|
mnottertail -> RE: Polar Bear Researcher Suspended, Under Investigation (8/1/2011 7:29:21 AM)
|
all of which was cited in the report. he used a common math conjecture, which was also indicated.
x is to y and m is to z.
4 bears floating after 11% survey.
there are 9.09 11s in 100.
If we had surveyed 100% of the area, we would have expected that it would indicate 9.09 X 4 dead bears or 36 dead bears in that area, all things being equal.
Read the fucking thing. Read the article, read the, oh the hell with it, you don't read.
ERIC MAY: Did they comment at all about any of the stats or –
CHARLES MONNETT: Uh, there’s no stats in there.
ERIC MAY: Well, calculations, for, for example, the 25 percent survival rate.
CHARLES MONNETT: Oh, well, that’s just a mindless thing. That’s in the discussion. Um, that is not a statistic. Um, that’s a ratio estimator. It’s a, it’s a fifth grade procedure. Do you have kids?
ERIC MAY: No.
CHARLES MONNETT: Okay, well, if you had kids, you would know that in about fifth grade, they start doing a thing called cross multiplication. "X" is to "Y" as, you know, "N" is to "M." And you can – there’s, there’s a little procedure you use to compare the proportions. And so that’s a, um, simply a calculation. It’s not a statistic.
ERIC MAY: Okay.
CHARLES MONNETT: And, uh, we were very careful, um, in how we presented that, to first make it clear that 6
we had – we didn’t have sufficient sample size, although a, a, a peer statistician type would probably argue we did. But we felt we didn’t have a sufficient sample size to do statistics and, you know, and to estimate, to do any estimators or confidence intervals or anything like that on. And we put caveats throughout that section, saying that, uh, "it’s possible." And we felt that, um, we didn’t want to leave the reader thinking that, "Okay, they went out, and they surveyed it, and there were four dead bears." Because this is a survey, and it only looks – it only covers a small part of the habitat.
When you’re out there flying in an airplane, uh, over this vast area, our transects were 100 kilometers or longer, many cases, and we were surveying an area 500, you know, kilometers wide. We appreciated that we had a very limited, you know, scope in this thing. We were only looking at a small percentage of it, and so we thought that it would be worthwhile, uh, letting them know essentially that we only looked at about 10 percent of the area. And so if you just kind of draw a circle around the area where the dead bears were, then if we looked at 10 percent of the area, um, it’s reasonable to think that if they’re distributed randomly, which we don’t have any reason not to think they are, that we would see 10 percent of what’s there. And that’s a standard thing that’s, um, used all the time and sometimes very rigorously.
ERIC MAY: Okay.
CHARLES MONNETT: But you, but you have to state your assumptions, which, you know, I think we did, so –. And that’s – that hasn’t been controversial. Nobody, nobody’s really complained about that that I’m – that I recall anyway. [IG Interview Transcript at pages 37-8]
Later, it became clear that the IG agents were focused on what in their mind was a disparity but was, in fact, their inability to understand the note. The issue was how many dead bears had been observed floating in open waters:
ERIC MAY: Well, actually, since you’re bringing that up, and, and I’m a little confused of how many dead or drowned polar bears you did observe, because in the manuscript, you indicate three, and in the poster presentation –
CHARLES MONNETT: No. 7
ERIC MAY: – you mentioned four.
CHARLES MONNETT: No, now you’re confusing the, um, the estimator with the, uh, the sightings. There were four drowned bears seen.
ERIC MAY: Okay.
CHARLES MONNETT: Three of which were on transects.
ERIC MAY: Okay.
CHARLES MONNETT: And so for the purpose of that little ratio estimator, we only looked at what we were seeing on transects, because that’s a – you know, we couldn’t be very rigorous, but the least we could do is look at the random transects. And so we based, uh, our extrapolation to only bears on transects, because we’re saying that the transects, the, the swaths we flew, represented I think it was 11 percent of the entire habitat that, you know, that could have had dead polar bears in it.
ERIC MAY: Um-hm [yes].
