How to perform a meta-analysis in R

[@braincraft]

Thank you so much for this!!! I'm performing my first meta-analysis and this video is an amazing resource. It's good to update research methods when necessary, but I wouldn't take the KZbin comments to heart about poor audio quality on a free educational resource 😅

[@dsquintana]

Thanks for your comment, I’m glad you found this video!

[@jamescutler428]

I'm so happy I found this video! Perfect intro for me for diving right into a meta-analysis quickly in R. This is exactly what I was looking for!

[@bradeymenz539]

Excellent video Daniel, thank you so much for detailing the code, data and rationale behind the analysis/method

[@dsquintana]

Glad you liked the video, thanks for the comment!

[@vangelis9911]

Daniel a meta-regession video would be really helpfull

[@mrlettel]

Dear Daniel, thank you so much for the video. I think it would be better if you take it more specify, e.g., this function is applied to random-effect or fix-effect model. Because the function for these 2 model is different and the result will also be varied a lot.

[@Toshitaka_work-ft3ip]

Thank you for creating this video. Great a supplement to textbooks!

[@dsquintana]

Happy to hear this!

[@mihiradhikary7075]

Thanks!

[@Prof_Julie]

Incredibly useful! Thank you very much for this and all the other videos on meta analysis. They are very useful to complement your paper. Not to mention that such a cute teacher helps a lot with concentration ;)

[@dsquintana]

Thanks for the comment, glad to hear these videos are useful!

[@alexfang3401]

Thank you so much! I just want to use your data materials to practice meta-analysis conducting steps, could you please share your data and code materials?

[@davidlau8211]

if I have collected the t-statistics, how do I convert them to the effect size "yi"? Please help.

[@sixucai]

First of all, thank you for the video, Daniel! I have a question tho regarding the dependent effect sizes. As you demonstrated, it is possible to account for this dependency using the robu() function, but what should the workflow look like? How should I follow up with other procedures such as finding the influential studies, plotting a forest plot, test for bias and so on? Or do I do the meta-analysis as if all the effect sizes are independent and only use the robu function to see if the results are robust? Thank you in advance for your time!

[@saadalhumaid3959]

Would have loved to watch how you make subgroup analysis to lower the heterogeneity

[@dsquintana]

Thanks for the suggestion! I hope to include this in a future screencast

[@terrorcamel3601]

This video saved me countless hours! Thank you so much!

[@saadalhumaid3959]

Thanks very much, mate! Best regards from Saudi Arabia

[@dsquintana]

Thanks for the comment, glad this was useful!

[@paulzauzau3131]

Thanks so much for this video. Its a very straight forward and it has helped me to understand how to perform a meta analysis in R.

[@dsquintana]

Thanks for taking the time to write a comment, happy to hear that this was helpful

[@rectorrector8637]

Dear Dr Quintana I have read with interest your paper on using R for meta-analysis published in 2015. I wanted to access the supplementary file but the link in the article simply takes me back to the original paper. Would you be so kind to assist with a copy? Tyrone

[@dsquintana]

Hi, I believe this is the link you're looking for as this contains all the supplementary scripts and data github.com/dsquintana/corr_meta

[@rectorrector8637]

@@dsquintana Thanks so much Doc

[@sampoornanandi5671]

Hello, is there a way to plot the moderator analyses? Thanks!

[@dsquintana]

Yes, it’s possible to plot both categorical and continuous moderator analysis. I hope to demonstrate this in a future video

[@sampoornanandi5671]

great, thanks!@@dsquintana

[@actuarialaman7753]

Doing a great job sir Have a great success in future

[@dsquintana]

Thank you 🙏

[@saeedewu129]

Thank you very much for you video. Just wanted to know if we can refer Lispay and Wilson (2001) if we use your way to do meta analysis?

[@jimpyjohn2474]

Hi Daniel, I am conducting an MA with only 4 studies included and the Isquare test is 0%. Is there an easy interpretation for this? I assume I'm probably missing some criterium for being able to run the test.

