# How can I compare the dispersion of two groups to reinforce the results of PERMANOVA ?
**URL:** https://forum.qiime2.org/t/how-can-i-compare-the-dispersion-of-two-groups-to-reinforce-the-results-of-permanova/24376
**Category:** General Discussion
**Created:** [October 7, 2022, 4:23pm UTC](https://forum.qiime2.org/t/how-can-i-compare-the-dispersion-of-two-groups-to-reinforce-the-results-of-permanova/24376 "2022-10-07T16:23:03Z")
**Posts on this page:** 3
**Page:** 1
### Author:  [@SingeunOh](https://forum.qiime2.org/u/SingeunOh)
#### Post date: [October 7, 2022, 4:23pm UTC](https://forum.qiime2.org/t/how-can-i-compare-the-dispersion-of-two-groups-to-reinforce-the-results-of-permanova/24376/1 "2022-10-07T16:23:03Z")
Hi!
I am currently working with Qiime 2 to analyze alpha & beta diversity with microbiome data of 48 sample.
(Animal stool samples)
I finished working of Beta diversity and got the data of significance. (Weighted UniFrac, Bray-curtis)
However, I received this message and need some help from you.
- The message is as below\*
" PERMANOVA can be sensitive to variation in dispersion - please examine whether the treatments differed in dispersion"
/ My question is, what type of variation test would be proper to compare the distance squares of microbial data and how can I do with python or R ? ( I am seriously considering Scheffe's test)
- As far as I know, the input data of dispersion comparison is the distance square of each object
If I choose the right type of test, then Can I get the distance square data of each group from Qiime2 ?
- And also, I want to hear any kind of recommendation to solve the question above from anyone who are familiar with the PERMANOVA statistics.
: the exact reason and meaning of the question or what kind of tools to solve the question are very welcomed because I can't understand that the reason to do another check for dispersion difference because dispersion test is already considered to examine the PERMANOVA results (F-test).
Just explanation about calculating the PERMANOVA is also welcomed !
How can I calculate the P, Fp in here?
Thank you !
---
### Author:  [@jwdebelius](https://forum.qiime2.org/u/jwdebelius)
#### Post date: [October 7, 2022, 5:05pm UTC](https://forum.qiime2.org/t/how-can-i-compare-the-dispersion-of-two-groups-to-reinforce-the-results-of-permanova/24376/2 "2022-10-07T17:05:00Z")
Hi @SingeunOh,
Welcome to the :qiime2: forum!
I would recommend using the permdisp test (it's called betadispr in R). It's implemented in the beta-group-significance function; check the documentation to see which flag you need to use. The one caveat to this test is that it requires categorical data.
Best,
Justine
Here's some more discussion of the test, and idea:
> [@adonis, betadiver, and betadisp](https://forum.qiime2.org/t/adonis-betadiver-and-betadisp/9364):
>
> An [earlier post](https://forum.qiime2.org/t/beta-group-significance-plots-not-representative-of-permanova-test/3320) by @CarlyRae first introduced me to the fact that when examining beta diversity I need to consider not only the results of a [multi](https://docs.qiime2.org/2019.1/plugins/available/diversity/adonis/?highlight=beta%20group) or [univariate](https://docs.qiime2.org/2019.1/plugins/available/diversity/beta-group-significance/?highlight=beta%20group) anova, but also examining if the within-group distances to a centroid were equivalent among my groups. I'm wondering how to best interpret the situation where everything turns out to be significant. That is, if the results of both PERMANOVA and PERMDISP suggest significant differences, where do most people go next? I've run these tests b…
> [@Determining dispersion for beta diversity](https://forum.qiime2.org/t/determining-dispersion-for-beta-diversity/12747):
>
> Hey all, I have read quite a few other posts on this but I am not sure I understand the solutions fully. To sum up my experiment, I have conditioned soil with different cover crop treatments for 3 months and I want to see if the fungal communities differ under those cover crop treatments. I understand the PERMANOVA results but is there a test to tell if the dispersion in one group is larger than another (if some treatments cluster tightly vs others being spread out on the PCoA)? You can visual…
> [@betadisper in R via QIIME2 results](https://forum.qiime2.org/t/betadisper-in-r-via-qiime2-results/16168):
>
> For one of our paper, we need to calculate betadisper before PERMANOVA analysis. Is there a way to calculate betadisper statistics from qiime2 results in R or in qiime2 directly? Thanks
---
### Author:  [@SingeunOh](https://forum.qiime2.org/u/SingeunOh)
#### Post date: [October 8, 2022, 6:53am UTC](https://forum.qiime2.org/t/how-can-i-compare-the-dispersion-of-two-groups-to-reinforce-the-results-of-permanova/24376/3 "2022-10-08T06:53:51Z")
Thank you so much jwdebelius !! It would be a big help ^^
I will try the recommended method.