CHARLES MONNETT: And, um, so by limiting it to the transect bears, then, you know, we could do that ratio estimator and say three is to, um, uh, "x" as, uh, 11 is to 100. I mean, it’s that kind of thing. You, you’ve, you’re nodding like you understand.
LYNN GIBSON: Yeah.
CHARLES MONNETT: Yeah, that’s pretty simple, isn’t confusing. I mean, it’s –
ERIC MAY: So, so, so you observed four dead polar bears during MMS –
CHARLES MONNETT: One of which was not on transect.
ERIC MAY: Okay, so that’s what –
CHARLES MONNETT: Yeah.
ERIC MAY: So is that considered an MMS survey, the one that was not?
CHARLES MONNETT: Yeah, because when we go out there, we don’t – we aren’t just limited to flying transects. We have to get there.
ERIC MAY: Okay.
CHARLES MONNETT: And we connect. We go between transects.
ERIC MAY: Right. [IG Interview Transcript at pages 43-4]
But despite saying that he understood, Agent May did not understand. Later in this interview this exchange illustrated that the investigation was premised on illusory points:
ERIC MAY: Three dead polar bears?
CHARLES MONNETT: Yeah, three dead. 8
ERIC MAY: Right.
CHARLES MONNETT: But the four swimming were a week earlier.
ERIC MAY: Okay.
CHARLES MONNETT: And, um, then we said if they accurately reflect 11 percent of the bears present so, in other words, they’re just distributed randomly, so we looked at 11 percent of the area.
ERIC MAY: In that transect?
CHARLES MONNETT: Yeah.
ERIC MAY: Right.
CHARLES MONNETT: In, in our, in our area there, um –
ERIC MAY: Right.
CHARLES MONNETT: – and, therefore, we should have seen 11 percent of the bears. Then you just invert that, and you come up with, um, nine times as many. So that’s where you get the 27, nine times three.
ERIC MAY: Where does the nine come from?
CHARLES MONNETT: Uh, well 11 percent is one-ninth of 100 percent. Nine times 11 is 99 percent. Is that, is that clear?
ERIC MAY: Well, now, seven of 11 – seven of what number is 11 percent? Shouldn’t that be – that’s 63, correct?
CHARLES MONNETT: What?
ERIC MAY: So you said this is –
CHARLES MONNETT: Seven/11ths this is –
ERIC MAY: No, no, no, no, no. This, this is, this is 11 – seven is what number of 11 percent?
CHARLES MONNETT: Seven?
ERIC MAY: Yeah.
CHARLES MONNETT: Is what number of 11 percent?
ERIC MAY: Eleven percent, right.
CHARLES MONNETT: Well, I don’t know. I don’t even know what you’re talking about. It makes no sense. [IG Interview Transcript at pages 51-2]
Agent May persisted but as the interview wore on it became less and less clear just what credible allegations prompted this investigation, as illustrated by this exchange:
ERIC MAY: "If seven total bears, four swimming, uh, and three drowned represents 11 percent of the population" –
CHARLES MONNETT: It doesn’t.
ERIC MAY: Okay, and we’ll – let me, let – "of bears before the storm, then the total number of bears after 9
the storm is 63," and that’s where I came up with the sixty –
CHARLES MONNETT: That’s just stupid. I – did you do that?
ERIC MAY: No.
CHARLES MONNETT: That is stupid.
ERIC MAY: I’m a, I’m just – I interview –
CHARLES MONNETT: In the first place, there’s – it’s 200 percent, okay?
ERIC MAY: So explain – tell me why that’s wrong.
CHARLES MONNETT: Well, because they’re acting like they were all seen at the same survey. We flew the whole thing twice to see that, right?
ERIC MAY: Right, and that’s, that’s different.
CHARLES MONNETT: Yeah.
ERIC MAY: That’s where the mistake is here –
CHARLES MONNETT: Yeah.
ERIC MAY: – because they –
CHARLES MONNETT: Yeah.
ERIC MAY: – they – it occurred on different trips.
