Except for the fact that authors who do not engage with the data by helping to improve it with ID’s, instead waiting for the identifiers to bring the data to a fit for publishing level, and only after others have put in a herculean effort to improve the set, at that point these authors walk in and scoop up the data for their own use. Especially annoying since they tend to mis-read the data and end up spreading the belief that this data is not reliable which reflects on the identifiers since we are the ones who framed the data.
A constant lament on this forum is that we need more reliable identifiers. But we’re not offering these experts anything for their work on here. It would possibly help get more experts on here if we insisted that papers wishing to use this data must credit the major reviewers for the set they are using.
This is something that effects the group working eastern North American Bombus. Right now, we have about half a dozen very reliable identifiers that will have that entire dataset finished between us soon. After putting in hundreds of thousands of ID’s and hundreds of hours of work, I would be rather insulted if that was recognized in a paper as casually as “we thank…those who contribute to the taxonomic and distributional knowledge of bumble bees on iNaturalist…”. That is how my identification work for 100% of almost 30K observations used in a recent Bombus paper was acknowledged.
Saying " well the data is free, and nobody said they’d give you anything to do this" is a good way to allow sloppy science to muddy the water. Remember, if these authors approached us in the early stages to vet our abilities, ask questions to help understand the data, and very importantly: request that we complete the dataset they wish to use. That only increases the credibility of the dataset, and demonstrates the care taken in ensuring it’s quality. An author who has never heard of me, but decides that my review, often of his entire dataset, should be taken as gospel without ever once contacting me should make everyone reading the paper ask themselves: “if these guys were this sloppy with data, were they as sloppy with conclusions?”
Do we have any way of generating a list of journals that do or do not typically have a policy of asking for a reviewer with inat experience on inat involved papers? If we could find out journals that do not do this we could write them letters.
I don’t know in the various fields that might use inat observations how many or what type of reviewers are typical; most of my own papers have had 1 reviewer, or sometimes 1 reviewer and additional mark ups from designated specialist editors who don’t write a full review, just some mark ups or comments. So maybe a model where a designated ‘citizen science editor’ marks up all papers that use citizen science data could be good if a journal usually only uses one reviewer and that reviewer needs to be a taxon specialist instead.
Did you write a Letter to the Editor of the journal in question?
I don’t know of any journal that has a policy like this (and I would doubt that there are any). As an editor, I can say that finding reviewers is a pretty ad hoc process. You’re trying real hard to get two, and sometimes you just have to take what you can get. For what it’s worth, I’ve never had a paper with <2 reviewers and the journal I edit for will not allow papers to be published with <2 reviews. For the few people that have in depth knowledge of how iNat works, I doubt that they would want to be reviewing a ton!
I do think that this is something that may improve with time - as iNat data becomes more common in usage (which it certainly is), there will be more reviewers with experience who can provide better feedback. That said, I think it would be great to write a paper about best practices for iNat data which could address some of the issues I have seen (and other people have talked about here and in other threads).
That is interesting! I wonder how the expectations have evolved so differently; these are some of the highest reputation journals in the field who’s stated policy is “typically 1 or 2 reviewers”. I guess one thing is that the big papers all pass internal reviews that are way more stringent than anything an anonymous reviewer could throw at them before they are even posted to a preprint server. And most people think that the real peer review happens when you post an article to a pre-print server; some journals will not even accept a paper to be sent out for review unless it has been posted to a pre-print server.
What I have encountered before is that, for example, in addition to the reviewer the journal included feedback from a separate statistics editor and data editor, where the data editor commented on our use of software and the presentation our figures and tables, and the statistics editor commented on our statistical methodology, but neither provided a full review or a formal recommendation for or against acceptance. So I was thinking a journal could have something like that but with a citizen science editor. I suppose it would be similar to how the journals already separate the journal editor, copy editors, and production editors.
I have contacted before, but this practice is in common use. If you read through the vast majority of papers that use iNat, it is quite common to see no iNat user as an author or any user thanked as an adviser. Most will just thank: “those who contribute to the taxonomic and distributional knowledge of ____ on iNaturalist”. Sometimes they will mention a few names in the acknowledgements, but there is no indication that the people mentioned actually reviewed the entire dataset, and the authors did not contact these people to vet their abilities. Like I said, our ID’s are being taken as gospel by people who have no idea who we are.
Have you tried roughly quantifying this? From my experience (published more than 10 iNat-related papers, reviewed ~10 iNat papers, and of course read + cited many more), I’ve seen the polar opposite, papers where none of the authors are iNat users seem to be proportionally rare from the papers that I read
Effectively like - I found it on FB and now it is mine. Or blog / photo scrapers.
