As many have surmised, the polling system has not changed with the times. The methodology that most pollsters use is badly outdated. This simple truth is finally being addressed by the website RealClearPolitics which is now changing how they report on the polls. Liberty Nation has been talking about poor polling ad nauseam. We’ve run article after article questioning the methodology of surveys, dug into the cross tabs, and repeatedly shown the undercounting of conservatives. Much of this data is off base, leading some to wonder whether this is on purpose or not. Are pollsters reflecting the electorate or trying to affect them, as the founder of The Trafalgar Group, Robert Cahaly, so succinctly put it? Either way, the broken system must be addressed, and it appears RCP poll reporting is doing just that.
RCP Poll Reporting Changes
Poll watchers may have noticed that RealClearPolitics has added some vital pieces of information to its daily reporting of averages. One new column is labeled “Polls Underestimated,” and another reads “Adjusted Poll Avg.” Statistical data based on the results of recent elections also appears on the chart along with which party was undercounted and by how much. These additions illustrate what conservatives already know – that the polls massively undercount Republicans and occasionally undercount Democrats. However, the undercounting of Democrats is usually a fraction of a point, while the Republicans have suffered a much larger error rate.
RealClearPolitics
The website now openly recognizes that “the public opinion survey industry has undergone momentous changes and faced a series of challenges in a profoundly altered political landscape.” This course correction in reporting polls is outlined in an article titled “RealClearPolitics Polling Accountability Initiative.” Its stated goal is to “improve public trust and confidence in political polling.”
The article discusses the changing landscape of how average Americans communicate in the last two decades: There was no social media back then, and the switch from landlines to smartphones has been nothing short of historic. RCP notes that only 40% of households maintain a landline in 2022 and 97% of Americans now own cell phones – a drastic difference from how people communicated in 2002. They point out what Liberty Nation and others have complained about; the average pollster is using outdated methods to survey the public.
Trafalgar’s Cahaly has been at the forefront of recognizing this problem. That RCP acknowledges this and is making changes in its reporting is a healthy start toward more accuracy and less slop in publishing and averaging out polls. As LN has previously reported, “averaging out rubbish – only gives you the average of rubbish.”
RealClearPolitics has also added a rating feature. This is not new, as pollster FiveThirtyEight has been doing this for some time, but it’s nice to have another source holding the pollsters’ feet to the fire. The hope is that this accountability will force pollsters to bring their methodology into the 21st century. As The Trafalgar Group has illustrated, new polling methods can work if an organization is willing to take the time and effort to get it right.
Many conservatives have known about the problem of inaccurate polls for some time now. More often than not, when discussing opinion surveys, one is likely to hear, “I don’t pay attention to polls anymore.” While this represents many Americans, millions of poll watchers are still out there. For those who like to consult the surveys of voters before an election, a move is afoot toward more accuracy and less wishful thinking. And that’s a step in the right direction – not just for RCP but for the American people.
