In the plane to Warwick on Monday, I was reading my latest issue of Nature and found an interesting editorial on the financial plight of many graduates and post-docs in both the US and the UK (and certainly elsewhere). Who, despite having a fellowship, cannot make ends meet. This is particularly true in expensive cities like London, Oxford or even Paris, where rents force those new researchers to face long commuting hours. The editorial suggests taking extra-jobs to make up for financial difficulties, but this does not sound to me like a particularly pertinent recommendation if it means taking time off one’s research, at the period in a researcher’s career where one’s energy should be mostly directed at the production of papers towards securing a (more) permanent job. Even teaching can prove too time consuming for finishing PhD students. An adequation between the needs of those young researchers and the institutional support they receive would sound like a natural requirement, while graduates looking for fellowship should truly assess the adequation in detail before accepting an offer.Which of course is not always easy. In countries where post-doctoral contracts are not negotiable and are set at a national level (like, e.g., France), checking with earlier fellows is a must. (As it happens or happened, I was quite lucky to spend my post-doctoral years in cheap places with decent support from the local universities, but this is not relevant in today’s environment!)
Archive for Nature
A very interesting issue of Nature I read this morning while having breakfast. A post-brexit read of a pre-brexit issue. Apart from the several articles arguing against Brexit and its dire consequences on British science [but preaching to the converted for which percentage of the Brexit voters does read Nature?!], a short vignette on the differences between fields for the average time spent for refereeing a paper (maths takes twice as long as social sciences and academics older than 65 half the time of researchers under 36!). A letter calling for action against predatory publishers. And the first maths paper published since I started reading Nature on an almost-regular basis: it studies mean first-passage time for non-Markov random walks. Which are specified as time-homogeneous increments. It is sort of a weird maths paper in that I do not see where the maths novelty stands and why the paper only contains half a dozen formulas… Maybe not a maths paper after all.
“In an age in which ancient genomes can reveal startling links between historical populations, we should ask not just whether remains should be reburied, but who decides and on what grounds.”
An article in Nature described the story of fairly old remains (of the Kennewick Man) in North America that were claimed for reburial by several Native American groups and that were found to be closer [in a genetic sense] to groups that were geographically farther (from South America and even Australian aboriginal Australians). What I find difficult to understand (while it stands at the centre of the legal dispute) is how any group of individuals can advance a claim on bones that are 8,000 year old. With such a time gap (and assuming the DNA analysis is trustworthy) the number of individuals who share the owner of the bones as one ancestor is presumably very large and it is hard to imagine all those descendants coming to an agreement about the management of the said bones. Or even that any descendant has any right on the said bones after so many generations which may have seen major changes in the way deceased members of the community are treated. I am thus surprised that a judiciary court or the US government could even consider such requests.
A paper in Nature this week that uses reversible-jump MCMC, phylogenetic trees, and Bayes factors. And that looks at institutionalised or ritual murders in Austronesian cultures. How better can it get?!
“by applying Bayesian phylogenetic methods (…) we find strong support for models in which human sacrifice stabilizes social stratification once stratification has arisen, and promotes a shift to strictly inherited class systems.” Joseph Watts et al.
The aim of the paper is to establish that societies with human sacrifices are more likely to have become stratified and stable than societies without such niceties. The hypothesis to be tested is then about the evolution towards more stratified societies rather the existence of a high level of stratification.
“The social control hypothesis predicts that human sacrifice (i) co-evolves with social stratification, (ii) increases the chance of a culture gaining social stratification, and (iii) reduces the chance of a culture losing social stratification once stratification has arisen.” Joseph Watts et al.
The methodological question is then how can this be tested when considering those are extinct societies about which little is known. Grouping together moderate and high stratification societies against egalitarian societies, the authors tested independence of both traits versus dependence, with a resulting Bayes factor of 3.78 in favour of the latest. Other hypotheses of a similar flavour led to Bayes factors in the same range. Which is thus not overwhelming. Actually, given that the models are quite simplistic, I do not agree that those Bayes factors prove anything of the magnitude of such anthropological conjectures. Even if the presence/absence of human sacrifices is confirmed in all of the 93 societies, and if the stratification of the cultures is free from uncertainties, the evolutionary part is rather involved, from my neophyte point of view: the evolutionary structure (reproduced above) is based on a sample of 4,200 trees based on Bayesian analysis of Austronesian basic vocabulary items, followed by a call to the BayesTrait software to infer about evolution patterns between stratification levels, concluding (with p-values!) at a phylogenetic structure of the data. BayesTrait was also instrumental in deriving MLEs for the various transition rates, “in order to inform our choice of priors” (!). BayesTrait has an MCMC function used by the authors “to test for correlated evolution between traits” and derive the above Bayes factors. Using a stepping-stone method I am unaware of. And 10⁹ iterations (repeated 3 times for checking consistency)… Reversible jump is apparently used to move between constrained and unconstrained models, leading to the pie charts at the inner nodes of the above picture. Again a by-product of BayesTrait. The trees on the left and the right are completely identical, the difference being in the inference about stratification evolution (right) and sacrifice evolution (left). While the overall hypothesis makes sense at my layman level (as a culture has to be stratified enough to impose sacrifices from its members), I am not convinced that this involved statistical analysis brings that strong a support. (But it would make a fantastic topic for an undergraduate or a Master thesis!)
