The Lancet pulls a fast one 10

[dik]

Not at all hokie...

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[dik]

My time is limited... I may not be going down with it.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[TugboatEng]

My time is limited... I may not be going down with it.

That statement applies to all of us.

 

[BridgeSmith]

SwinnyGG, I understand what the report says, but that really doesn't matter. Whether including the graph in that form was sloppy or intentional, it's bound to be misused.

 

[SwinnyGG]

Your snarky attitude is destroying your credibility. BridgeSmith is just keeping this discussion real in the face of dik's tedious harping.

I absolutely could not care less about your opinion of my credibility on this issue based on an argument with people who find whatever narrative they want regardless of fact.

I understand what the report says, but that really doesn't matter.

At least you're being transparent that you don't care about what the facts are.

This logic would mean that every scientific paper on every topic should never be published unless the conclusions are clear to every single person who might ever read it. Ridiculous.

If you publish a set of drawings and a contractor grossly misinterprets them to their own benefit, is that your problem? No.

 

[BridgeSmith]

At least you're being transparent that you don't care about what the facts are.

I hope that was just given as an example of 'gross misinterpretation'. Otherwise, your reading comprehension needs some work.

 

[BridgeSmith]

If you publish a set of drawings and a contractor grossly misinterprets them to their own benefit, is that your problem? No.

Yes, it is. We still end up figuring out how fix what they screwed up. If it's not clear that they were supposed to do it a different way than they did, we end up paying them to fix it, too.

 

[dik]

Depends on the quality of the drawings and the contractor. I was on a project 30 years back where the contractor was a total incompetent... it happens... There was a Far Side cartoon at the time that described him perfectly... about a contractor building a moat on the wrong side of the castle wall.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[rb1957]

ok, how about this. If that chart wasn't included in the report, but someone added it to a review, would that be accurate (or inaccurate) ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[dik]

It depends on the chart and I don't know what sort of addendum/revision is required to include added data to a report.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[rb1957]

I didn't say "addendum", I said review. If someone reviewed this report and presented this chart in isolation, would that be accurate ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[dik]

Misunderstood you... then simply just add it as a footnote reference...

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[rb1957]

sorry, don't follow you, dik. I said if a review included this chart in isolation would it be accurate ?

Or if this chart wasn't in the report, and the review created it to represent the report ... would that be accurate ?

But, really, it doesn't matter ...

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[LionelHutz]

Since everyone crying about one chart

Oh get off your high horse. You have ABSOLUTELY NO idea what the thought process was behind making the graph. They CHOSE to purposely represent that data in that fashion. We can chose to say it appears to be an attempt to misrepresent the data if we CHOSE.

 

[SwinnyGG]

You have ABSOLUTELY NO idea what the thought process was behind making the graph. They CHOSE to purposely represent that data in that fashion. We can chose to say it appears to be an attempt to misrepresent the data if we CHOSE.

Neither do you or anyone else crying about said chart.

Absolute best case is that we're on equal footing.

If you read the entire report - which I'm willing to bet you didn't - you'd have the context needed to actually interpret the chart. You didn't, so you don't. Which means you're CHOSING to see what you want to see. That's intellectual dishonesty, plain and simple.

 

[BridgeSmith]

If you read the entire report - which I'm willing to bet you didn't - you'd have the context needed to actually interpret the chart.

Again, it's not about those here, or elsewhere, who will take the time to read the report. It's about those who won't. In a well written report, that isn't an attempt to mislead the reader, the charts should not need context. They should accurately and clearly convey the results, without needing further information to "interpret" them. As I wrote earlier, this was either very sloppy work, or an intentional attempt to mislead.

 

[SwinnyGG]

It's about those who won't.

I understand that that is your opinion.

I disagree.

Context always matters. Going back to the drawing analogy - if I try to build a job with just your detail pages, I'm going to get a bunch of stuff wrong. I need the full set of drawings, which clarify what detail applies to what, where there are exceptions, etc. That context is required. I can't even remember how many times an engineer has responded to a question about conflicting details/layouts/whatever with something to the effect of 'the drawing set is to be taken as a whole' in an RFI response or email, because the complete set of information is required to correctly evaluate design intent.

I also still argue that the reader failing to correctly read the chart is not an indicator - even a potential indicator - of malicious intent on the part of the author. This whole thread was started because one dude failed at comprehension, and then a bunch of people piled on without doing a single second of exploring context or even questioning the conclusion because a bad assumption matched their narrative. I don't see how that is in any way defensible.

 

[wheaney]

BridgeSmith has a good point here - isn't the whole point of a chart, plot, or other graphic to communicate data? Take the politics and interpretation away, and what you're left with in this case is a chart which reads differently to the casual observer and the invested one. That ambiguity in communication is never a good thing, especially when people are all too quick to throw vitriol after it.

Granted, I'm still in school, but I am well into the thousands of pages of reports and thousands of slides of slide decks between internships, classes, and design projects with clients. Among others, some stand-out rules thus far:

1: Elements present in charts, plots, and slides should all serve a purpose. Don't call the audience's attention to something that doesn't serve an intentional purpose.
2: When communicating information, make every reasonable effort to remove ambiguity. If it isn't possible to do so, make it abundantly clear to the audience by way of footnotes or other explanation.
3: Visual elements should stand alone, meaning that between legends, labels, and captions, all figures should be just as meaningful and clear alone as in the original context.

While The Lancet clearly abided by rule 1 and potentially by rule 3, rule 2 seems to have been forgotten. A discontinuity in one axis is fairly common and, in this case, labeled. The fact remains that this axis break is still not abundantly clear to the audience. Several options have been proposed in this thread, but simpler than all of them would be to add a simple footnote to clarify the break. Regardless of intent, regardless of politics, regardless of leaning, omitting something as simple as that speaks to a lack of academic and professional integrity.

 

[SwinnyGG]

what you're left with in this case is a chart which reads differently to the casual observer and the invested one.

'Casual observers' are not the intended audience of papers published in scientific journals.

If it isn't possible to do so, make it abundantly clear to the audience by way of footnotes or other explanation.

If you read the report the caption of the chart IS abundantly clear about what the death rates are and how they compare.

Directly from the description of the chart in the report that no one has apparently taken the time to read:

For both cold and heat, the effect was noticeably larger for the oldest age group, with 82 (72 to 91) and seven (six to eight) excess deaths per 100 000 person-years

there was around one death per 100 000 person-years in the youngest age group for cold, and less than one per 100 000 person-years for heat

The impact of cold is important everywhere........with a maximum of 240 (151 to 327) raw excess deaths per 100 000 person-years due to cold in Latvia

There is wider heterogeneity in the effect of heat........with a maximum of 37 (25 to 49) excess deaths per 100 000 person-years in Croatia

 

[rb1957]

that's what you take from the report ? that old people have a higher death rate than young people ? ok, I do hope not much money was spent determining that.

Mind you the variable between nations is interesting ... some have more extreme extremes, some have better (or worse) health care. And since cold is significantly more significant than heat (something that "surely" deserves a mention) the price of fuel and the wealth of nations is also relevant.

but surely a key finding is that cold is significantly more deadly than heat. In part because this brings the economic factors to more light.

It would be interesting to see this analysis on previous decades.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.