How the news media erroneously reported the latest survey as an increase in the President’s trust rating

 How the news media erroneously reported the latest survey as an increase in the President’s trust rating (marichulambino)

Based on the reports and media monitor of student serendipity and Mark Sing

     Student “serendipity” quoted the Rappler story on the latest SWS survey on the President’s trust rating, while student Mark Sing reiterated the discussion in class that the sponsors of most surveys are never  disclosed and that the exact wording of  questions in opinion polls are, most of the time, not released to the public.

    What could be added to these is a discussion on how to apply the margin of error for a more accurate reporting of the survey, as follows: 

 Blog admin’s comments: The news media and the SWS (Inquirer, Rappler, ABS-CBN, etc.) storified the latest SWS survey on the President’s trust rating as an increase in the President’s popularity, as follows:

 From Rappler: headline: “Aquino gov’t ratings rebound from record low” xxx “moderate’ +29 net score in the previous quarter, the current government’s satisfaction rating rose to a ‘good’ +35.”

From the Inquirer: “Aquino Administration Satisfaction Rrating Improves   – SWS

        That is, the figures are: from the previous quarter: a +29 or positive 29 , and up to the present quarter: a +35 or positive 35.

       The science or methodology may remain the  same through all surveys,  or may even be  unassailable without the entire results disclosed, but the manner by which the survey is reported out or storified, or spinned by the news media, should be examined by media practitioners and the audience (viewers and readers).

       The margin of error is +/-3% or plus or minus 3 per cent.

     This means that the figures could also be 3 per cent more or 3 per cent less.

      Applying the +/-3% margin of error, or plus or minus 3 per cent margin of error,  that means the +29 could range from a +26 to a  +32  (the President’s trust rating in the previous quarter).

       On the other hand,  the +35 in the current quarter could range from +32 to +38.

        See where the figures meet?

      Because of the plus or minus 3% margin of error, the figures actually meet at +32%.

     In other words, the difference is not statistically significant.

     In other words, Another way of reporting this more accurately is:  the trust rating remains statistically the same between the previous quarter and the current quarter.

       But the news media, or most of the mainstream news media organizations, reported this as: “The President’s trust rating rebounds”  “The President’s trust rating improves” etc.

        It is a sneak peek of things to come. In the next 18 months running up to 2016, there will be more and more storification of surveys on  how this personality or that presidentiable  is leading in the polls, or has rebounded, or how the ratings have improved, even if the differences are statistically insignificant or the small leads are actually statistical ties. 

    the oft-quoted description of surveys are that they are “self-fulfilling prophecies” — when, in reality however, it is the news media that turn  surveys into “self-fulfilling prophecies”  by  distended storification and “spinning” to create a “bandwagon effect”.  

     Perhaps, a paradigm-shift or a metaphor-shift is in order:

      The next 18 months leading up to the 2016 presidential elections is  honeyfication time : when survey firms and the news media turn and spin  favored presidentiables into honey , for donors, campaign contributors, and power-brokers to swarm to —  like ants to sweets.