous

Anonymous Wed 30 Oct 2019 13:09:16 No.11105075 Report

Quoted By:

OP here. So I think I calculated p-hat to 12,59519, but I don't feel like this number makes sense. Used the mean of population1 and population2 (keep in mind that both these pops were denoted in procent)

Anonymous

Anonymous Wed 30 Oct 2019 15:02:32 No.11105347 Report

Quoted By: >>11105402 >>11105424

>>11105023
Did you divide by 100 to get a proper proportion between 0 and 1?

Anonymous

Anonymous Wed 30 Oct 2019 15:22:26 No.11105402 Report

Quoted By:

>>11105347
>100 to get a proper proportion between 0 and 1?
No, but I did now. Thanks.

It's says that I should use a two-sided t-value, but isn't this a Z-test? I'm confused

Anonymous

Anonymous Wed 30 Oct 2019 15:28:45 No.11105424 Report

Quoted By:

>>11105347
>Did you divide by 100 to get a proper proportion between 0 and 1?
What I did was to sum all of pop1 (remember, the data is in procent ex 15,4%) and sum all of pop2. Then I divided pop1 by N and did the same for pop2.

And with the help of your comment I have divided pop1 and pop2 again with 100 this time to get a proportion between 0 and 1. Is this correct?

Anonymous

Anonymous Wed 30 Oct 2019 16:39:56 No.11105595 Report

Quoted By:

bump

Capcode	All Only User Posts Only Moderator Posts Only Admin Posts Only Developer Posts
Show Posts	All Only With Images Only Without Images
Deleted Posts	All Only Deleted Posts Only Non-Deleted Posts
Ghost Posts	All Only Ghost Posts Only Non-Ghost Posts
Post Type	All Only Sticky Threads Only Opening Posts Only Reply Posts
Results	All Grouped By Threads
Order	Latest Posts First Oldest Posts First

Your latest searches