Why is this Observation Casual/Needs ID/Research Grade? - "Official" Topic

[tiwane]

A common issue brought up on the forum is an observation which looks like it meets the criteria for a certain data quality grade (i.e. Casual, Needs ID, or Research Grade), but iNaturalist does not label the observation as having that grade. While this is sometimes caused by a software bug, it is usually due to other reasons. This tutorial goes over what to check before filing a bug report.

Here are the criteria of each data quality grade on iNaturalist.

---

To start sleuthing, first scroll down the observation’s page until you reach the Data Quality Assessment (DQA):

One the left you will see the observation’s current data quality grade, and below that the various Research Grade qualifications. Any member of the community can vote on the qualifications with the thumbs up or thumbs down icons.

On the right is an explainer detailing why the observation has its current data quality grade. In the example above, you’ll see that this particular observation is lacking a community agreement on the ID and therefore cannot be Research Grade. More often than not, you will find your answer here - look for the matching icons on the left- and right-hand sides.

Perhaps the most common reason an otherwise worthy observation is not labeled Research Grade is due to this:

Someone has voted that the Community Taxon can still be improved. This is often due to the voter misunderstanding its purpose. In situations like this, “No, it’s as good as it can be” votes can balance out the “Yes” votes and make the observation Research Grade, but it’s often best to also comment on the observation and ask the voter why they voted that way. Clicking on the number next to a vote will show you who voted that way:

---

If the Data Quality Assessment section looks good, then it’s possible the observation was not “reindexed” properly, and the system has not registered its true current data quality state. When a part of an observation changes, like its ID, a DQA vote, someone faving it, iNaturalist reindexes the observation to reflect the change. However, this sometimes doesn’t happen due to a glitch or a bug. You can force an observation to be reindexed by voting in the DQA or faving the observation. Give it a few seconds and the observation’s data quality grade might change to an accurate one.

Please make sure to retract any vote or fave you don’t mean to keep.

---

Another common cause that an observation has not attained Research Grade is that the observer has opted out of Community Taxon. This means that the observation Wil be labeled at the top with the observer’s ID and not the community’s ID. Take a look at the Community Taxon section of the observation. If the user is has opted out of the Community Taxon, you will see that here:

In situations like this, the observation will not become Research Grade unless the observer’s ID matches the Community Taxon.

---

If none of the above situations fit the observation in question, the issue may be a software bug, so please file a Bug Report for the observation.

[tiwane]

The above post is a wiki, so please feel free to add other common reasons why an observation has its current data quality grade!

[thomaseverest]

OK I’ve been trying forever to figure this out today: observations were casual because they had been checked “No, it’s as good as it can be.” Is this a bug? Or an intended feature? They were all observations that couldn’t be IDed further than, say, family, but the check made them Casual instead of Research Grade. The more I try to find a pattern the more mystified I get because it’s not consistent. It seems to be that observations I have identified get marked as casual if I check No, but it’s not entirely consistent either. It doesn’t matter whether the user has opted out of Community ID or not. It doesn’t matter when the observation was observed or added, but it seemed to be worse with older observations.

Here’s an example:
https://www.inaturalist.org/observations/2402042
I’ve checked yes to bring to Needs ID, but if I remove my check it reverts to Casual. I can replicate it by voting No on this observation:
https://www.inaturalist.org/observations/2396067
I can’t replicate it here (same user, newer observation), nothing happens:
https://www.inaturalist.org/observations/44400934
I didn’t ID any of those observations.

It’s also the case here where the user hasn’t opted out of the Community ID:
https://www.inaturalist.org/observations/533751

But voting No here doesn’t change anything:
https://www.inaturalist.org/observations/48842313

This isn’t limited to family level or oysters (even though that’s what all of these were) but could it be a mollusk thing? That’s almost the only thing I identify. I’ve tried with other groups and I’ve never gotten it to give me Casual. Then again any time I try checking No for a banana slug it goes to Research Grade. I’m very confused but maybe I’m missing a very simple explanation. If so, I’d love to add it to the Wiki! If not, I can make a bug report.

[sgene]

The examples you gave are all family-level, so they can’t become Research Grade. In the last paragraph in section 7 in the link @tiwane provided above, it says that genus-level can become Research Grade, but it doesn’t say family-level can.

[thebeachcomber]

@sgene @tiwane that link might actually need updating re genus and below can become RG

I have an observation that’s at subfamily and RG, so it would seem like that’s the upper threshold
https://www.inaturalist.org/observations/45670041

[aisti]

It even has the big red X in the DQA box for “community taxon is genus or lower”. I wonder if this should be posted as a bug report.

