Well, not really: Psychology Journal Bans Significance Testing
This is perhaps the first real crack in the wall for the almost-universal use of the null hypothesis significance testing procedure (NHSTP). The journal, Basic and Applied Social Psychology (BASP), has banned the use of NHSTP and related statistical procedures from their journal. They previously had stated that use of these statistical methods was no longer required but can be optional included. Now they have proceeded to a full ban.
The type of analysis being banned is often called a frequentist analysis, and we have been highly critical in the pages of SBM of overreliance on such methods. This is the iconic p-value where <0.05 is generally considered to be statistically significant.
I’m not entirely sure what to make of this. It’s psychology so, eh. It’s true there is pressure to get to the Magic P-value, that is, to show an effect. And I believe I’ve posted a couple of articles about how the vast majority of research that shows “significance” ends up, upon replication, eventually devolving into no effect at all. And yes, in all of my published biomedical research I’ve used p-values. And let’s not even get into the publication bias about showing an effect.
But at least in biomed circles (probably most circles), I believe people don’t generally take the Magic P-value as an end; it’s merely a statistical beginning. Ferinstance, in my latest alcohol research I found an effect (positive) of alcohol consumption. I haven’t treated it as an “Aha!” moment though; in fact, in the Discussion I really minimized it. What you end up doing is attempt to explain the results. In my case, I suspect that the “significant effect” isn’t really an “effect” at all; I think it’s an indicator of something else, probably a more complicated set of factors. And then throw in the usual statement about “more research needs to be done”, which is actually not just a throwaway statement at all: as I suggest above, if you do more and more research and keep getting significant results, while controlling for more and more other factors, you get more confidence that the effect you’re seeing is really an effect.
Then again, it is psychology, so maybe they just want to be able to argue whatever they think is politically and socially correct without having to reference the Tyranny of Numbers.
Megan McArdle, over at Bloomberg, has an article up regarding corporate natural selection: The Church of Wal-Mart
I got a lot of responses to my post last week on Wal-Mart’s decision to raise the minimum wage many of its employees earn to $10 an hour next year. One variety of response stood out: the folks who said “Wal-Mart is doing this because it’s good for its business.”
It stood out because it is almost right, but not quite. The correct statement is that “Wal-Mart is doing this because it thinks it’s good for its business.” Never ignore the possibility that Wal-Mart could be completely wrong.
I remark on this because some of the arguments I saw verged upon what I’ve come to think of as “corporation theology”: the belief that if a corporation is doing something, that thing must be incredibly profitable. This is no less of a faith-based statement than the Immaculate Conception of Mary. Yet it is surprisingly popular among commentators, not just on the right, but also on the left.
What this does is confuse the ideas of intentionality and natural selection. Selection doesn’t really work on intentions, it works on things. No matter your best intentions, if you try to make a car with square wheels it’s not going to work very well (in most contexts). But as humans we like to think of ourselves as problem solvers — “necessity being the mother of invention” and all that — but all that really is, is a way of producing variation that natural selection can work on. It doesn’t matter, from selection’s perspective (to anthropomorphize a little), where the variation comes from, be it random mutation or intentional alteration. It just acts on the results.
But she makes an important point:
Corporations, like all human institutions, are great engines for making mistakes. The only reason they seem so competent is that companies who make too many mistakes go out of business, and we don’t have them around for comparison.
Which links to the above rather nicely: We see as end results those intentional things that worked and so assume the inventor had some kind of foresight that it would work (which may well be the case sometimes), but we forget all the myriad other attempts that never got anywhere. But let’s also remind ourselves that this isn’t an optimality game either, as the QWERTY keyboard makes clear.
Large Hoard of Gold Coins Found in Israel
Almost 2,000 gold coins, discovered by amateur divers near the port city of Caesarea in Israel, form the largest single hoard of medieval gold coins ever found in the country.
Amateur divers and professional underwater archaeologists form the Israel Antiquities Authority have found what they say is the largest hoard of medieval gold coins ever found in Israel. Image credit: Israel Antiquities Authority.
