What is the Bonferroni Correction? 

Much clearer than my lecturer, thanks!

Simple and easy to understand. Thank you so much for sharing!

I hope you are blessed for all of eternity. Without you, I may have been living on the streets with $50,000 worth of student debt.

Concise and crystal clear!! Thanks so much!!

thank you, you may have just helped me to pass my exam

Thank you very much for making this topic. It is very helpful.

Thank you! Very comprehensive explantation!

Excellent presentation!

Thanks for making the video, it is really helpful!!

Amazing! I prefer approach 2 because I just like to always use the traditional 0.05 as my guide.

Thank you for making it easy to understand

Thank you so much for your great explanationt~ it really helped a lot

Beautiful explanation

Great explanation! RU-vid is a rescue when you're stuck with a bad teacher at uni ^^

infuriating that my lecturer manager to convolute such a basic concept!! thankyuou so much for simplifying it and making it so so much more clear

very useful explanation

Many thanks, do we do the same for the Wilccoxon signed rank and the Mann-Whitney U test?

Thank you so much!

Thanks for sharing

Subbing for sure. Thank you!!

hi, thank you for the video!!-.recently I read a trial that calculates the relative risk fo two procedures but the result is lower than you would expect (I used epi info to calculated with IC). the authors mentioned that they used the Bonferroni correction. Could this difference be related to Bonferroni´s correction?

Nice lecture

Hello I am comparing gene expression data between two groups (control and diseased) and have to test seven genes.... for each gene the CT values are different for each group. I have used Mann Whitney U test. The sample size is small, Diseased: 16 and controls: 8. Kindly please suggest I need to do the correction test for the observed p values for each gene or not?

excellent.

Hello, thank you for the video! What if I conduct one-way ANOVA between a,b,c groups on y variable and then regression with a/b/c groups on and some other parameters on z variable, shall I correct the analysis P value,since technically a,b,c are used in both tests?

Thanks for this nice and easy-to-follow tutorial. I have questions... I compared whether Group 1 is smaller than Group 2. To get the data for this comparison, I used a formula with two parameters to play around. Let's say, I have 12 sets of data for comparison, i.e., G1 vs. G2 using scenario #1, G1 vs. G2 using scenario #2, ..., and G1 vs. G2 using scenario #12. I don't want to compare which parameter combination is the best. The main goal is to test if G1 < G2. Should I use Bonferroni correction?

I think R does the second approach right? It adjusts the p-values of the estimates rather than changing the acceptance level

where can I find that formula or proof for the alpha familywise? Neve saw it before. Thanks

I have a follow up question. I conducted a within group study where all participants performed 3 variation of a task. At first, I ran Friedman's ANOVA and found statistical significance. Secondly I ran wilcoxon signed rank test between two of the three variations. I did it twice (according to my initial hypothesis) between variation A and B and between A and C . So should I divide the p value by 3 for all the three tests? I mean the p values from friedman ANOVA, and two wilcoxon tests. Thanks in advance for your reply

Is Bonferroni correction: original p value set at 0.05 divided by the number of dependant variables?

How do we apply Bonferroni to t-tests and their non-parametric equivalents? SPSS has the procedure for one way analysis of variance

Great video, thanks! I have a follow-up question. What happens when there are multiple comparisons but not on the same sample of data? For example, when two groups (control, treatment) are compared across several outcome measures (e.g. language, memory, attention, behaviour, etc.). In this case you run multiple comparisons but not on the same sample each time. Does this require correction as well? Thanks!

The Bonferroni correction is an adjustment made to P values when several dependent or independent statistical tests are being performed simultaneously on a single data set. Please explain what is the meaning of single data set there....

25% (57/229)

what does it mean that you conduct test on the same samples? Why is this familywise, or how are the tests dependent on each other? :)

Here after Dream's response

Hi, Suppose I have a customer level dataset and each customer belong to a segment and I have to calculate the uplift for each segment using Anova. Now, I want to understand the application of Bonferroni correction - since I will be performing Anova for each segment seperately, do I need to apply the correction or not? What I think is since the customer base for each of the segment is mutually exclusive we don't need to apply it. Please let me know what should be actually done.

Help please. I'm stuck from the beginning. What does "same sample of data" mean? Why would measure test with "same sample of data"??

Minecraft speed running fraud making me a stat major

Is 0.5 really good enough? This means that 1/20 statistics studies that accept this alpha value are wrong.

Thank you for the explanation. But 3 commercials in a 10 minute video is too many - you should reconsider how to profit from these.

I mmmmm

Amazing! Now it's clear to me. I am using pairwise.wilcox.test in R with Bonferroni correction and it's using the second approach. but unfortunately, I can't find a reference to talk about this approach. could you please provide a reference for the citation of the second approach?