William M. Briggs

Statistician to the Stars!

Page 151 of 708

Der Spiegel Asks About Global-Warming-Causes-Syrian-Civil-War Paper

Part of the paper's Fig 1.

Part of the paper’s Fig 1.

That peer-reviewed Colin Kelley et al. paper “Climate change in the Fertile Crescent and implications of the recent Syrian drought“, which the traddy media is portraying like Wired did—e.g. How Global Warming Helped Cause the Syrian War—is making the rounds. The Independent (surely not taking sides) said “Climate change key in Syrian conflict — and it will trigger more war in future“.

Axel Bojanowski, Science Editor at Der Spiegel, asked some questions about the paper. Here they are, with my responses (just as I gave them to him).



Hope these comments made it to you in time.

I find the paper poor, full of contradictory assumptions, misunderstandings and misuses of statistics, and with a conclusion which is not justified.

1) Is the time scale (115 years) really sufficient when it comes to “unprecedented changes”?


Especially considering the data is not the data. What I mean is that the data they present (like in Fig 1) are the result of a model, extrapolations from a computer using input from changing stations inside Syria. See the second paragraph of the supplementary material for words about this.

That means they are not justified in drawing single blue lines, as in Fig 1A. These lines should be accompanied by some measure of uncertainty, some kind of ± bounds. That Fig 1A blue line is just a guess, in other words, which has uncertainty attached to it.

The most important thing to get from this is that the uncertainty in the data *must* be carried forward in *all* analyses. That means that red line they drew (from a regression model) on top of Fig 1A is itself too certain. I mean, that “statistical significance” they claimed with a wee p-value might not be, and very probably isn’t, true.

Incidentally, did you notice they described “statistical significance” by p-values less than 0.05 and later by less than 0.1? Grasping at straws here. Anyway, p-values are awful measures of model performance. See this: http://wmbriggs.com/post/15465/

2) Do you expect the climate data of Syria beeing sufficient especially in the first half of the 20th century?

No, and for the same reasons. We barely have any real observational data. What we have model outputs as guesses of data.

3) Does the argument that models expected the detected change really satisfy when models often fail when it comes to regional climate change?

No, no satisfaction. We don’t even know if their models are any good. The only way we can is if they can predict new data, data never before seen. They give no indication they have paid any attention to actual model performance. They instead ask us to take their word for it that the models they use are fine and that the causes they ascribe are real causes. Why should I trust them?

They ask us to believe that their inefficient regressions (bottom of p. 3) “suggest anthropogenic influence” on the drought. It does nothing of the kind. Even giving them the “significance” (which we can’t because the data they use is not the data), it is not evidence that humans caused the drought. Statistical models do not prove causality, though to believe they do is a common mistake.

The burden is on them to say exactly how this drought could have been *caused* by humans by showing us the causal mechanisms. Their burden is a heavy one because observations show that world-wide temperatures have not done much these past two decades. I mean, the warming we were promised by climate models never materialized. That means—it absolutely means—those models are broken and cannot be trusted. Would you trust a weatherman who forecasted temperatures always too high?

4) Do you consider the climate connections named in the paper beeing plausible?


On p. 2 they say “1.2-1.5 million Iraqi refugees…arrived…at the beginning of the drought”, and that these refugees make up “roughly 20%” of Syria’s urban population. They also say Syria’s population rose. And some of this is from the Iraqi influx.

They also say on that same page “The rapidly growing urban peripheries of Syria, marked by illegal settlements, overcrowding, poor infrastructure, unemployment, and crime, were neglected by the Assad government and became the heart of the developing unrest”.

Can a drought exacerbate internal unrest? Of course. Who would doubt that? The authors even say (p. 5) “An abundance of history books on the subject tell us that civil unrest can never be said to have a simple or unique cause. The Syrian conflict, now civil war, is no exception.” But they start their next sentence, “Still…” and then go on to blame the civil war on the drought. And not only on the drought, but on a drought “made worse by human-induced climate change”.

But like I said, they haven’t even come close to proving, or even suggesting, that this drought had anything to do with global warming.

