Find my observations that have IDs from someone else but are not RG

I recently got a question from someone that wanted to display/view all their observation that are not yet research grade but wich have at least one ID from another user.
The idea would be to query and them review ones observations which are not yet RG but for which others have already left an ID.

To date I have not figured out a way to doing this, nor a way to sort search results by the number of IDs attached to them (this was the best "workaround I managed to find:

Maybe someone has an idea on how to approach this.


I seem to be able to do this by going to “my observations” and choosing the filter options “Needs ID” in the upper left, and then selecting “species” for both high and low ranks (seen under categories). The problem is that it is also the things I have identified to species, not just things that other people have identified to species… so it doesn’t quite fulfill all your parameters. It also wouldn’t capture those that are at genus level or less, if they are interested in those (but you could change the filter for whatever level you wanted). But maybe it’s a start?

1 Like

The API exposes an identifications parameter which accepts values of most_agree, most_disagree, or some_agree. I don’t know what the definitions of those values are, but some_agree seems to be a subset of most_agree, so perhaps it’s agreements > disagreements, disagreements > agreements, and agreements >= disagreements, respectively, and relative to the OP’s original ID. Perhaps you could union these sets together and discard duplicates to get all observations with an ID.

Here are the individual sets for your user:


I’m very interested in the answers you receive (although I’m not optimistic). Similarly, I’d like to be able to search for Research Grade observations with only one ID (not counting the observer). In my experience, observations with only one ID are most likely to be incorrect.

I don’t suppose you’re planning to do this but as I observe this behaviour regularly on iNat, I’ll leave it as a general comment: in case there is ID by someone else it is not advised to “agree” with it if you are not able to ID specimen yourself:

1 Like

most agree is num_identification_agreements > num_identification_disagreements
some agree is num_identification_agreements > 0
most_disagree is num_identification_agreements < num_identification_disagreements

However, the number of agreements and disagreements is set by some rather old code and can be a little inconsistent.

Search by number of IDs was requested in the Explore brainstorm. Though you could probably add a request specifically for number of IDs not including the observer’s.

The short answer is that there isn’t an easy and reliable way to do this currently. I think if you wanted to be thorough, you would download the data from all the user’s needs_id observations and write a script of some sort to check each one for an ID made by someone other than the user.

I think the next best thing, which is a bit unreliable because the ID count isn’t always correct, is to bring up needs_id observations in Identify, and look for the ID count icon. If the user identified their own observations to start, then cards with a shield should be observations that other users have also added IDs to.


I fully agree with what you are saying, it should definitely not be “misused” to simply agree with what other IDed. I would also understand if such a functionality should not be easily accessible, to prevent what you are describing

Personally for me it would be a way to “revisit” my older observations an maybe evaluate the IDs I added back then, or also use the skills I learned over time to add better IDs to my own observations.


Thanks for this suggestion, I did not think of that.

Maybe in the future we will get the “number of IDs” counter.


This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.