Emmanuel Didier

Sampling and Democracy: Representativeness in the First United States Surveys

How did statistical representativeness first appear in the United States? To answer this question we turn to the archives of the Department of Agriculture. Since the 1860s, agricultural statisticians have been working to come up with representative groups of farmers able to answer questions about crop production. The archives describe the methods and tools used to devise representative samples. But the explicit justifications of why a given sample is particularly representative are nearly always missing. Where can we find explanations for why one sampling method is representative and not another? Where can we find the why when we have only the how? This paper argues that we can turn to the theorists of democracy for answers, because selecting a sample and selecting a group of representatives have a lot in common.

L'évolution de la délinquance d'après les enquêtes de victimation. France, 1984-2005

Cet article s’attache a construire un indicateur de l’evolution de la delinquance en France au cours des dernieres decennies. Il utilise a cette fin les enquetes nationales de victimation disponibles apres avoir explicite tous les prealables necessaires a cette mise en serie. Afin de solidifier l’operation, les auteurs mobilisent egalement d’autres enquetes nationales plus ou moins proches de leur objet, ainsi que des enquetes de victimation regionales ou locales. L’apport de l’article est double: methodologique quand il parcourt les problemes a resoudre pour mettre en serie les donnees tirees des enquetes de victimation; de fond quand il permet un diagnostic sur l’evolution de la delinquance sur le moyen terme.

The new watchdogs’ vision of science: A roundtable with Ivan Oransky (Retraction Watch) and Brandon Stell (PubPeer):

On March 3rd, 2016, the authors of this note hosted a conference entitled ‘Destabilized Science’ at the University of California, Los Angeles, to which we invited two representatives of core actors within the new science watchdog pack: Ivan Oransky, co-founder in 2010 of Retraction Watch, and Brandon Stell, co-founder in 2012 of PubPeer. After the formal conference, we organized a roundtable to discuss these invitees’ experience and their vision of contemporary science. Mario Biagioli (University of California, Davis), Michael Chwe (UCLA) and Aaron Panofsky (UCLA) participated to the conversation. An edited transcript of the discussion and a short podcast version are being published on Transmissions (ssstransmissions.org) the new blog associated with Social Studies of Science.

Cunning Observation: US Agricultural Statistics in the Time of Laissez-Faire

What does it mean for a small bunch of civil service statisticians to observe something as huge as the whole American agriculture? What are the eyes that can see this? We answer this question focusing on the actual practice of statistical observation at the beginning of the twentieth century. We show that aggregating heterogeneous pieces of observation is necessary— even though it might seem paradoxical—and it requires the statistician be a very cunning observer.

Alain Desrosières’s spectacles: one lens realist, the other constructivist

This paper focuses one central problem that Alain Desrosieres has tackled during his whole career, namely how to articulate the two contradictory natures of statistics, that they are at the same times real and constructed. To understand the different strategies that he followed to solve it, we focus on his writing techniques. He was acutely aware of the importance of rhetoric, as shown among other things by his activity as journal editor at the INSEE and by his paper on the importance of the “writing styles” in the genre of the history of statistics. We show that he wrote first and foremost narratives, in which the subject of the action is very often downplayed of even absent. His narrative techniques allowed him to build a very solid bridge between statistics and politics.