I wonder. As younger scientists (and journal editors) bring experience across using social media, how that will change (perhaps @beachcomber 's comment answers that for me). But with peer review and networking, scientists will sort each other into reliable, or flaky?
Citizen science projects are uncharted territory for old dogs who are not willing to learn new tricks.
A scientist wouldn’t use a graph and its data from an earlier published paper as - I found it, used it, now it’s mine. Oh yes, thanks, by the way.
Every paper I’ve read using bee data with two exceptions is as I described. There is another coming out soon that isn’t in that category.
My favorite was where the authors described Research Grade as “ID’d by two experts”. So glad to see that the authors took the time to research the data before using it.
it must oddly be a largely bee-specific thing then
It might be. I don’t read articles that aren’t bee related. But it is something that I see quite a bit, I had an author who had an account, but wasn’t a regular user send me a few others their paper for review (thankfully) which was filled with data misunderstanding. If they had worked with one of us in the early stages of the paper, it would have saved having to re-write large sections of it. If they had published without approaching us, it would have been terrible.
The fact that they approached us made me hope that will happen more often.
Maybe part of the problem of not properly acknowledging or consulting iNat contributors is that many are anonymous behind usernames. Yes, the authors could still reach out to them by direct messaging, but if the contributor of a record can’t be readily identified as an individual with a real name, it’s perhaps easier to not consider them as such.
An online 'nym is not anonymous.
People have good reasons to use 'nyms.
Some scientists, who use their professional name online, have a disclaimer. ‘This is my personal opinion’. Others may need an absolute separation between work and iNat. A 'nym is not an excuse to dump ethics and courtesy.
Question: If I wanted to cite this discussion (or rather, certain replies in it) in a paper, what is the proper way to do that?
No, it’s not an excuse and, yes, there are reasons for using a pseudonym. But many usernames are anonymous … you don’t know who the person is, what their background is, where they live. I think it’s human nature to view something on the internet that is posted under the person’s real name as more authentic (if that’s the right word) than if it’s under a pseudonym because that person’s reputation is attached to their name and what they post. If I want to publish something, I use my real name. If I’m acknowledged in a paper, the author uses my real name (unless I’m a peer reviewer selected by the editor to critique a manuscript, in which case I’m thanked as an anonymous reviewer).
As far as I know, no scientific journal knowingly allows a person to publish under a pseudonym/username. But maybe that’s coming. I suspect pseudonyms can currently be included in acknowledgments but have not seen that myself.
Scientists do often use data from previous papers. In my field (ecology), many journals now require that the data underlying the paper be published in a repository (like Dryad, Figshare, etc.), and the proportion of journals that require this is increasing. That data is then free to use by other authors (though generally with acknowledgement/citation). The data and analyses can also be checked/assessed by peers. I think this is a positive change. One instance where it is very useful is when authors do meta-analyses to examine the variation in support for a result over many conditions (species, experiments, whatever unit of replication).
You do see authors using or adapting figures from previous works, though that is comparatively less common. Generally you need to ask for permission from the journal/publisher of the original article containing the figure, but it is usually granted (many have a form somewhere on their website where you can submit a request). I’ve done this a couple of times (once for one of my own figure from an earlier paper), and had no issue. I think journals often like it, as it is guaranteed citations, they just don’t want you to take so much content from a previous article that people stop citing/accessing the original. If a graph is published in a CC licensed paper, you generally can use it with acknowledgement and not request permission, just like iNat observations/photos with CC licenses.
Depending on how you consider this, citations/acknowledgements of iNat users are often whatever their username is. There’s no guarantee that they have their real name anywhere on the site, so that can’t be required. I’ve generally cited the real names of iNat user’s I’ve worked with, but this is because I communicated individually and asked them what they preferred to have listed. I think all of the folks I’ve worked with individually have ended up using real names, but I’m not 100% sure.
I’ve occasionally seen people acknowledged by pseudonum. Even an example in a biology book that came out this year where a person is acknowledged by pseudonym for use of a particularly striking phrase he produced in a comment on a blog. Perhaps this will become more common.
I believe it would depend on the style guide for the particular field or publication. The one I’ve used in the past is the American Anthropological Association’s Style Guide – https://americananthro.org/publications/publishing-style-guide/ – which includes a section for citing ‘online resources’ such as websites.
Yeah when I was listed as a coauthor based on an inat obs I made I obviously had to use my real full name and institutional affiliation, but in just an acknowledgement/citation I’d be completely fine being acknowledged by just my username; it isn’t my field anyway.