Utah Birds, Utah Birding, and Utah Birders. Promoting the sharing of information, and the conservation of habitat for birds in Utah and elsewhere. We are a group of people who want to share what we know, and create a positive birding experience in Utah.

BIRDERS BLOG

a blog by and for Utah Birders

eBird v.s. Record Committees: A New Controversy

posted by Tim Avery at
on Friday, April 29, 2011 

This topic previously got a lot of responses and generated some good conversation. I am bringing it back up due to a rumor I recently heard where in eBird supposedly is telling regional reviewers that records in eBird for rare and out of range birds should not be accepted in eBird unless they are accepted by a state records committee!

SHAME ON YOU EBIRD.

Let me start by saying that I don't know how true the rumor is, and that my sources will remain unnamed. What I can say is I find this to be ridiculous! If you have read my previous posts you are aware of my perception of the records committee. It has been at least 3 or 4 years since I submitted a record, and I have no plans of doing it going forward. For those in support of the committees, more power to you, I think it's great that you use them as an archive of sightings. What I don't like the idea of is eBird basically not including valid sightings because the person submitting the list doesn't see the need, doesn't know about, or doesn't care to submit a record to a state bird records committee. This seems to defeat the purpose of having the eBird reviewer even do their job. At that point why not appoint the records keeping to the state committee?

Bollocks I say! eBird this is for you from me. If you do this, there is a good chance I will stop using the program to submit my records. I will go back to using a spreadsheet and Microsoft Excel. It is absurd to think that valid data wouldn't be included based off this ridiculous idea.

Why isn't eBird being proactive in coming up with their own committee? When I create a checklist, I am giving you the ID of the birds I observed, a time, a place, and notes if I see fit. I can even include a link to a photo. If eBird is worried about how valid some records are, and having something relate to an archive, why not, create a form for more information when a rare bird is flagged in the system?

I see where eBird is trying to go with this, in that a data point submitted by me has no other information regarding the sighting. But is it better to completely dismiss that even when correct that to include it and at least have the data that is provided? It seems to me that some information is better than none--and none is surely doing a disservice to the countless 1,000's who submit data, use the database to derive information, and to future generations who are trying to learn more about birds from what is the biggest database of information regarding bird sightings int he world.

eBird, I surely hope this isn't true. If it is, I am shaking my head in disappointment. I outlined a basic solution to the problem. It will require adding some features--but those features would not only improve the quality of the data, but give the database more of an archive feel with the records that are noteworthy. So eBird, what's going on? Would anyone care to comment on the situation?

Labels: , , ,

6 Comments:
Blogger Greg Gillson said...

Trolling for controversy, Tim?

The instruction for eBird Reviewers is simply to yield to the records committee decision unless we have a really good reason to disagree on the identification.

eBird Reviewers are LOCAL EXPERTS. If a rare bird is reported, Reviewers are, by definition, no longer experts.

As an eBird Reviewer I make decisions based on the evidence--just like a records committee would do. The rarer the bird, the better quality the evidence should be. But I am only one person, and more likely to make a mistake than a records committee. -- I am more likely to accept a poorer quality description than would a Records Committee.

eBird records are reviewed soon (often just hours) after the birds are reported on eBird, and rare birds take months to a couple of years to go through the records committee (longer if the decision is not unanimous). So, no. eBird Reviewers don't necessarily wait for the records committee. A good description or a photo will do for most birds.

If eBird accepts the record initially, but the records committee later determines the description doesn't meet minimum documentation requirements, then it is a simple mouse click to unaccept the bird from the eBird public data view. The record remains in eBird with Reviewers comments and can be viewed by researchers, etc.

Frankly, Tim, your attitude toward records committees is puzzling. A person who says they saw incredible things, but refuses to (or can't) provide any proof is not very believable. You know what happened to the reputation of the boy who cried wolf, right?

One way to look at it is that Records Committees protect birders' reputations.

April 30, 2011 at 1:46 AM  
Blogger Tim Avery said...

@Greg: Thanks for your response. Trolling? This is a commentary on a blog, so I don't see how that is trolling. This blog was created to talk about all things birding related-some controversial. I would say in this case the controversy was created by eBird, when they didn't publicly announce this information. As a user I would like to know how my data is being processed if the process seems questionable.

"The instruction for eBird Reviewers is simply to yield…"

I can think of good reasons to disagree with the records committee. But this post wasn't about yielding to their decision; it was about the fact that if a record isn't submitted to the committee then the eBird record should be invalid.

"eBird Reviewers are LOCAL EXPERTS.”

I don't agree with that comment or believe it is true. A local expert can be an expert in a broader sense. I have seen the same things on records committees. If a diagnostic photo is taken there is NO doubt to the ID of the bird, there is no need to have another body govern the decision. I know you commented on that later, but I wanted to reiterate that point, as it was one of my main concerns.