5) What else do you’d like to point out?

I won’t even get started on their section “Frequency of Multiyear Droughts”. More confusing probability for causality, and in an unnecessarily complicated way.

Let me know if you have more questions,


Journal Bans Wee P-values—And Confidence Intervals! Break Out The Champagne!

I'll drink to that!

I’ll drink to that!

Well, it banned all p-values, wee or not. And confidence intervals! The journal of Basic and Applied Social Psychology, that is. Specifically, they axed the “null hypothesis significance testing procedure”.

I’m still reeling. Here are excerpts of the Q&A the journal wrote to accompany the announcement.

Question 2. What about other types of inferential statistics such as confidence intervals or Bayesian methods?

Answer to Question 2. Confidence intervals suffer from an inverse inference problem that is not very different from that suffered by the NHSTP. In the NHSTP, the problem is in traversing the distance from the probability of the finding, given the null hypothesis, to the probability of the null hypothesis, given the finding. Regarding confidence intervals, the problem is that, for example, a 95% confidence interval does not indicate that the parameter of interest has a 95% probability of being within the interval. Rather, it means merely that if an infinite number of samples were taken and confidence intervals computed, 95% of the confidence intervals would capture the population parameter. Analogous to how the NHSTP fails to provide the probability of the null hypothesis, which is needed to provide a strong case for rejecting it, confidence intervals do not provide a strong case for concluding that the population parameter of interest is likely to be within the stated interval. Therefore, confidence intervals also are banned from BASP.

Holy moly! This is almost exactly right about p-values. The minor flaw is not pointing out that there is no unique p-value for a fixed set of data. There are many, and researchers can pick whichever they like. And did you see what they said about confidence intervals? Wowee! That’s right!

They continue:

…The usual problem with Bayesian procedures is that they depend on some sort of Laplacian assumption to generate numbers where none exist. The Laplacian assumption is that when in a state of ignorance, the researcher should assign an equal probability to each possibility…However, there have been Bayesian proposals that at least somewhat circumvent the Laplacian assumption, and there might even be cases where there are strong grounds for assuming that the numbers really are there…thus Bayesian procedures are neither required nor banned from BASP.

Point one: they sure love to say Laplacian assumption, don’t they? Try it yourself! Point two: they’re a little off here. But they were just following what theorists have said.

If you are in a “state of ignorance” you can not “assign an equal probability to each possibility”, whatever that means, because why? Because you are in a state of ignorance! If I ask you how much money George Washington had in his pocket the day he died, your only proper response, unless you be an in-the-know historian, is “I don’t know.” That neat phrase sums up your probabilistic state of knowledge. You don’t even know what each “possibility” is!

No: assigning equal probabilities logically implies you have a very definite state of knowledge. And if you really do have that state of knowledge, then you must assign equal probabilities. If you have another state of knowledge, you must assign probabilities based on that.

The real problem is lazy researchers hoping statistical procedures will do all the work for them—and over-promising statisticians who convince these researchers they can deliver.

Laplacian assumption. I just had to say it.

Stick with this, it’s worth it.

Question 3. Are any inferential statistical procedures required?

Answer to Question 3. No…We also encourage the presentation of frequency or distributional data when this is feasible. Finally, we encourage the use of larger sample sizes than is typical in much psychology research…

Amen! Many, many, and even many times you don’t need statistical procedures. You just look at your data. How many in this group vs. that group. Just count! Why does the difference exist? Who knows? Not statistics, that’s for sure. Believing wee p-values proved causation was the biggest fallacy running. We don’t need statistical models to tell us what happened. The data can do that alone.

We only need models to tell us what might happen (in the future).

…we believe that the p < .05 bar is too easy to pass and sometimes serves as an excuse for lower quality research.

Regular readers will know how lachrymose I am, so they won’t be surprised I cried tears of joy when I read that.

We hope and anticipate that banning the NHSTP will have the effect of increasing the quality of submitted manuscripts by liberating authors from the stultified structure of NHSTP thinking thereby eliminating an important obstacle to creative thinking. The NHSTP has dominated psychology for decades; we hope that by instituting the first NHSTP ban, we demonstrate that psychology does not need the crutch of the NHSTP, and that other journals follow suit.

