# Metadata error message-no categorical data
**URL:** https://forum.qiime2.org/t/metadata-error-message-no-categorical-data/13028
**Category:** User Support
**Created:** [January 1, 2020, 6:08pm UTC](https://forum.qiime2.org/t/metadata-error-message-no-categorical-data/13028 "2020-01-01T18:08:27Z")
**Posts on this page:** 5
**Page:** 1
### Author:  [@bloc\_gen](https://forum.qiime2.org/u/bloc_gen)
#### Post date: [January 1, 2020, 6:08pm UTC](https://forum.qiime2.org/t/metadata-error-message-no-categorical-data/13028/1 "2020-01-01T18:08:27Z")
Hi,
I have been working on a fecal microbiome study and I am a beginner. I got this error message saying that there are no categorical data in my metadata file. I have a categorical data column and I still got this error. I have attached pictures of the error message and the metadata file. Please let me know how to rectify this error!
Thanks a million!
 
---
### Author:  [@fedarko](https://forum.qiime2.org/u/fedarko)
#### Post date: [January 1, 2020, 9:10pm UTC](https://forum.qiime2.org/t/metadata-error-message-no-categorical-data/13028/2 "2020-01-01T21:10:04Z")
Hi Sai,
That definitely is a categorical column! Another possibility that could be causing this error is that this categorical column only contains "unique values", as the error message says.
Looking at the one categorical column in the screenshot of your metadata (`body-site`), all of the three shown values are unique -- instead of just being `feces` three times, you have `feces0`, `feces35`, and `feces70`. It looks like this is because each row's `body-site` value also includes its `Day` value. **I think this might be causing the error you're seeing.**
However, fixing that (by changing these values all back to `feces`) might not be enough. Are these three samples (`FecalMicrobiome6`, `FecalMicrobiome7`, and `FecalMicrobiome9`) the only ones in your metadata file, or is the screenshot just a snippet of the full metadata? Looks like this visualizer also requires that the categorical column "doesn't consist of exactly one value" -- so if all of your samples have `feces` as their `body-site`, and that's the only categorical column in your metadata, this will still fail again.
From googling the error message, I'm guessing you're trying to run `qiime diversity alpha-group-significance`? Here are a few other discussions from the forum about this error in that visualizer.
> [@Faith\_pd visualization in Alpha diversity](https://forum.qiime2.org/t/faith-pd-visualization-in-alpha-diversity/8033):
>
> Hi, I having some problem to generate alpha diversity. Until this step everything is ok, but when I do this command: qiime diversity alpha-group-significance --i-alpha-diversity fiath\_pd\_vector.qza --m-metadata-file metadatafile.tsv --o-visualization faith\_pd\_vector.qzv I receive error message: Plugin error from diversity: Metadata does not contain any columns that satisfy this visualizer's requirements. Theremust be at least one metadata column that contains categorical data, isn't emp…
> [@Plugin error from diversity](https://forum.qiime2.org/t/plugin-error-from-diversity/3335):
>
> I am having problem to generate alpha diversity. Until this step everything is ok, but when I do this command: qiime diversity alpha-group-significance --i-alpha-diversity core-metrics-results/faith\_pd\_vector.qza --m-metadata-file mapping\_fileJ793.txt --o-visualization core-metrics-results/faith-pd-group-significance.qzv I receive this message: Plugin error from diversity: Metadata does not contain any columns that satisfy this visualizer's requirements. There must be at least one meta…
Hope this helps!
---
### Author:  [@bloc\_gen](https://forum.qiime2.org/u/bloc_gen)
#### Post date: [January 2, 2020, 5:44pm UTC](https://forum.qiime2.org/t/metadata-error-message-no-categorical-data/13028/3 "2020-01-02T17:44:36Z")
Hi,
Thank you so much for getting back. That is my metadata in entirety. i tried changing the categorical column in to 2 variable 'gut' and 'feces' and it worked. And yes, I was trying to do an alpha significance.
If possible, can you suggest me a method where I can use 3 variables for the 3 days?
Thanks a million!
---
### Author:  [@fedarko](https://forum.qiime2.org/u/fedarko)
#### Post date: [January 3, 2020, 1:21am UTC](https://forum.qiime2.org/t/metadata-error-message-no-categorical-data/13028/4 "2020-01-03T01:21:17Z")
Understood! Ok, a few things.
> i tried changing the categorical column in to 2 variable ‘gut’ and ‘feces’ and it worked.
If these samples were really taken from the same `body-site` (i.e. they're all fecal samples), I don't think doing this is a good idea.
The reason `qiime diversity alpha-group-significance` will fail if all of your samples have a `body-site` of `feces` is because this visualization involves a comparison between groups of samples. If your only categorical variable is `body-site`, and all of the samples have the same `body-site`, then you only have one group of samples (`feces`) -- you'll need to think "what am I comparing these samples against?" This comment by @jwdebelius goes into detail about this sort of issue:
> [@Comparing beta diversity between samples (not groups)](https://forum.qiime2.org/t/comparing-beta-diversity-between-samples-not-groups/7981/2):
>
> Hi @devonorourke, I want to address your goals. As you've described them, I don't think they're possible. There is no statistical test that I can come up with that will let you compare categorical data when you only have one sample per category. The test - regardless of the type of test - relies on a distribution of data. And, the only way to characterise this distribution is to have multiple samples. (You can look at continuous variables because they're essentially sampling a distribution by…
> If possible, can you suggest me a method where I can use 3 variables for the 3 days?
I'm not sure there are a lot of (or any?) statistical tests that will tolerate three samples total, or 3 groups each with n = 1. There are certainly some ways you can analyze these samples, but I really doubt that three is enough samples to where you could consider assigning statistical significance to anything. Quoting @Nicholas_Bokulich in [the thread I linked above](https://forum.qiime2.org/t/comparing-beta-diversity-between-samples-not-groups/7981/5):
> As @jwdebelius pointed out, there is no method that will allow you to determine whether individual samples are significantly different, unless if you sample them multiple times. You can aggregate your temporal samples (or bin them into groups of sequential weeks) to test for differences between site. But you cannot say “site X and week Y” is significantly different from other sites/weeks if you have N=1 at that site/week. Unless if temporal variation is high, I recommend aggregating (drop the week term)/binning to test for site-specific differences.
As I understand it, your dataset has three samples taken from the same site over time? Since doing statistical analyses on alpha diversity differences isn't really feasible in this case, you could try just visualizing things about them -- for example, you could use q2-longitudinal to [show how the alpha diversity of your samples varies over time](https://docs.qiime2.org/2019.10/tutorials/longitudinal/#volatility-analysis), as one example. The plot you'll get from that will only have three points, but it's at least something you could check out! :qiime2:
---
### Author:  [@system](https://forum.qiime2.org/u/system)
#### Post date: [February 3, 2020, 7:21am UTC](https://forum.qiime2.org/t/metadata-error-message-no-categorical-data/13028/5 "2020-02-03T07:21:20Z")
This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.