"If eBird accepts the record initially…”

So only researchers can see the data? I consider myself a citizen scientist, & I would love to see that data. I cant without having to show Cornell I am doing research? I don't even know how that process works. Some of that data is important to birders trying to understand distribution, vagrancy, & patterns. If it is removed from their view, even when valid then a huge disservice has been done to them. The simple solution for eBird is to leave these sightings visible to the public, but have them flagged in the database & the public view as "not confirmed". Allowing the observer to attach or link to photos & post comments that are visible to the public would also be valuable. At that point the public could make the decision on the data on their own.

"Frankly, Tim, your attitude…"

My "attitude"? It's not an attitude, its an opinion, which is perfectly acceptable to have (& part of why this blog was created). A person who says they saw incredible things, photographs, & has documentation of them, CLEARLY saw incredible things. That person doesn't need a committee to decide whether or not their sighting is "credible" or "believable". I would say about 90% of birders are then not "believable" as I would venture the vast majority don't submit records. I obviously don't have a study or stats to prove that number, but based off my personal encounters & speaking with others, I can do some basic math & come up with that number. This has nothing to do with "the boy who cried wolf", it has to do with FACTS. A committee is no more "believable" than any Joe Schmo when their decisions are questionable.

"One way to look at it is that Records Committees protect birders' reputations."

So in order to have a good reputation as a birder, I need a records committee to "protect my reputation"? Frankly, that's bullshit. Sorry to be so blunt, but a records committee has ZERO justification over my reputation as a birder. In fact that is NOT their duty at all. If anyone feels the need for the records committee to validate their reputation, that's their choice-I don't rely on the committee to boost me up & make me feel validated.

The point of this post was to bring up an apparent recent policy change within eBird that was not brought up to the public. This decision seems like something eBird should have announced so that it's users were aware of it. It also has provided several solutions & ideas that could be used as an alternative.

Cheers
Tim

April 30, 2011 at 9:10 AM  
Blogger Tim Avery said...

Just one quick follow up. This discussion is a perfect example of why this blog was created, to talk about and discuss these types of things in a forum that is open to the public, and allows for such discussions. We welcome comments and thoughts from EVERYONE!

Good Birding
Tim

April 30, 2011 at 9:21 AM  
Blogger Brian Sullivan said...

Tim et al.

I just spent .5 hr responding to this and it vaporized when I clicked 'post comment'. Oh well. The short version is this: eBird adheres to state BRCs on decisions for any record that has been submitted to them. In cases where users do not submit to a committee, our reviewers render a decision on the record based on the documentation available. We do not force our users to submit to BRCs, but we do encourage it. We do not wait for BRCs to render a decision before validating obviously well-documented rarities in eBird, but we may choose to wait if a record is controversial or difficult to assess. If the record is not submitted to a BRC, we have to make a decision based on the evidence, and do the best we can without their support.

We value the work that BRCs do for data vetting, and look to them for guidance on difficult records. eBird is built in the spirit of collaboration, and we intend to continue that moving forward.

All eBird data (valid and invalid) are available for download through the Avian Knowledge Network (www.avianknowledge.net). Here you can access all the raw data and use it for whatever research purposes you have.

Thanks

Brian Sullivan
eBird Project Leader
Western Region

April 30, 2011 at 10:47 AM  
Blogger Tim Avery said...

@Brian, Thanks for your response. I understand that this has long been part of what the eBird reviewers do. What I hope is not true is that reviewers have been told NOT to accept records if a records has not been submitted, or a record has been rejected, by a records committee. I would love to see eBird add a way for users to add detailed reports along with rare sightings. It would surely be a valuable tool, and make the data even better!

Cheers
Tim

May 3, 2011 at 9:24 PM  
Blogger Greg Gillson said...

Tim,

An observer may add whatever comment they wish to each sighting. So, for instance, if they know they are reporting something unusual, they can add a description or notes about nesting, sex, age.

In fact, in an upcoming new eBird sightings entry form, any bird flagged as unusual in the filter will automatically request the observer to provide more details, if possible.

Everyone from professional scientists to classes of first grade school children report bird sightings to eBird. This is wonderful! But with such free access comes some questions about data accuracy.

Inaccuracy is not just about mistaken ID. There are also typos (wrong species or wrong numbers [110, when they meant 10]), list building (observations from multiple areas or life lists rather than a single location and date), and just plain maliciousness (April Fool's lists, practice lists, etc.).

Agreement with records committee decisions is just one very small part of the overall data-checking done in eBird to assure as much accuracy as possible.

May 5, 2011 at 5:34 PM  

Post a Comment

Subscribe to Post Comments [Atom]

<< Back to Previous




UTAH BIRDERS FLICKR POOL


    SEE MORE AND SHARE ON FLICKR