A US study by the Pew Research Center finds that pre-election polls favor Republican candidates when the pollster only calls landlines, and not cell phones. The gap appears to be growing as more people abandon land lines for cell service. [S]upport for Republican candidates was significantly higher in samples based only on landlines than in dual frame samples that combined landline and cell phone interviews. The difference in the margin among likely voters this year is about twice as large as in 2008. And then there's Skype. This calls to mind the 1948 US presidential election, in which polls (and pundits) predicted a...

23 November, 2010

Harper spokespeople argue that sending the voluntary census long form to a larger number of people will compensate for any loss of data quality due to the newly voluntary nature of the form. Milan Ilnyckyj explains the fallacy. One of the biggest challenges in statistics is collecting a representative sample: finding a subset of the population that will do a good job of approximating the whole group. When a dataset contains a lot of sampling bias and is not reflective of the general population, it is essentially worthless as a guide. That cannot be fixed by using a larger sample size, nor...

23 July, 2010