Pollster.com

Articles and Analysis

 

Obama Approval House Effects

Topics: House Effects

Obama Approval:
AllHousesApp.png


While many pollsters have yet to do very many presidential approval polls, we have finally built up enough to take a look at house effects. The red lines & dots are each pollster, the blue line is our standard trend estimate. I'll let the pictures have their thousand words and leave it at that for now.

Obama Disapproval:
AllHousesDisapp.png

 

Comments
poughies:

These charts just make you ask why Zogby even bothers with these internet polls. And it isn't just non-probability samples in general, it's theirs. You would think that a non-probability sample bias against Obama would be seen across the board...

What I find especially interesting is Zogby's polling for many conservative causes as of late (League of American Voters for instance). No, I'm not saying they are doing something with polling to support conservatives (Zogby ran for Mayor? or some post in the 80s as a very liberal Democrat), but perhaps it is something with the phrasing of questions.... or their sampling frame. Something with likely voters... something related with that. If so, we see the same thing reflected with Rasmussen's likely voter frame.

Just thinking aloud here.

____________________

Art:

Note that Democracy Corps, in spite of its partisan leanings, pretty much nails it on both approval and disapproval. They must be doing something right.
NBC/WSJ is dead on on approval but slightly off on disapproval.

____________________

Art:

A further note. I'm not going to try the math but visually it appears Gallup's daily and Rassmussen's daily essentially cancel each other out.

____________________

Art-- you nailed it. The reason including Rasmussen has very little effect on our overall trend is that we also include Gallup and the two have (so far) very nearly balanced each other on approval. If you estimate the trend without either Rasmussen or Gallup, the resulting trend estimate is within a couple of points (often less that a point) of the trend with them both included. Of course the trend is noisier when you take them both out, so it jumps around a bit more, and the balance between them could change. But since we've been able to compute a trend for Obama job approval, this balance has held up. You can try it yourself with the interactive approval charts-- just uncheck Rasmussen and Gallup Daily and see what you get.

As of right now, our standard trend is 48.6 and that becomes 51.3 without Rasmussen and Gallup. That's one of the higher discrepancies we've seen since January. Right now we see a dip in Gallup and Rasmussen plus Fox. If other non-daily polls start to show the same drop this week, the two estimates should be closer later in the week. (And of course they could show the decline in the last week is ephemeral.)

Thanks all,

Charles

____________________

student1776:

Rasmussen sample reflects people more likely to vote (registered, has voted in past). As a consequence Rasmussen is more accurate in predicting electoral outcome but is less accurate in reflecting the overall population zeitgeist - since they de-emphasize groups less likely to vote (e.g. the young) who tend to be more leftist.

____________________



Post a comment




Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR