{"id":13359,"date":"2012-11-08T10:18:32","date_gmt":"2012-11-08T08:18:32","guid":{"rendered":"http:\/\/davidbosman.fr\/blog\/2012\/11\/08\/datacratie\/"},"modified":"2012-11-08T10:26:36","modified_gmt":"2012-11-08T08:26:36","slug":"datacratie","status":"publish","type":"post","link":"https:\/\/www.davidbosman.fr\/blog\/2012\/11\/08\/datacratie\/","title":{"rendered":"Datacratie ?"},"content":{"rendered":"

The new megafile didn\u2019t just tell the campaign how to find voters and get their attention; it also allowed the number crunchers to run tests predicting which types of people would be persuaded by certain kinds of appeals. Call lists in field offices, for instance, didn\u2019t just list names and numbers; they also ranked names in order of their persuadability, with the campaign\u2019s most important priorities first. About 75% of the determining factors were basics like age, sex, race, neighborhood and voting record. Consumer data about voters helped round out the picture. \u201cWe could [predict] people who were going to give online. We could model people who were going to give through mail. We could model volunteers,\u201d <\/p>

(\u2026)<\/p>

The analytics team used four streams of polling data to build a detailed picture of voters in key states. In the past month, said one official, the analytics team had polling data from about 29,000 people in Ohio alone \u2014 a whopping sample that composed nearly half of 1% of all voters there \u2014 allowing for deep dives into exactly where each demographic and regional group was trending at any given moment. <\/p>

(\u2026)<\/p>

The polling and voter-contact data were processed and reprocessed nightly to account for every imaginable scenario. \u201cWe ran the election 66,000 times every night,\u201d said a senior official, describing the computer simulations the campaign ran to figure out Obama\u2019s odds of winning each swing state. \u201cAnd every morning we got the spit-out \u2014 here are your chances of winning these states. And that is how we allocated resources.\u201d<\/p>

Inside the Secret World of the Data Crunchers Who Helped Obama Win<\/a><\/p> <\/blockquote>\n\n

Ca pourrait donner :<\/p>\n\n

Le nouveau super-fichier n’a pas seulement indiqu\u00e9 comment trouver des \u00e9lecteurs et retenir leur attention, il a aussi permis aux analystes de r\u00e9aliser des essais pr\u00e9disants quels types de personnes seraient convaincues par quels types d’appels. Les listes d’appels dans les bureaux de campagne, par exemple, ne listent pas seulement  les noms et les num\u00e9ros, elles classaient aussi les noms par ordre de sensibilit\u00e9 \u00e0 la persuasion, avec les priorit\u00e9s les plus importantes de la campagne en premier. Environ 75% des facteurs d\u00e9terminants \u00e9taient basiques, du genre \u00e2ge, sexe, race, voisinage et historique de vote. Les donn\u00e9es de consommation des \u00e9lecteurs ont aid\u00e9 \u00e0 compl\u00e9ter le tableau. "Nous pouvions [pr\u00e9voir] les personnes qui allaient donner en ligne. Nous pouvions mod\u00e9liser les personnes qui allaient donner par la poste. Nous pouvions mod\u00e9liser les b\u00e9n\u00e9voles,"<\/p>

(…)<\/p>

L’\u00e9quipe d’analyse a utilis\u00e9 quatre flux de sondages pour dresser un tableau d\u00e9taill\u00e9 des \u00e9lecteurs dans les Etats cl\u00e9s. Au cours du mois dernier, a d\u00e9clar\u00e9 un officiel, l’\u00e9quipe d’analyse disposait des donn\u00e9es de sondage d’environ 29.000 personnes, rien que pour l’Ohio–un \u00e9norme \u00e9chantillon de pr\u00e8s de la moiti\u00e9 de 1% de tous les \u00e9lecteurs l\u00e0-bas–permettant des analyses en profondeur des tendances de chaque groupe d\u00e9mographique et r\u00e9gional \u00e0 un moment donn\u00e9.<\/p>

(…)<\/p>

Chaque nuit, les r\u00e9sultats des sondages et les informations de contact des \u00e9lecteurs \u00e9taient trait\u00e9s et retrait\u00e9s pour prendre en compte tous les sc\u00e9narios possibles. "Nous avons rejou\u00e9 l’\u00e9lection 66.000 fois chaque nuit", a d\u00e9clar\u00e9 un responsable, d\u00e9crivant les simulations informatiques r\u00e9alis\u00e9es pour d\u00e9terminer les chances qu’avait Obama de gagner chaque Etat ind\u00e9cis. "Et chaque matin, nous avions le r\u00e9sultat–Voici vos chances de gagner ces \u00e9tats. Et voici comment nous avons allou\u00e9 les ressources."<\/p><\/blockquote>\n","protected":false},"excerpt":{"rendered":"

The new megafile didn\u2019t just tell the campaign how to find voters and get their attention; it also allowed the number crunchers to run tests predicting which types of people would be persuaded by certain kinds of appeals. Call lists … Read more →<\/a><\/p>","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[11],"tags":[4],"class_list":["post-13359","post","type-post","status-publish","format-standard","hentry","category-chose-a-dit","tag-societe"],"_links":{"self":[{"href":"https:\/\/www.davidbosman.fr\/blog\/wp-json\/wp\/v2\/posts\/13359","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.davidbosman.fr\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.davidbosman.fr\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.davidbosman.fr\/blog\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.davidbosman.fr\/blog\/wp-json\/wp\/v2\/comments?post=13359"}],"version-history":[{"count":1,"href":"https:\/\/www.davidbosman.fr\/blog\/wp-json\/wp\/v2\/posts\/13359\/revisions"}],"predecessor-version":[{"id":13360,"href":"https:\/\/www.davidbosman.fr\/blog\/wp-json\/wp\/v2\/posts\/13359\/revisions\/13360"}],"wp:attachment":[{"href":"https:\/\/www.davidbosman.fr\/blog\/wp-json\/wp\/v2\/media?parent=13359"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.davidbosman.fr\/blog\/wp-json\/wp\/v2\/categories?post=13359"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.davidbosman.fr\/blog\/wp-json\/wp\/v2\/tags?post=13359"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}