Stultifying structure of hypothesis testing! I was blubbering by this point.


Thanks to the multitude readers who pointed me to this story.

Coming Clean On My Global Warming Funding

This is everything.

This is everything.

Since the “Why models run hot” flap started a witch hunt in Congress, I thought I’d give the complete story of my funding for any work of any kind done on climatology since after I graduated Cornell. When I was a student I made about $13,000 a year.

I have received $110 in direct cash payments from donors (for which I am immensely and most sincerely grateful). That’s one hundred ten. United States dollars. This all came within the last three months. So I’m gathering up speed.

A generous reader sent me to the Heartland Conference last year, picking up the tickets, registration, and hotel bill (was it one night or two?). This ran to, say, six or seven hundred. I didn’t speak (wasn’t invited) but I did blog about the event, as regular readers will remember.

When I was a professor at the Cornell Medical School, they sent me to three or four AMS meetings and the like. Call it $3,000.

And one I nearly forgot. I was at the first Heartland Conference many years ago. I got some kind of honorarium for a talk on hurricanes (and boy was I boring: this was before I learned what I now know about statistics), the exact same talk I had given to the AMS annual conference a couple of months before that. This was around $1,000. I did the work on the hurricane papers on my own (I don’t recommend reading them).

Round it up to, say, $5,000. Spread over more than a decade. Because you’re my friends, I can tell you I blew that $110 all on whiskey and cigars.

How much did I get for “Why models run hot”? Nothing.

Of course, you have to balance that hefty five grand by the amount I’ve lost. For instance, I was being groomed to take over a spot at Lawrence Livermore lab a few years back (Ben Santer’s playground). And then, one day, the powers-that-were there suddenly forgot who I was. Emails and phone calls were never returned. Right, Bill?

I had another job lined up with a firm whose purpose was to expose bad science. But they backed off and said they didn’t want themselves to become known as working with a “denier.” I’d tell you who they were, but I signed a non-disclosure. They weren’t the first or last.

An interview I had with a small teaching college (“our students come first” kind of place) is typical. I did so well that I was brought into the office of the chair and shown just where to buy houses (there is obviously more to this story than I’m telling here). But later a department member found my website. This kind of thing has happened more than once.

Although, to be fair, this particular incident was exacerbated because I admitted that I would not write any government grants. I might have been forgiven for being a “denier”. But to refuse to bring in government money? Never. That’s academic freedom for you. It also hasn’t helped me that I don’t want to teach frequentist statistics to impressionable kids and fill their skulls with fallacies. Academic freedom insists you teach what you’re told. Of course, my stance that Diversity equates Mandated Uniformity is somewhat of a taint. Academic freedom again.

Understand I’m not complaining: it’s these schools’ money and they’re free to spend it any way they wish. Until the government makes it spend it the way they insist. I just don’t think the term “academic freedom” has any meaning.

That new CATO group was going to hire me on, at least to write some pieces for them. But I got a call that said one of the VPs there couldn’t work with me because of my stance on gay so-called marriage. Guess I’m a “denier” on that one, too.

And this is why I say on my “Who’s WMB” page that except for two weeks a year I am “completely independent”, a pleasant enough euphemism. At least it’s giving me time to finish this damned book about a better way to think about and do probability and statistics (almost there; stay tuned).

That’s about it. I do not jest when I say I’m thisclose to using what I learned by working with my father and setting up as a handyman.

Fair’s fair. Now let’s hear from the other side. If they have the guts.

Summary Against Modern Thought: That God Is One II

This may be proved in three ways. The first...

This may be proved in three ways. The first…

See the first post in this series for an explanation and guide of our tour of Summa Contra Gentiles. All posts are under the category SAMT.

Previous post.

Finishing the God is one argument this week. Many words, but boy do they flow. If you’ve been following the discussion, this should be a breeze. But if not, review the subjects of essence and existence, and recall the in God they are the same.

