Climate Change - The Scientific Debate (Vol. II)

[wc98]

kerplunk said:

Any evidence that the USCRN is giving different results to the regular stations yet? Quite the opposite I believe. If USCRN doesn't give different results then that kind of undermines the imperative to do it doesn't it.

it has only been running since 2004 so not a length of time anyone would be happy to conclude indicative of climate.my understanding is there is no statistically significant warming in that time period. looking at the data there is no warming trend in it as i am not fussed about the statistically significant part
https://www.ncdc.noaa.gov/temp-and-precip/national...

[kerplunk]

wc98 said:

kerplunk said:

Any evidence that the USCRN is giving different results to the regular stations yet? Quite the opposite I believe. If USCRN doesn't give different results then that kind of undermines the imperative to do it doesn't it.

it has only been running since 2004 so not a length of time anyone would be happy to conclude indicative of climate.my understanding is there is no statistically significant warming in that time period. looking at the data there is no warming trend in it as i am not fussed about the statistically significant part
https://www.ncdc.noaa.gov/temp-and-precip/national...

You've answered a question I didn't ask, but luckily the page you link to does answer the question.

Click again to see USCRN and USHCN data on the same graph:

https://www.ncdc.noaa.gov/temp-and-precip/national...

[LoonyTunes]

kerplunk said:

wc98 said:

kerplunk said:

Any evidence that the USCRN is giving different results to the regular stations yet? Quite the opposite I believe. If USCRN doesn't give different results then that kind of undermines the imperative to do it doesn't it.

it has only been running since 2004 so not a length of time anyone would be happy to conclude indicative of climate.my understanding is there is no statistically significant warming in that time period. looking at the data there is no warming trend in it as i am not fussed about the statistically significant part
https://www.ncdc.noaa.gov/temp-and-precip/national...

You've answered a question I didn't ask, but luckily the page you link to does answer the question.

Click again to see USCRN and USHCN data on the same graph:

https://www.ncdc.noaa.gov/temp-and-precip/national...

Could that be any more identical?

[ludo]

wc98 said:

kerplunk said:

Any evidence that the USCRN is giving different results to the regular stations yet? Quite the opposite I believe. If USCRN doesn't give different results then that kind of undermines the imperative to do it doesn't it.

it has only been running since 2004 so not a length of time anyone would be happy to conclude indicative of climate.

I have a strange feeling of deja-vu

ludo said:

wc98 said:

it's just a chosen methodology (pha method is not smoothing afaiui ?) that is neither right or wrong. what is wrong with having a record of individual pristine stations and monitoring their results ? in fact that appears to be what is being created with uscrn ,so someone thinks it's a good idea.

simple, we don't have a time machine to go back to the start of the instrumental record and replace the WEATHER stations with pristine climate monitoring stations optimally space across the planet. Sometimes you have to make the best of what you have (and discuss the uncertainties).

This is of course risible sophistry. First ask why we don't use a network of pristine stations, and then when someone wants to discuss that say that the record is too short to draw any conclusions, answering his own question. This is climate skepticism in a nutshell. Ignore any data where there is any uncertainty if you don't like what it suggests and only look at datasets that you know a-priori can't be used to argue convincingly against your case.

This is why it is not worth trying to discuss science here.

wc98 said:

my understanding is there is no statistically significant warming in that time period. looking at the data there is no warming trend in it as i am not fussed about the statistically significant part
https://www.ncdc.noaa.gov/temp-and-precip/national...

... and then goes on to imply a conclusion from the record that is too short. Statistical hypothesis tests are not symmetric. A lack of statistically significant evidence for warming does not imply that there is no warming, it isn't even necessarily evidence that it isn't warming. Climate skeptics have been demonstrating their ignorance of statistics on this one for years. For a more detailed explanation, see here (waits for ad-hominem against the source).

[robinessex]

Statistics. Mathematical guessing. So I was told at the start of my statistics stuff at college many years ago!

[anonymous-user]

robinessex said:

Statistics. Mathematical guessing. So I was told at the start of my statistics stuff at college many years ago!

It rather makes any kind of future planning difficult if you’re glibly dismissing models and statistics and their role in predicting future outcomes.

[ludo]

robinessex said:

Statistics. Mathematical guessing. So I was told at the start of my statistics stuff at college many years ago!

Unfortunately statistics is sometimes taught by someone that doesn't really understand it (c.f. Haller and Krauss).

[LoonyTunes]

robinessex said:

Statistics. Mathematical guessing. So I was told at the start of my statistics stuff at college many years ago!

I've never heard that. I agree that the misuse of statistics is something to be guarded against but "guessing"?

You'd have to prove that and simultaneously remove a whole branch of mathematics from the globe.

[ludo]

LoonyTunes said:

robinessex said:

Statistics. Mathematical guessing. So I was told at the start of my statistics stuff at college many years ago!

I've never heard that. I agree that the misuse of statistics is something to be guarded against but "guessing"?

Ironically those arguing a lack of statistically significant warming over a period too short to reasonably expect to see statistically significant warming (see the link I gave for an explanation of why) is interesting/meaningful is a misuse of statistics. But for some reason skeptics don't seem too concerned about "mathematical guessing" in that case ;o)

[robinessex]

ludo said:

LoonyTunes said:

robinessex said:

Statistics. Mathematical guessing. So I was told at the start of my statistics stuff at college many years ago!

I've never heard that. I agree that the misuse of statistics is something to be guarded against but "guessing"?

