They Added One Dropdown Option. Reports Tripled In Nine Weeks.
Eleven years of submissions, read against the form's own version history. A contractor added one menu item in March 2019 to match a neighbouring state. Free-text reports of the same animal, in the same counties, did not move.
Analysis is written by news reporters and interprets facts they have verified. It is not opinion, and it takes no position on what should be done.
SALEM, Ore. — The commit is four lines long. It was pushed at 15:42 on 14 March 2019 by a contractor closing a ticket, and it added one string to a list.
Nine weeks later, monthly submissions reporting that animal had gone from a mean of fourteen to a mean of forty-seven. They stayed there for thirty-two months. In November 2021 the option was taken out during a site redesign, and the mean fell to sixteen.
Free-text reports describing the same animal, in the same eleven counties, did not move in either direction. Not up in 2019. Not down in 2021.
The ticket
The ticket text is nineteen words: Parity with WA form — add species options we're missing. See attached screenshot. Low priority.
It was raised by a programme assistant at the agency who had been sent a link to the neighbouring state's public reporting page and noticed the two lists were different.
Devon Achterberg-Rios wrote the change. He works for the vendor that maintains the form, has never been to any of the eleven counties, and had to be told twice why we were calling.
"I don't have an opinion about the animal," he said. "I had a ticket, and there were four options on it."
The other three were Rocky Mountain elk (unconfirmed), melanistic bobcat, and feral swine.
What the menu did
The agency released every submission the form has taken since it went live: 41,806 rows, in two tranches, under the state public records act.
Before March 2019 there was no category for this animal. A person who wanted to report one had to choose Other — not listed and then type a description into a box. Fourteen a month took that trouble.
After March 2019 they could choose it from a list, in one click, and say nothing else at all.
Forty-seven a month did, and sixty-one per cent of those left the description box empty, because the menu had already said it for them.
The thing that stayed still
We coded the free-text field. Two coders, working blind to each other, marked every submission from 2015 to 2026 that described a large upright hair-covered animal, whatever category the submitter had picked. They agreed on 94 per cent of cases; the disagreements are published.
That series has a mean of 8.6 reports a month before the change, 8.9 while the option existed, and 9.1 after it was removed.
The animal did not arrive in 2019. It did not leave in 2021. A menu item did, and we can date both events to the minute.
The two papers
A 2022 paper treats the 2019–2021 window as a genuine temporal and spatial cluster. A 2023 paper in a regional wildlife journal reads a northward shift of 68 kilometres out of the same window.
Neither set of authors was told the form had changed. Nobody withheld it. Nobody thought to mention it, because nobody at the agency thought of the form as an instrument — it was a web page, maintained under a support contract, by people who were not asked.
The 68-kilometre shift is driven by two northern counties. In April 2019 the agency's outreach officer put a link to the reporting form in a county newsletter that goes to about 9,000 households in those two counties. That is a second instrument effect, from a second document nobody archived, and we found it by accident.
Dr. Solveig Trevanion-Baptiste, who wrote the 2022 paper, has rerun it. Her reply is printed at the foot of this page, entire, at her insistence and at more than twice the length we offered her. It is the most useful thing on this page and it is not friendly to us.
What a form is
Her central objection is fair and this desk accepts it. There is no purer version of these data sitting behind the form. The form is how the reports exist. Calling its behaviour an artefact implies a clean signal underneath, and there isn't one.
But the practical point stands, and it is small and dull and fixable. Eleven years of a public instrument changed at least fourteen times — we counted the commits — and not one of those versions was archived by the agency that owns it. The only reason this story exists is that the vendor happened to keep its repository public.
The agency's public information officer confirmed, in writing, that no dated copy of any version of the form has ever been kept.
There is a whole literature resting on a web page that nobody was watching, maintained by a man who closed the ticket, went home, and does not remember doing it.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
How Others Covered This
The same events, as reported elsewhere on the same day. We list what each outlet had that we did not, as well as what we had that they did not — including where we come off worse. Why we print this.
-
The Meridian TelegraphSightings Surge Across Three Counties As Something Moves North
Ran the 2019–2021 rise in 2021 as a live wildlife story, with a map.
Had that we did not
The county-level counts, accurately, straight from the agency dashboard.
Left out
That a dashboard counts submissions. Nobody at the paper asked what a submission was, and the agency was not asked either.
-
Signal & LedgerState Wildlife Portals Move To Shared Taxonomy Vendor
Trade coverage of the 2021 redesign contract, not a word of it about what the form asks.
Had that we did not
The contract value, the migration schedule, and the sentence 'the species picker was rationalised'.
Left out
What was rationalised out of it, and that two published papers rest on the difference.
-
They Buriedthis newspaperThey Added One Dropdown Option. Reports Tripled In Nine Weeks.
Read eleven years of submissions against eleven years of the form's source code.
Had that we did not
The commit, the ticket, the developer, and the free-text series that stayed flat throughout.
Left out
Our headline says reports tripled. Submissions reporting this animal tripled; descriptions of the animal itself went from 8.6 a month to 8.9, which is inside the noise. That distinction is the entire finding and the headline loses it. — V. Ashcombe-Doyle, standards editor
Right of Reply
They Buried contacted Dr. Solveig Trevanion-Baptiste, quantitative ecologist on 22 April 2026, in writing, with the full analysis and the raw export attached, eleven days before publication. Replied 3 May 2026. She asked that it be printed entire or not at all. Printed entire and unedited.
Your finding is correct and your framing is wrong, and I am going to spend most of this on the second thing, because the first takes one line.
Yes. The option was added in March 2019 and removed in November 2021, and no, nobody told me. I learned it from your reporter's e-mail. I have rerun the 2022 analysis with the instrument period as a covariate and I will publish the result whichever way it falls. As of this week the spatial term survives and the temporal term does not, which means half of what you think you have knocked down is still standing and the other half is genuinely gone.
Now the framing.
You have written this as though the instrument were contaminated and the science therefore void. Every observational dataset in my field is a record of what people found it easy to report. Nobody working in this area believes a sighting report is a sighting. We believe it is a report, and we spend our careers modelling the difference, and we call the difference reporting effort, and it is not a scandal, it is the job.
What is genuinely new in your piece — and I want to be exact, because it is worth more than the rest of it — is the timestamp. I have argued about reporting effort for fifteen years with nothing to point at. You can point at a commit. You can say the words 'fifteen forty-two, the fourteenth of March, by a man closing a ticket'. Nobody has ever been able to put a minute on a change in reporting effort before. I will be citing that for the rest of my working life.
But then you call the cluster an artefact. An artefact of what? Of a form. The form is how these data exist. There is no cleaner version of them sitting behind it, waiting. You are describing the only instrument there is and calling its behaviour an error.
The correction is not to throw the cluster away. It is to archive the instrument. No agency in this country keeps a dated copy of its own public reporting page. I have written to two of them this month asking for exactly that. Both said they would look into it. Neither has a mechanism, a budget line, or a person whose job it is.
One complaint, and then I am finished. You gave me eleven days and offered me five hundred words on a finding about my own paper. I have taken more than five hundred. You said you would print it entire. I am holding you to it.
Published unedited under our right-of-reply guarantee.
How was this story?
We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.
Readers' Letters 0
Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.