Political pollsters who solicit public opinion via mobile phones are facing huge obstacles. Chief among these is the tendency for mobile phone users to screen out pollsters’ calls. If their samples don’t reflect the views and behaviors of the population of interest, their results have little predictive value. And non-response bias, says Fred Conrad – or the tendency for people who respond to pollster calls to be different from those who don’t – is a significant consideration in phone polling.

More about ISR’s Conrad says mobile phone polling faces non-response bias >