Ironically those arguing a lack of statistically significant warming over a period too short to reasonably expect to see statistically significant warming (see the link I gave for an explanation of why) is interesting/meaningful is a misuse of statistics. But for some reason skeptics don't seem too concerned about "mathematical guessing" in that case ;o)

A bit of net searching

“Correlation does not imply causation.”

https://www.fastcompany.com/3030529/hilarious-grap...

http://www.tylervigen.com/spurious-correlations

[LoonyTunes]

So are Clinical Trials results all wrong then or to be ignored as it's only statistics? Strange, as my daughter runs these in different countries all over the world and tells me that they are of the utmost importance.

She didn't say this but it summarizes my grasp of what she said to me.

"Statistics play a very important role in any clinical trial from design, conduct, analysis, and reporting in terms of controlling for and minimising biases, confounding factors, and measuring random errors. A grasp of statistical methods is fundamental to understanding randomised trial methods and results.

Statistical methods provide formal accounting for sources of variability in patients’ responses to treatment. The use of statistics in clinical trials allows the clinical researcher to form reasonable and accurate inferences from collected information, and sound decisions in the presence of uncertainty. Statistics are key in preventing errors and biases in medical research."

So the next time you are offered a pharmaceutical solution to what ails you I'd suggest you either rethink your attitude to statistics or turn the remedy down.

[ludo]

robinessex said:

ludo said:

LoonyTunes said:

robinessex said:

Statistics. Mathematical guessing. So I was told at the start of my statistics stuff at college many years ago!

I've never heard that. I agree that the misuse of statistics is something to be guarded against but "guessing"?

Ironically those arguing a lack of statistically significant warming over a period too short to reasonably expect to see statistically significant warming (see the link I gave for an explanation of why) is interesting/meaningful is a misuse of statistics. But for some reason skeptics don't seem too concerned about "mathematical guessing" in that case ;o)

A bit of net searching

“Correlation does not imply causation.”

https://www.fastcompany.com/3030529/hilarious-grap...

http://www.tylervigen.com/spurious-correlations

We weren't talking about correlation and causation, so that looks very much to me like rhetorical evasion to avoid addressing the point I raised. Plus ca change...

[anonymous-user]

Robinessex has been stating that “it’s all guesswork” for years on the climate politics thread, it’s the central pillar of his scepticism.

First time he’s ventured in here with it though, looks like it’s going well.

[GroundEffect]

robinessex said:

Statistics. Mathematical guessing. So I was told at the start of my statistics stuff at college many years ago!

Yet all of science relies on it. And it has this really monotonous ability to work as well. A colossal bore.

[grumbledoak]

GroundEffect said:

Yet all of science relies on it. And it has this really monotonous ability to work as well. A colossal bore.

All of science? Hardly. And I would happily bet that modern reliance on it is well correlated with the fall in reproducibility.

"If your experiment needs statistics, you ought to do a better experiment."

[LoonyTunes]

grumbledoak said:

GroundEffect said:

Yet all of science relies on it. And it has this really monotonous ability to work as well. A colossal bore.

All of science? Hardly. And I would happily bet that modern reliance on it is well correlated with the fall in reproducibility.

"If your experiment needs statistics, you ought to do a better experiment."

You certainly can't do clinical trials without statistics. There are a thousand other crucial uses for statistics in everyday life including Weather Forecasting, Emergency Preparedness, Disease Prediction, Genetics and Quality Testing to name just 5.

However, in the context of data science this might help
https://link.springer.com/article/10.1007/s41060-0...

In short anyone dismissing statistical mathematics is a fool.

[ludo]

grumbledoak said:

GroundEffect said:

Yet all of science relies on it. And it has this really monotonous ability to work as well. A colossal bore.

All of science? Hardly. And I would happily bet that modern reliance on it is well correlated with the fall in reproducibility.

"If your experiment needs statistics, you ought to do a better experiment."

It is hilarious that you use that quote from Rutherford, given that there is no science that relies on statistics more than modern particle physics (ask CERN). LOL.

[ludo]

More importantly, Guinness wouldn't be what it is today without statistics.

[hairykrishna]

The concept that any science which relies on statistics can be dismissed out of hand is perhaps the best illustration yet of the level of scientific understanding on the sceptic side. The idea is nonsensical.

[robinessex]

LoonyTunes said:

grumbledoak said:

GroundEffect said:

Yet all of science relies on it. And it has this really monotonous ability to work as well. A colossal bore.

All of science? Hardly. And I would happily bet that modern reliance on it is well correlated with the fall in reproducibility.

"If your experiment needs statistics, you ought to do a better experiment."

You certainly can't do clinical trials without statistics. There are a thousand other crucial uses for statistics in everyday life including Weather Forecasting, Emergency Preparedness, Disease Prediction, Genetics and Quality Testing to name just 5.

However, in the context of data science this might help
https://link.springer.com/article/10.1007/s41060-0...

In short anyone dismissing statistical mathematics is a fool.

Use with caution

In statistics, many statistical tests calculate correlations between variables and when two variables are found to be correlated, it is tempting to assume that this shows that one variable causes the other That "correlation proves causation" is considered a questionable cause logical fallacy when two events occurring together are taken to have established a cause-and-effect relationship. This fallacy is also known as cum hoc ergo propter hoc, Latin for "with this, therefore because of this", and "false cause". A similar fallacy, that an event that followed another was necessarily a consequence of the first event, is the post hoc ergo propter hoc (Latin for "after this, therefore because of this.") fallacy.