Simon Gächter and Jonathan Schulz recently published a paper in Nature attempting to link intrinsic (individual) honesty with a measure of corruption in the subject home country. Out of more than 2,500 subjects in 23 countries. [I am now reading Nature on a regular basis, thanks to our lab subscribing a coffee room subscription!] Now I may sound naïvely surprised at the methodological contents of the paper and at a publication in Nature but I never read psychology papers, only Andrew’s rants at’em!!!
“The results are consistent with theories of the cultural co-evolution of institutions and values, and show that weak institutions and cultural legacies that generate rule violations not only have direct adverse economic consequences, but might also impair individual intrinsic honesty that is crucial for the smooth functioning of society.”
The experiment behind this article and its rather deep claims is however quite modest: the authors asked people to throw a dice twice without monitoring and rewarded them according to the reported result of the first throw. Being dishonest here means reporting a false result towards a larger monetary gain. This sounds rather artificial and difficult to relate to dishonest behaviours in realistic situations, as I do not see much appeal in cheating for 50 cents or so. Especially since the experiment accounted for a difference in wealth backgrounds, by adapting to the hourly wage in the country (“from $0.7 dollar in Vietnam to $4.2 in the Netherlands“). Furthermore, the subjects of this experiment were undergraduate students in economics departments: depending on the country, this may create a huge bias in terms of social background, as I do not think access to universities is the same in Germany and in Guatemala, say.
“Our expanded scope of societies therefore provides important support and qualifications for the generalizability of these theories—people benchmark their justifiable dishonesty with the extent of dishonesty they see in their societal environment.”
The statistical analysis behind this “important support” is not earth-shattering either. The main argument is based on the empirical cdfs of the gain repartitions per country (in the above graph), with tests that the overall empirical cdf for low corruption countries is higher than the corresponding one for high corruption countries. The comparison of the cumulated or pooled cdf across countries from each group is disputable, in that there is no reason the different countries have the same “honesty” cdf. The groups themselves are built on a rough measure of “prevalence of rule violations”. It is also rather surprising that for both groups the percentage of zero gain answers is “significantly” larger than the expected value of 2.8% if the assumption of “justified dishonesty” holds. In any case, there is no compelling argument as to why students not reporting the value of the first dice would naturally opt for the maximum of the two dices. Hence a certain bemusement at this paper appearing in Nature and even deserving an introductory coverage in the first part of the journal…
While in Warwick this week, I borrowed a recent issue (Oct. 08, 2015) of Nature from Tom Nichols and read it over diners in a maths house. Its featured topic was reproducibility, with a long initial (or introductory) article about “Fooling ourselves”, starting with an illustration from Andrew himself who had gotten a sign wrong in one of those election studies that are the basis of Red State, Blue State. While this article is not bringing radically new perspectives on the topic, there is nothing shocking about it and it even goes on mentioning Peter Green and his Royal Statistical Society President’s tribune about the Sally Clark case and Eric-Jan Wagenmakers with a collaboration with competing teams that sounded like “putting one’s head on a guillotine”. Which relates to a following “comment” on crowdsourcing research or data analysis.
I however got most interested by another comment by MacCoun and Perlmutter, where they advocate a systematic blinding of data to avoid conscious or unconscious biases. While I deem the idea quite interesting and connected with anonymisation techniques in data privacy, I find the presentation rather naïve in its goals (from a statistical perspective). Indeed, if we consider data produced by a scientific experiment towards the validation or invalidation of a scientific hypothesis, it usually stands on its own, with no other experiment of a similar kind to refer to. Add too much noise and only noise remains. Add too little and the original data remains visible. This means it is quite difficult to calibrate the blinding mechanisms in order for the blinded data to remain realistic enough to be analysed. Or to be different enough from the original data for different conclusions to be drawn. The authors suggest blinding being done by a software, by adding noise, bias, label switching, &tc. But I do not think this blinding can be done blindly, i.e., without a clear idea of what the possible models are, so that the perturbed datasets created out of the original data favour more one of the models under comparison. And are realistic for at least one of those models. Thus, some preliminary analysis of the original or of some pseudo-data from each of the proposed models is somewhat unavoidable to calibrate the blinding machinery towards realistic values. If designing a new model is part of the inferential goals, this may prove impossible… Again, I think having several analyses run in parallel with several perturbed datasets quite a good idea to detect the impact of some prior assumptions. But this requires statistically savvy programmers. And possibly informative prior distributions.