Hi all, thanks for watching! I have more tutorials using Excel for meta-analysis and systematic review practices, the full list is found here: th-cam.com/play/PLMDcoG6cxhI0EaLoG9jG7fmf22YIUjD-o.html
This video and the rest of the series have been fantastic and so helpful. Thank you! Your passion for this topic is clearly evident as well, which makes it that much better. Is there a way to make funnel plots for publication bias through Excel? Not sure if you made a video on this already, but I didn't see one. Thanks again for uploading these!
Hi I would like to know how to make a Forest plot of the log of the odds ratio and its 95% confidence interval for comparison of categorical predictor variables in logistic regression? I have multiple variables I need to compare. I would like to have 6 forest plots combined together. How can I do this in excel?
Hi and thank you very much for the amazing work. QUESTION: why in hell my title crossbars don't show the negative part of the bar? I've tried everything, including "show both sides" obv... I feel like my excel is just bugged sometimes, like when I select a series of values or bars in the graph not all of them appear selected (light blue dots around them) but the actually modify if I change values 😅. I'm lost guys😞
Thanks for the great video - would be lost without it! I have one more questions: can negative effect sizes also be included in a meta-analysis? Bc. when i want to use the formula for the SE = ES/SQRT(ES*N) you cant take the SQRT of a negative effect...
Yes you need to include negative effect sizes -- since sampling error alone can create study outcomes with positive, null, or negative effects. Calculating the SE of effect sizes is not conventional, SD and 95%CI are the standard presentation approach. I am not familiar with the equation you are using for meta-analysis.
Thank you for providing the video. It really helpful. In my experience, to do practice while watching video, it much easier to learn using two screens, one for practices, another for watching the video. Then, for different excel version, it may have different location to find the function, a little bit stressful in the beginning. Also, need a little bit manually adjustment for y axis (can not use 1, 2, 3, etc) and perform risk ratio value (manually move to the right).
Sir I'm trying to learn R but I have no idea how to make a forest plot on it. I can make l'abbe plot but that's it. Also can you use the odds ratio for this?
yes, any effect size (e.g., odds ratio, hedges' d, correlations) can be used, this video focuses on log scale effects, but you can ignore that section for regular effect sizes
Once again sir, I doff my hat to you. This video has been a life saver! My only error was looking elsewhere before remembering to check your channel. You're right, this process is tedious but until I get R, or any other software, these Excel tutorials you upload are my lifeline to doing meta-analysis. Thank you for spending half an hour for my sake Dr. Marc 😅
I am making a forest plot with 3 subgroups. How do I calculate the point estimates for each subgroup on excel? (i.e. how do I calculate the diamond's location for each subgroup?)
This is a more challenging task, and perhaps too difficult to discuss here in the comments -- but maybe I can do another video to help with these special cases :-)
Really helpful! I'm a basically R user, but the OR plot i made with packages was not satisfactory. Your video enables me to make a more 'customized' plot for me. thx a lot!🍺
Any suggestions how you can change the size of square boxes (ie effect sizes), in the R plot you have shown the boxes are different sizes but in the final excel file they are all the same size.
Excel is not friendly with boxes of different shapes or sizes, but it is possible -- you'll need to begin with a bubble plot rather than a scatter plot
These are estimated using meta-analysis techniques, I have a video on how to calculate these, if you wish to use excel: th-cam.com/video/i66Lf2a_Pa8/w-d-xo.html
@@LajeunesseLab Thanks so much for this ! Can I create a forest plot just to view a range of numbers for quality of life values reported in different studies: example , for diabetes related complications ? Not meant to do a meta analyses but just narrative synthesis ?
Awesome tutorial and great teaching! Q: how would you address the weights issue? both the calculation (haven't seen you talk about that in any of the other videos) and in the plot?
For the weights, excel is still not friendly to customize in this way and easily have different sizes. But excel gets updated all the time, so maybe in the near future this issue will be address :-)
Confidence Intervals are calculated from the standard deviation of each effect size, the the confidence of pooled means are estimated using meta-analysis
![[TH] FS vs BME - VCT Ascension Pacific - Lower Bracket Final](http://i.ytimg.com/vi/wnygqsR2Fig/mqdefault.jpg)
Hi all, thanks for watching!
I have more tutorials using Excel for meta-analysis and systematic review practices, the full list is found here:
th-cam.com/play/PLMDcoG6cxhI0EaLoG9jG7fmf22YIUjD-o.html
Wow! Incredible use of x,y scatter plot. Transforming what would otherwise be the vertical axes as data points and using data labels is brilliant.
This was SO helpful! I didn't have time to learn R for the paper I'm working on, and this is EXACTLY what I needed. Thank you!
You are so welcome!
This was the first time I've ever had to create a forest plot and this video was extremely helpful! THE BEST!
Awesome! Thank you!
Thank you so much for this video! Finally a good source for making a forest plot on Excel. It helped me a lot. Greetings from Germany!
You're very welcome!
This is art level of using excel!!! Thank you us much!!
This video and the rest of the series have been fantastic and so helpful. Thank you! Your passion for this topic is clearly evident as well, which makes it that much better. Is there a way to make funnel plots for publication bias through Excel? Not sure if you made a video on this already, but I didn't see one. Thanks again for uploading these!
May I ask where did you get the 7.39? Thank you so much
overall great vid but why is the error bar for Comstock & webster (6.53) going all the way to 7.39?
great description sir...very helpful
Thanks!