Chapter 42 That God is One Two (as in Part two)

[9] Furthermore, given two gods that are posited as agreeing in the necessity of being, either that in which they differ is in some way required for the completion of their necessity of being, or it is not. If it is not, then it is something accidental, because that which accrues to a thing without contributing to its being is an accident. Hence, this accident has a cause, which is, consequently, either the essence of the necessary being or something else. If its essence, then, since the necessity itself of being is its essence, as is evident from what was said above, the necessity of being will be the cause of that accident. But the necessity of being is found in both gods. Therefore, both will have that accident, and thus will not be distinguished with reference to it. If, however, the cause of the accident is something else, it follows that, unless that something else existed, this accident would not exist; and unless this accident existed, the aforesaid distinction would not exist. Therefore, unless that something else existed, these two supposed necessary beings would not be two but one. Therefore, the proper being of each depends on the other, and thus neither of them is through itself a necessary being…

[11] It is, therefore, not possible to posit many beings of which each is through itself a necessary being.

Notes In other words, it isn’t and can’t be turtles all the way down, which each one giving something to another which the other doesn’t have.

[12] What is more, if there are two gods, either the name God is predicated of both univocally, or equivocally. If equivocally, this is outside our present purpose. Nothing prevents any given thing from being equivocally named by any given name, provided we admit the usage of those who express the name. But if it be used univocally, it must be predicated of both according to one notion, which means that, in notion, there must be in both one nature. Either, therefore, this nature is in both according to one being, or according to a being that is other in each case. If according to one, there will not be two gods, but only one, since there cannot be one being for two things that are substantially distinguished. If each has its own being, therefore in neither being will the quiddity be its being. Yet this must be posited in God, as we have proved. Therefore, neither of these two beings is what we understand by the name God. It is, therefore, impossible to posit two gods…

Notes Shorter version: since God’s existence and essence are one, as previously proved, to say there are two (or more) gods is to speak equivocally.

[13] …therefore there cannot be several beings of which each is a necessary being. It is, consequently, impossible that there be several gods.

[15] Furthermore, either the nature signified by the name God is individuated through itself in this God, or it is individuated through something else. If through something else, composition must result. If through itself, then it cannot possibly belong to another, since the principle of individuation cannot be common to several, It is impossible, therefore, that there be several gods.

[16] If, again, there are several gods, the nature of the godhead cannot be numerically one in two of them. There must, therefore, be something distinguishing the divine nature in this and in that god. But this is impossible, because, as we have shown above, the divine nature receives the addition neither of essential differences nor of accidents. Nor yet is the divine nature the form of any matter, to be capable of being divided according to the division of matter. It is impossible, therefore, that there be two gods.

Notes Of course, this follows even if you’re not yet convinced God exists. And if you are not yet convinced, you need to go back and re-read especially Chapter 13. And the material proving God is pure act, actuality only, and in Him there is no potential or accidents (if you like, parts that are not essential).

[17] Then, too, the proper being of each thing is only one. But God is His being, as we have shown. There can, therefore, be only one God…

[19] Furthermore, we notice in each genus that multitude proceeds from some unity. This is why in every genus there is found a prime member that is the measure of all the things found in that genus. In whatever things, therefore, we find that there is an agreement in one respect, it is necessary that this depend upon one source. But all things agree in being. There must, therefore, be only one being that is the source of all things. This is God.

[20] Again, in every rulership he who rules desires unity. That is why among the forms of rulership the main one is monarchy or kingship. So, too, for many members there is one head, whereby we see by an evident sign that he to whom rulership belongs should have unity. Hence, we must admit that God, Who is the cause of all things, is absolutely one…

Notes This is St Thomas being complete. I don’t think these last arguments are convincing on their own. [19] relies on earlier material on being-in-act, and where that ultimately arises. [20] seems to be missing premises about what God is up to. But, of course, we don’t need either of these two and have enough both from last week and this to prove (with certainty) there is only one God who is the ground of all being, the Unmoved Mover, etc.

« Older posts Newer posts »

© 2016 William M. Briggs

Theme by Anders NorenUp ↑