[@dsquintana]

I squared is not sensitive to the number of studies in a meta-analysis, but rather, the precision of each study (i.e., sample size). I would recommend reporting a range of heterogeneity measures as all common measures have pros and cons bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/heterogeneity.html

[@tommasob.jannini3899]

Dear Dr. Quintana, thanks for this very useful video. I just have a rapid question. In the Likelihood Ratio test, is it correct that for rejecting the null hypothesis you need a p-value of below 0.1 (in your example p-value was 0.08 and you said that the test was significant)? Thanks again.

[@akshittapillai]

where is the ci dataset

[@janinemedinahuerta3096]

Thank you so much for this video, Daniel. It has been super helpful! Would you mind sharing a citation for the publication bias test using the weightr function? Thank you!!

[@dsquintana]

Glad this video was helpful! Here is the official citation for the weightr package: Kathleen M. Coburn and Jack L. Vevea (2019). weightr: Estimating Weight-Function Models for Publication Bias. R package version 2.0.2. CRAN.R-project.org/package=weightr. For a number of useful citations related to this method, see the 'selection models' section of our recent preprint psyarxiv.com/75bqn/

[@janinemedinahuerta3096]

@@dsquintana You're amazing!! Thank you so much!! I'm working on my first meta analysis and I'm scared and excited at the same time. I have probably watched your video and read your script and accompanying paper about 15 times

[@dsquintana]

@@janinemedinahuerta3096 Also check out this excellent (and free) textbook bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/ Good luck!

[@SPC700.]

Is it possible to attach risk of bias table by the side of a forest plot?

[@DNeuropsych]

Such a helpful video! Saving my disseration! Thank you :) One question, I've followed this step by step but my estimators (tau^2 etc) keep coming back as 0, and H^2 as 1. I feel that this isn't correct, can you (or anyone) think where I may have gone wrong for this to occur? Thanks :)

[@DNeuropsych]

Never mind I figured it out haha

[@mariamnaser5981]

which r version i can find packages plz

[@richelle-joychia3991]

Hi Daniel, thank you very this insightful video, it is super helpful! I was wondering if the same moderator analyses (res.modage

[@mikehector1]

Dear Daniel, this was really helpful. Great tutorial indeed. Learned a lot. Thanks for the tips about the 'publication bias', really helpful. I followed your tutorial and was able to reproduce the whole thing!! However, just a little bit of problem I found, probably nothing significant to you, was unable to get rid of printing of 'Common effect model' in the forest plot picture. Both the 'Common effect model' and 'Random effect model' label showing in my Forest plot graph, but can see only 'random effect model' label in your your forest plot graph though. Any suggestion would be highly appreciated. Thank you.

[@fionnualarogers8443]

Does this method also work for Cohen's d?

[@tiyashuda]

can i join research with you? im interest in meta analysis research

[@U2006gmx]

You're the boss! Thanks a lot.

[@laxmanbisht2638]

Very useful video.

[@dsquintana]

Thanks for your comment, glad you found this useful!

[@sivaranjanmurugesan2419]

hello sir whether you can take class on full meta analysis for me privately

[@CS-si8kl]

Why such good audio quality ? It distracts me from the technical contents of the video. JK 😂😂. Thanks for the video BTW.

[@mariamnaser5981]

First, thank you very much for your good effort,I have this problem can u help me

[@georgegl3192]

OMG, such poor audio quality. Have you been recording this with a potato? ;)

[@dsquintana]

Haha the audio quality of the original video pretty much sounded like that. Happy that a new version is out!

[@kevind.shabahang]

give 'er

[@jorgemmmmteixeira]

Why such a poor video quality? jk

[@dsquintana]

😂

[@actuarialaman7753]

@@dsquintana yeh sir the quality is not good we are not able to see properly Plz do something