Good News from Rasmussen Polls

[louis c]

The daily tracking polls have Bush with just a 1 point lead nationally 48%-47%. In fact, the premium members have it broken down to 47.5 for Bush and 46.8 for Kerry.

http://www.rasmussenreports.com

Further, in the paid Premium feature that I subscribe to, it has even better news on a three day tracking poll conducted over the last three days. Florida is a dead heat at 47% for both Bush and Kerry. Kerry leads in Ohio 49%-46%. Kerry leads in Michigan 48%-42%. Bush has a huge lead in North Carolina at 55%-40%.

However, in the three day tracking poll of the all important 16 battleground states, Rasmussen is reporting a Kerry lead of 50%-45%. In the same battleground states President Bush has a 48% favorable rate to 51% unfavorable rate, while Senator Kerry in the same 16 state survey stands at 53% favorable to 45% unfavorable. Please note that these numbers are more up-to-date than those that appear on the Rasmussen home page and are from his Paid, Premium, site.

If this is accurate, we are more than just in the game, we're ahead.

[kikiek]

1. Thank you thank you thank you..Did I say thank you. Trying to not

worry about polls, but sure love seeing good ones. I can't believe there are still people out there voting for Bush.

[lancdem]

2. I pray this is true

I especially love the Ohio number since Gallup and Zogby gave Bush a large lead there.

[louis c]

8. Remember that Gallup and Zogby

as most major polls, are conducted over a short period of time and subjected to the news of one day.

In addition, their polls are always the result of data collected a week before the numbers become public.

Tracking polls are the only poll's that can give you a snapshot that is consistent and up to date.

[tritsofme]

12. Remember bush winning by 8 in November 2000

I hope to be able to rely on Rasmussen in the 2006 midterms and the 2008 presidential election, but as of right now I cast a wary eye on everything he releases.

[DarthDem]

13. Yeah, But

. . . his problem in 2000 was clear Rethug bias and oversampling. He is a Republican, after all. However, he really seems to have cleaned up his methodology this time around, and his results show it, at least in terms of how well they've been tracking with Zogby, which I use as the benchmark.

If Rasmussen is showing these sorts of leads for Kerry, I completely trust that they're legitimate.

[tritsofme]

14. I'm aware of his changes in methodology

but the fact that he blew 2000 by such a huge margin still looses a bunch of credibility in my book.

For state polls I look for Mason/Dixon and Survey USA first, then Rasmussen.

Research 2000 did fairly well back in 2000, but ARG was just horrible.

While I have a lot of respect for Zogby's national polling, I put his state interactive polls dead last in terms of credibility.

[DarthDem]

15. Makes Sense, Fair Enough

Although I will say that the Zogby state-by-states have been fairly reliable and in tune with common sense, whatever the polling method - - that strange result from Ohio, however, made me wonder if someone (i.e., a Freeper or Freepers) figured out a way to jigger the results.

[tritsofme]

16. I could put out polls that are in tune with common sense

But that doesn't make them scientific or reliable.

I'd be willing to bet that there are quite a number of Republicans from swing states that have found the light and are voting for Kerry on this forum, and vice versa in places like freak republic.

[DarthDem]

18. Yes . . . .

I'm aware that polls can be slanted. I'm missing your gripe here, I think - - if you don't want to believe Rasmussen, I respect your opinion. :) I choose to give it weight this time around, however, and don't see the same problems with Zogby interactives.

[tritsofme]

19. Interactives aren't scientific, that's my problem with them

Like I said, I give Rasmussen more weight in state polling than Zogby Interactive.

[senseandsensibility]

3. Thanks for sharing your paid

premium poll data with us. I'm a poll junkie myself, and thought of subscribing to the poll you cited, but I guess I was too cheap. GOOD NEWS!!! YESSSS!!!!

[louis c]

5. I forgot to add Pennsylvania

to the list. Kerry leads 48%-46%. I added on edit.

[DarthDem]

4. Right

I posted this earlier after subscribing myself - - I had a feeling that the daily tracking polls would contain good news for us. The ridiculous 55-40 result in NC is an obvious outlier; I'll be interested to see if that one gets closer over the next few days as the last days of the RNC drop off the weighted average. Being up three points in OH and tied in FL over that time period, though, is awesome news indeed. Really takes the steam out of the new trend in national polling - - oversampling Rethuglicans dramatically. :eyes:

[brook]

6. I appreciate that.

It's music to my heart - and wonderful that you share it with us.

[City Lights]

7. Thanks for sharing with us.

And please let the sharing continue. :-)

[DarthDem]

10. It Was Only $49

I'll be sure to post it every day if I don't see it done by someone else first . . . .

[meti57b]

9. Thank you!!!!!!!!!!!!!!!!

[louis c]

25. You're welcome

[Andromeda]

11. Thank you for the good work!

This is such a volatile campaign season that I'm a nervous wreck. Seeing polls like this helps me keep my hopes up.

[DemMother]

17. Another thanks

I'm not entirely sold on Rasmussen, but he's far more believable than Survey USA, which showed Bush only 3 points behind Kerry in California.

I did write Rasmussen a quick note tonight, however. When Kerry was at 46.5 percent, he rounded him down to 46, giving Bush a two-point lead, instead of less than two points. Now that Bush is at 47.5, he's rounding his total up to 48. I'm not sure that's evidence of his so-called bias (and I admit it's pretty petty, but it annoyed me.

I do check in on the tracking poll once a day and will appreciate seeing your updates on state polling here on DU.

[louis c]

21. It's the rounding of the rounding

for example, if Kerry is really at 44.48%, the Premium members get 44.5%, the Home Page will show 44%. If Kerry is at 44.53%, the Premium members will get 44.5%, but the Home page will show 45%.

If we're going to follow polling data, we have to trust someone. I subscribed to Rasmussen three months ago, so I just don't pick and choose who to believe.

I follow this poll because it is tracking 1,000 voters a day, and rolling that average over a three day period so you get the results of 3,000 respondents every daily announcement.

It used to be 500 and 1500, but he increased it after labor day. His tracking of the Key battleground states is at a rate of 500 a day, rolling for three days for a total of 1500 per state. You can't ask for better information than that.

[blue neen]

20. Thanks for the good news.

Now could someone do me a favor and explain "daily tracking polls" to me. What is the difference between that and the other polls?

Thanks.

[louis c]

22. Well, here goes

We start with a sample of 500 voters on day one. On day two, we poll 500 more, On day three, we do the same. We then give a result of the 1,500 respondents, say, for example Kerry 49% to Bush 45%.

Then we poll 500 more on day four, and drop the data from day one, so your average is from the 1,500 respondents from day 2,3, and 4. We come up with the new figure. Kerry 48% to Bush 46%

On day 5, another 500 folks are polled, and we drop day 2 and average the 1,500 from day 3,4,and 5. And so on and so on. You can judge the movement from day to day.

Rasmussen has had this tracking, or rolling data, every day since early March. It takes out a statistical anomaly, a one day bump, or hot news item that may skew a single poll. It is more like a moving picture, or time lapse photography, instead of a single snapshot.

[blue neen]

23. Seems like it would be apt to be more accurate that way.

Thanks for the explanation--you explained it very well!

[louis c]

24. By the way

the home page is free access, which gives you the daily results at noon, seven days a week.

just click on, and save the site, and you can enjoy the daily tracking poll, free of charge, every day.


http://www.rasmussenreports.com