In this Econtalk episode, Doug Rivers explains how random phone calling is not a reliable technique because people don’t like to be called at home or on their cells. Instead he is pioneering a new way of polling where large number of people of whom he knows a lot about participate to input their opinions into a database and are rewarded for their time. However, Bryan Caplan’s theory of rational irrationality posits that the cheaper it is to express beliefs the more irrational they tend to be. With this in mind the fact that people don’t like to participate in polls means that they value their time more than the opinion that is being fished out of them. My reaction is to say: if an individual doesn’t value his opinions very highly himself, why do we have to count them with the same weight as someone who is quite certain of what he believes and is willing to put his money where is mouth is?