This was soo good. I like how excel can do a lot of things (if we know how to use it). Thanks for the video
Glad you enjoyed it!
Not all heroes wear capes
How you created RF model and the other model state that too
Thx for your precious efforts. Please try to share the excel that you used as an attachment so we can use it as a template.
Hi I would like to know how to make a Forest plot of the log of the odds ratio and its 95% confidence interval for comparison of categorical
predictor variables in logistic regression? I have multiple variables I need to compare. I would like to have 6 forest plots combined together. How can I do this in excel?
Thank you so much! You saved me!!
You're welcome!
Hi and thank you very much for the amazing work. QUESTION: why in hell my title crossbars don't show the negative part of the bar? I've tried everything, including "show both sides" obv... I feel like my excel is just bugged sometimes, like when I select a series of values or bars in the graph not all of them appear selected (light blue dots around them) but the actually modify if I change values 😅. I'm lost guys😞
Thanks for the great video - would be lost without it!
I have one more questions:
can negative effect sizes also be included in a meta-analysis? Bc. when i want to use the formula for the SE = ES/SQRT(ES*N) you cant take the SQRT of a negative effect...
Yes you need to include negative effect sizes -- since sampling error alone can create study outcomes with positive, null, or negative effects. Calculating the SE of effect sizes is not conventional, SD and 95%CI are the standard presentation approach. I am not familiar with the equation you are using for meta-analysis.
Thank you for providing the video. It really helpful.
In my experience, to do practice while watching video, it much easier to learn using two screens, one for practices, another for watching the video.
Then, for different excel version, it may have different location to find the function, a little bit stressful in the beginning. Also, need a little bit manually adjustment for y axis (can not use 1, 2, 3, etc) and perform risk ratio value (manually move to the right).
wonderful.. thank you a lot..
Thanks for watching!
Hello, please I would like to ask you some questions. How can I contact you?
Sir I'm trying to learn R but I have no idea how to make a forest plot on it. I can make l'abbe plot but that's it. Also can you use the odds ratio for this?
yes, any effect size (e.g., odds ratio, hedges' d, correlations) can be used, this video focuses on log scale effects, but you can ignore that section for regular effect sizes
Is it possible to make a forest plot video for R? Like with event rates and stuff extracting from excel? Just can't find any decent tutorial online.
Thank you so much !!!
Thank you so much
Glad it was useful!
Once again sir, I doff my hat to you. This video has been a life saver! My only error was looking elsewhere before remembering to check your channel. You're right, this process is tedious but until I get R, or any other software, these Excel tutorials you upload are my lifeline to doing meta-analysis. Thank you for spending half an hour for my sake Dr. Marc 😅
Also you're a god you saved my dissertation
Thank you so much!
in the original graph there are different box sizes for different weights of the studies, does anyone know how can we incorporate this into the graph?
Thank tou very much, amazing explanation!
Super I appreciate your teaching and skill
Thanks Amir!
Great tutorial, if you want to use excel
I am making a forest plot with 3 subgroups. How do I calculate the point estimates for each subgroup on excel? (i.e. how do I calculate the diamond's location for each subgroup?)
This is a more challenging task, and perhaps too difficult to discuss here in the comments -- but maybe I can do another video to help with these special cases :-)
Really helpful! I'm a basically R user, but the OR plot i made with packages was not satisfactory. Your video enables me to make a more 'customized' plot for me. thx a lot!🍺
Glad I could help!
Any suggestions how you can change the size of square boxes (ie effect sizes), in the R plot you have shown the boxes are different sizes but in the final excel file they are all the same size.
Excel is not friendly with boxes of different shapes or sizes, but it is possible -- you'll need to begin with a bubble plot rather than a scatter plot
Thanks for the video, that is very useful. RevMan software is also a useful toolkit to use and straightforward to use.
Great tip!
didn't want to use R so this was massively helpful!!
haha, this is great!
great demonestration. Thanks a lot for your time
Glad it was helpful!
What does the RE and FE Models mean and how can I compute them?
These are estimated using meta-analysis techniques, I have a video on how to calculate these, if you wish to use excel:
th-cam.com/video/i66Lf2a_Pa8/w-d-xo.html
Thanks for this video. This is amazing. Is the forest plot specific to meta-analysis? Can we use it for observational studies?
Yes you can use it, as long as the outcome has confidence intervals
@@LajeunesseLab Thanks so much for this ! Can I create a forest plot just to view a range of numbers for quality of life values reported in different studies: example , for diabetes related complications ? Not meant to do a meta analyses but just narrative synthesis ?
Great video and super useful! Thank you so much!
Glad you enjoyed it!
This is a great video. Very helpful
Glad it was helpful!
you helped me do well in my report. thanks
Glad I could help!!
How to calculate risk ratio in excel?
How to make diamond shape, sir? I made everything except this part
unfortunately mimicking the multi-sized diamonds is not possible unless you format the marker with your own built-in marker-type or size
Awesome tutorial and great teaching!
Q: how would you address the weights issue? both the calculation (haven't seen you talk about that in any of the other videos) and in the plot?
For the weights, excel is still not friendly to customize in this way and easily have different sizes. But excel gets updated all the time, so maybe in the near future this issue will be address :-)
How did you get the confidence intervals?
Confidence Intervals are calculated from the standard deviation of each effect size, the the confidence of pooled means are estimated using meta-analysis