Parents frequently say that kids should not believe everything they hear on TV.  Is this true?  And does it apply to what people read in newspapers, too?  We sought to test the powers of prediction possessed by political pundits and politicians, evaluate their accuracy, and test the qualities that made for good predictors.  TV may (or may not) harm kids’ brains, but it turns out that parents probably should not believe most of what they hear on TV, either.

          Millions of Americans tune in for Sunday morning talk shows and read weekly op-ed columns to gain insight into current events and global politics and to anticipate what is coming next.  They watch politicians, pundits, and journalists pontificate around the “round table” and read the opinions of newspaper columnists each week, but what do we stand to gain from this?  It is possible that prognosticators offer deeper analysis of current issues and can more accurately predict future outcomes.  It is possible that prognosticators serve solely as entertainers.  Alternatively, prognosticators may present opposing viewpoints that are representative of the wide range of ideologies in our political spectrum.  Regardless, one has to wonder whether certain prognosticators are more valuable than others.

          We explored these questions through a large-scale analysis of the predictions made by politicians, pundits, and journalists in both TV and print.  Taken together, we called predictors from all groups prognosticators.  We evaluated the predictions of prognosticators over a 16-month period, from the beginning of September 2007 to the end of December 2008.  This was designed to encompass the 2008 election season and capture the largest possible number of predictions.  By analyzing the prognosticators’ statements in transcripts and printed columns, we sought to find and test their predictions.  Our data yields results with implications that begin to answer some of our initial questions. 

          It turns out that we, as consumers, should be watching and reading certain prognosticators over others based on their predictive ability. In addition, political topics tend to elicit certain predictive qualities, such as being phrased using positive/negative or extreme language.  When we created an absolute, holistic scale on which we could place individual prognosticators, certain individuals did markedly better than others.

          To our knowledge ours is the first attempt at creating a “consumer report” for the average American who wants to know when to tune in and who to take most seriously.  Our hope is that with enhanced accountability, prognosticators will become better--and with enhanced visibility, citizens will be more discriminating, listening to more accurate predictors.