CHARLES MONNETT: Yeah, it, it, it’s, it’s three out – uh, three is to 11 to 100 percent, and then four is to 11 to 100 percent. It’s another 100 percent. And so I, I don’t even still follow what they did to get the 60 percent. That, that’s –
ERIC MAY: The 63 percent.
CHARLES MONNETT: Yeah, that’s just goofy.
ERIC MAY: Okay.
CHARLES MONNETT: But you should at least be – if you were trying to, uh, uh, document the rate at which we saw something, dead or swimming, it would be seven out of 200 percent. [IG Interview Transcript at page 60]
After a few more minutes of mystifying questions from the IG agents, they indicated they were about to conclude. This exchange then occurred with one of the PEER legal team, Executive Director Jeff Ruch:
JEFF RUCH: Um, but, uh, Agent May indicated to, um, us all that he was going to lay out what the allegations are, and we haven’t heard them yet, or perhaps we don’t understand them from this line of questioning.
ERIC MAY: Well, the scientif- – well, scientific misconduct, basically, uh, wrong numbers, uh, miscalculations, uh –
JEFF RUCH: Wrong numbers and calculations? 10
ERIC MAY: Well, what we’ve been discussing for the last hour.
JEFF RUCH: So this is it?
CHARLES MONNETT: Well, that’s not scientific misconduct anyway. If anything, it’s sloppy. I mean, that’s not – I mean, I mean, the level of criticism that they seem to have leveled here, scientific misconduct, uh, suggests that we did something deliberately to deceive or to, to change it. Um, I sure don’t see any indication of that in what you’re asking me about.
ERIC MAY: No, no, no further comment on my part. We, we’re – I’m just about complete with my – the interview, so –
CHARLES MONNETT: Really? Oh, good. That’s it?
ERIC MAY: Like I said, we receive allegations; we investigate.
CHARLES MONNETT: Don’t you wonder why somebody that can’t even do math is making these allegations and going through this stuff? [IG Interview Transcript at pages 83-4]
. . . in the 11 percent of the habitat. And so you could set up a, um, a ratio here, three is to “x” 25 equals 11 over 100, right? And so you end up with – you can cross-multiply. You know algebra?
ERIC MAY: Um-hm [yes], yeah.
CHARLES MONNETT: You can cross-multiply. Okay, so you end up with 300 equals 11x, and I am sure that that‟s – equals 27, okay?
ERIC MAY: Right, right, got that.
CHARLES MONNETT: And if you stick four in here instead, you end up with –
ERIC MAY: Thirty-six.
CHARLES MONNETT: – whatever that number was, yeah, 36. Now, um, those numbers aren‟t related, except we made the further
assumption, which is implicit to the analysis. Seems obvious to me. We went out there one week, and we saw four swimming on the transect, which we estimated could have been as many as 36.
CHARLES MONNETT: – that right after we saw these bears swimming, this storm came in and caught them offshore, all right? And so if, um, if you assume that the, the, the 36 all were exposed to the storm, and then we went back and we saw tentially 27 of them, that gives you your 25 percent survival rate. Now that‟s, um, statistically, um, irrelevant. I mean, it, it‟s not statistical. It‟s just an argument. It‟s for, it‟s for the sake of discussion. See, right here, “Discussion.”
ERIC MAY: Um-hm [yes].
CHARLES MONNETT: That‟s what you do in discussions is you throw things out, um, for people to think about. And so what we said is, look, uh, we saw four. We saw a whole bunch swimming, but if you want to compare them, then let‟s do this little ratio estimator and correct for the percentage of the area surveyed. And just doing that, then there might have been as many as 27 bears out there that were dead. There might have been as many as 36, plus or minus. There could have been 50. I don‟t know. But the way we were posing it was that it‟s serious, because it‟s not just four. It‟s probably a lot more. And then we said that with the further assumption, you know, that the bears were exposed or, you know, the ones we‟re measuring later that are carcasses out there, it looks like a lot of them, you know, didn‟t survive, so – but it‟s, it‟s discussion, guys. I mean, it‟s not in the results. …
|
|
|
|