[jwidness]

@thebeachcomber

at subfamily and RG

Yes, the actual threshold when the observer hasn’t opted out of community ID is lower than (but not including) family, see here.
So the full list of what can be made RG by checking that box is:
subfamily
supertribe
tribe
subtribe
genus
genushybrid
subgenus
section
subsection
complex

The threshold when the observer has opted out of community ID is defined in a different spot and specifies only genus (i.e. not subfamily, supertribe, etc.). I don’t know why it’s done that way, but it’s already been reported.

Edit: this was fixed and now observations at these ranks should be able to convert to RG regardless of whether the observer opted out of CID.

[thomaseverest]

Ah OK so that explains why those can become Research Grade. But what’s the pattern with checking No at a level below genus or at/above family? In some instances nothing changes and the observation stays at Needs ID. In others it reverts to Casual.

[sgene]

These that you gave as examples https://www.inaturalist.org/observations/2396067 and https://www.inaturalist.org/observations/44400934 have a “yes” in the DQA, so that would cancel your “no” vote.

[thomaseverest]

Hm I suppose I didn’t notice that. But voting No on this one makes it Casual:
https://www.inaturalist.org/observations/49032602

[bouteloua]

The community taxon of that one is only at rank superfamily.

[thomaseverest]

Ohhhh. OK that makes sense. So let me get this straight.

If the Community Taxon (not the observation taxon) is lower than family and the user hasn’t opted out of the Community Taxon, voting “No, it’s as good as can be” will make the observation Research Grade. If the Community Taxon is family level or higher, it will become Casual. The only difference with users who have opted out of the Community Taxon is that voting “No…” will only make the observation Research Grade if the Community Taxon is at genus level. At all other levels voting “No…” will make the observation Casual. Note that nothing happens when “Yes” or “No…” is checked if no other user has provided an identification on an observation by a user who has opted out of the Community Taxon because there is no Community Taxon.

Should that go in the Wiki? I didn’t try checking “Yes” in all those different cases but I assumed they would make it casual unless the user had opted out?

So I guess my ultimate question is what’s the justification for any relevant observation becoming Casual? Or is that a bug?

[bouteloua]

You might be interested in voting for this feature request :)
https://forum.inaturalist.org/t/say-what-the-community-taxon-is-in-the-dqa/8836

@thomaseverest

The only difference with users who have opted out of the Community Taxon is that voting “No…” will only make the observation Research Grade if the Community Taxon is at genus level.

Not quite. If the observer has opted out of community ID, and the observer’s ID is maverick to what the community has decided, the observation will never be research grade.

@thomaseverest

nothing happens when “Yes” or “No…” is checked if no other user has provided an identification on an observation by a user who has opted out of the Community Taxon because there is no Community Taxon.

Right.

And sorry, I’m not following your last few sentences - which part of all that do you think should go in the wiki? Which cases are you referencing when you say “checking ‘yes’…would make it casual”, and what you mean by relevant observation? :)

Some of these explanations are “in” the wiki by way of linking to the FAQ, but maybe it’d be helpful to transfer some of that over here. Kind of a lot of places to keep updated though. https://www.inaturalist.org/pages/help#quality

The data quality info is also on each observation page:

[thomaseverest]

OK thanks. I just wondered if any of that should go at the top of this page. So far this is the first thing I haven’t been able to figure out by myself so I thought it might be helpful to put where others might find it easier. But I suppose it’s rather complicated and obscure enough that it doesn’t need to.

[sgene]

@thomaseverest

I suppose it’s rather complicated and obscure enough that it doesn’t need to

I think it’s so obscure and complicated that it needs to be presented succinctly in each observation, under “The Below Items are Needed to Obtain Research Grade.” Just a list, so a person wondering about a particular observation can just go down the list and easily see which ones haven’t been met. Along with a prominent statement in that section of what the Community ID currently is.

[krokozavr]

Seems like pretty rare case:
https://www.inaturalist.org/observations/3629050

[benkendrick]

Platform** (Website)

Browser, if a website issue (Chrome) :

https://www.inaturalist.org/observations/50964158

Seems to be automatically declaring casual.
I don’t think it was captive/cultivated.

[draposo79]

It’s marked as captive by… iNat?!

[jwidness]

From https://www.inaturalist.org/pages/help#quality:

“The system will vote that the observation is not wild/naturalized if there are at least 10 other observations of a genus or lower in the smallest county-, state-, or country-equivalent place that contains this observation and 80% or more of those observations have been marked as not wild/naturalized.”

In this case there was a group recording captive mussels at the Haley Institute of Animal Science and Aquaculture, so there are already a bunch of captive observations in this county – and not very many wild.