For you to graph the confidence interval properly, you have to create two columns for "Graph upper 95% CI (Upper CI minus RR)" and Graph for lower 95% CI (RR minus lower CI). These are the variables you need when specifying the values for your error bars.
Hello, Karina. I'd watch this even if I don't need it. But I just want to thank you so much for making a difference in the lives of many. Thank you so much for helping me during my thesis write-up. You made the learning process more enjoyable! 😃
Hardly ever comment on UA-cam videos, but this has been so helpful, thank you. Really like how quick and to the point you are, and that you don't ask for anything. Weirdly makes me want to support you on Patreaon or Kickstarter - are you an evil genius, lol? Many thanks again!
you need to subtract odds value with upper bound for CI+ value and lower bound with odds ratio for CI- value. then only you can get the actual plot of lower and upper confidence interval
This is correct. The goal is to specify the length of the error bar on the negative and positive sides, which will involve subtracting the lower value (of the estimate vs. upper or lower bound of CI) from the higher value.
I think confidence intervals we see are not intervals per se....but actually how much to add to or subtract from main effect estimates (odds ratios) to obtain the confidence intervals.
My values are effect sizes, and some of the lower bound of the confidence intervals are negative. When I select the data for the positive and negative error values, the error bars do not show negative values. Is there something else I need to do? By the way thank you for your video
For you to graph the confidence interval properly, you have to create two columns for "Graph upper 95% CI (Upper CI minus RR)" and Graph for lower 95% CI (RR minus lower CI). These are the variables you need when specifying the values for your error bars.
Hello, Karina.
I'd watch this even if I don't need it. But I just want to thank you so much for making a difference in the lives of many.
Thank you so much for helping me during my thesis write-up. You made the learning process more enjoyable! 😃
Hardly ever comment on UA-cam videos, but this has been so helpful, thank you. Really like how quick and to the point you are, and that you don't ask for anything. Weirdly makes me want to support you on Patreaon or Kickstarter - are you an evil genius, lol? Many thanks again!
Great video! Very clear and concise. Thank you!
This is exactly what I'm looking for. Thanks Karina!
So short, so simple, very good explained! Look so many explanation but only yours bring light into the dark.
Thank you for saving my meta-analysis 🙏🏽🙏🏽🙏🏽
you need to subtract odds value with upper bound for CI+ value and lower bound with odds ratio for CI- value. then only you can get the actual plot of lower and upper confidence interval
This is correct. The goal is to specify the length of the error bar on the negative and positive sides, which will involve subtracting the lower value (of the estimate vs. upper or lower bound of CI) from the higher value.
Wow! This was so simple and straight forward. Thank you so much for saving my brain!
That was awesome. Simple and straightforward. Thank you!
Thank you for this video.
Hi, I'm wondering why the lower and upper confidence intervals don't match up with the line we created.
Hi you should calculate the different value
Did you figure this out?
I think confidence intervals we see are not intervals per se....but actually how much to add to or subtract from main effect estimates (odds ratios) to obtain the confidence intervals.
whay do you mean "different value"?
@@xiaoli7352
have you resolve the issue
My values are effect sizes, and some of the lower bound of the confidence intervals are negative. When I select the data for the positive and negative error values, the error bars do not show negative values. Is there something else I need to do? By the way thank you for your video
this is very helpful, thank you!
It's really helpful!! This works well compared with other methods introduced by other youtuber
Thank you!
INCREDIBLE! Thank you so much!!
Hi Karina, how can i get the CI+ and CI- i have all the raw data and i also calculated the odds ratio
What kind of error bars do you add? Standard error or STD?
dude you are awesome, thanks
Thank you
Hi. Could you help with creating forest plot please?
I have question how to match exact letters or word in excel like eg: Anil-C1 AnilPaul-C1 so i need C1 status and confirmed as C1 status
How to match exact words or letters in two data like example: anil-C1 and AnilPaul-C1 so i need C-1 status is exact match
Hi, I'm wondering why the lower and upper confidence intervals don't match up with the line we created. any one knows?
i found it...
put the figures extracted from ci values to HR or OR for error bars
@@intercardioLSDHi, I am trying to work out how to do this, what are HR and OR?
I have question how to match exact letters or word in excel like eg: Anil-C1 AnilPaul-C1 so i need C1 status and confirmed as C1 status