The hoard, dating from the Fatimid period (10th – 12th century CE), was found by a group of amateur divers who immediately reported the find to the Israel Antiquities Authority (IAA).
“The discovery of such a large hoard of coins that had such tremendous economic power in antiquity raises several possibilities regarding its presence on the seabed,” said Dr Kobi Sharvit, director of IAA’s Marine Archaeology Unit
Hard to miss ‘em, as they wouldn’t have corroded at all (photo shows some in situ).
Ancient cities grew like their modern counterparts, archaeologists say
According to archaeological data from a comprehensive study of countless ancient sites in Mexico, a team of archaeologists have discovered ancient cities grew like their modern counterparts.
A team of researchers from the Santa Fe Institute in New Mexico, led by Luis Bettencourt, have been investigating urban development and construction through the lens of social interaction. Bettencourt says that city development leads to the creation of so-called “social reactors,” or the opportunity to magnify social interaction through growth. Larger cities become more efficient and more productive – research shows that doubling a city’s population brings a 15 percent per capita in GDP, wages, and other not-so positive aspects like violent crime through a process that Bettencourt and his team refer to as “urban scaling.”
I’ve seen similar analyses before, but I’d be interested to see how they got “GDP” from the archaeological record.
Archaeology lecture in Stuart, free concert by Kellie Pickler
That would totally bump up the attendance at my next lecture. . . . .
Otherwise known as “click bait”: New archaeological discovery in Israel exposes Neolithic sex, fertility rituals
Archaeologists believe find to be part of a fertility cult,includes stones from over 8,000 years ago
Israeli archaeologist Dr.Uzi Avner discovered stones in the shape of female genitalia and phallic symbols along with figurines of men and women in the Arava desert in southern Israel reported the Israeli news site Ynet Thursday.
The find, which is spread over 102 sites in a 12 kilometer stretch of the Eilat Mountains was dated to be over 8,000 years old to the Neolithic period.
Kind of an oddly written piece and it’s a bit hard to make anything out of it. The photos are also not the best. But hey, the headline is worth the link.
Too bad I just got around to this one, I should have posted it last Saturday: Greek archaeologists find remains of embracing couple in Neolithic tomb
The remains of a man and woman locked in an embrace, which were found at an archaeological site near the Diros Caves in the Peloponnese, constitute a “stunning discovery” dating to around 3,800 BC, the head of the dig has told Kathimerini.
. . .
The remains, which were unearthed last year and in 2013, were analyzed using radiocarbon dating. The results of the analysis, which were made public last week, showed that the bones belonged to a man and a woman, both aged between 20 and 25.
The main reason you don’t find this more often is that the people have to die at the same time. Although I suppose one could somewhat gruesomely pose them like that at different times. But I’ve probably seen this a dozen times so far since starting this thing.
Here’s a pic of it if it comes across:
Prehistoric grave may be earliest example of death during childbirth
Archaeologists say they’ve made a grim discovery in Siberia: the grave of a young mother and her twins, who all died during a difficult childbirth about 7,700 years ago.
The finding may be the oldest confirmed evidence of twins in history and one of the earliest examples of death during childbirth, the researchers say.
Common in that death during childbirth was probably an all-too-often occurrence but rare that you actually find direct evidence of it. Also they say it’s a hunter-gatherer community and it’s rare to find actual cemeteries of HG’s.
Remains of bustling city found near new bridge
“The occupation is heavy,” says Tamira Brennan, the interim field station manager for the Illinois State Archaeological Survey in Fairview Heights. Although the scientists haven’t yet done a population study for the area, they can say it was “densely populated.”
“This is the first big city in North America,” said Brad Koldehoff, the chief archaeologist for the state. “Now we have details, and it’s – wow. Some conjecture had been that all Cahokia moved to East St. Louis, but that’s not it.
“When Cahokia was big, East St. Louis was big; and this was even bigger than people thought.”
Afraid I know nothing about this but wanted to pass it along.