Professor Rohas News Today

Think campaigns -- you mentioned the study. for candidates who cant afford, who dont have access to what we would describe as quality polling. that this is a way to judge sentiment. is this the kind of thing you would expect campaigns to use, not solely but supplementary in campaigns to come? >> thats absolutely right. one of the biggest implications of this research is not in prediction. you can always, you know, work with different models. you can always get one prediction over another predict...

Joining me now, one of the authors of the study, professor rohas. lets start with the critique. im sure when you published this, you knew a critique was doing. he basically says this conclusion flies in the face of logic. give me your quick take on that. >> so let me start by saying i appreciate what he wrote about us. i appreciate the attention. and id like to start with the response by saying that random samples are not the only way to learn about the world. there are other ways you can learn ...