June 25, 2009

Inferences Derived From Rasmussen Data

Rasmussen regularly reports very useful information in the polling universe. Frequently bloggers (myself included) have blogged about his Presidential approval index. The index measures, on a daily basis, the percentage of the American public that strongly approves minus the percentage of the American public that strongly disapproves of the President's performance. In other words a positive rating means more people strongly approve than disapprove.

The problem with these numbers is that their recency, while giving current snapshots, creates a myopic view that encourages an eye away from the broader trends (not from Rasmussen himself, but rather from the reader). With that in mind, I took a quick snapshot of the Presidential Tracking polls provided by Rasmussen and applied some higher level viewing to it in the graphs below.

In the first graph, I looked at the monthly average of 'strongly approve' versus the 'strongly disapprove' results. Clearly at this level, a trend becomes apparent.

Click to enlarge.

Beyond the existing trend, the point of no return intersection of Strong approval versus strong disapproval occurs in July if you forecast the data outward. Personally I think that's a bit soon but based on the existing trend, there is a real rate of change that is occurring.

[Note that the linear logarithmic forecast was used for Disapprove but a third order polynomial regression was used for the Strong Approval. The reason - anything else for strong approval had it falling before now or falling too far, while anything else for strong disapproval had it climbing beyond 100% before year end. Therefore a mixed trending was required.]

The second graph, looks at total approval versus total disapproval. The results are quite similar.

Click to enlarge.

These results show the President falling below the 50/50 threshold some time in September. This is all theoretical of course, as the biggest driving factor is the performance of the economy. If the current appearance of a recovery firms up into something solid, it will solidify his ratings. It would be ironic given that the stimulus money he has planned for has hardly touched the economy yet - in other words the economy would have recovered pretty much without any help from the President.

Looking at graphs like the above is inherently dangerous. Politics and economics are not linear, or for that matter logarithmic or polynomial on any level calculable level. Still these graphs point to a trend that has occurred since Inauguration Day, and they cannot be ignored.

No comments:

Post a Comment

Disagreement is always welcome. Please remain civil. Vulgar or disrespectful comments towards anyone will be removed.

Related Posts Plugin for WordPress, Blogger...

Share This