stats · politics · spain

El problema del PP y la ventaja de Podemos en un solo gráfico

Una cosa es que no te voten quienes eran antes tus simpatizantes. Otra muy distinta es que no se vean votándote otra vez en el futuro. Y esa es la idea que refleja el siguiente gráfico: De cuando en vez, el CIS pregunta por la intención de voto en forma de probabilidad. En lugar de poner al entrevistado en una situación similar a la que enfrentará en las urnas y hacer que escoja una única opción dentro del menú de partidos, el cuestionario le permite asignar una probabilidad de voto a cada uno de los partidos. Podría parecer que la... Read more ›

politics · data · technology

To save everything click here

Morozov’s discussion of technical solutionism in politics in To Save Everything Click Here is unfair. It is difficult to disagree with his position about the dangerous naiveté of the interaction between technology and society in authors like Clay Shirky or businessmen like Jeff Bezos. But he tries to make his arguments pass for an overaching discussion of solutionism when in fact he is limiting himself to a very distinctive kind of political theory for which technology is supposed to provide the golden bullet to transition to a very particular utopia. In this view, the difference between old and new political... Read more ›

politics · stats

No, not liking your yogurt will not bring police to your door

Reading Morozov’s To Save Everything, Click Here, I came across this nugget: Thus, even tweeting that you don’t like your yogurt might bring police to your door, especially if somone who tweeted the same thing three years before ended up shooting someone in the face later. Morozov is talking about predictive models of criminal behavior based on “tweets or Facebook updates that exploit nonthreatening verbal cues that tend to precede criminal acts.” He is obviously trying to be provocative but it points to a misunderstanding of how prediction works that makes him target the wrong problem on an relevant issue.... Read more ›

politics · stats

Campañas electorales cuantitativas

Hay una distinción que parece escaparse a los artículos que he visto últimamente en prensa sobre la transformación tecnológica en las campañas electorales. Big data y experimentos, aunque estén intimamente relacionados, se refieren a aspectos cualitativamente diferentes del mismo proceso de penetración del instrumental analítico en la gestión de las campañas electorales. Las campañas han aprendido a usar los experimentos para evaluar la efectividad de diferentes los modo de contacto con la masa de votantes. Eso ha servido para cuantificar retornos sobre la inversión de diferentes estrategias de contacto y para estimar que, por ejemplo, que cada voto adicional cuesta... Read more ›

methodology · surveys

A study in the complementarity of phone and online surveys

The drop in the penetration of landlines has a clear effect on the ability of pollsters to reach some segments of the population using the standard phone surveys. Namely, younger people (although the problem is obviously not exclusive to that group) not only are less likely to have a landline, they are also more likely to not answer the phone. I very much like this experiment that the GESOP carried last year on the complementarity of phone and online surveys as a way to handle this problem in Spain. As they state in the presentation of the report: The test... Read more ›