Tuesday, November 23, 2004

Where's the NOW in the poll as reported by Nagourney & Elder?

Still wading through the New York Times/CBS News polling data and a new question arises.
 
In "Americans Show Clear Concerns on Bush Agenda" (http://www.nytimes.com/2004/11/23/national/23poll.html?hp&ex=1101272400&en=3ea63a4e6a76d17d&ei=5094&partner=homepage) Adam Nagourney and Janet Elder end their jointly written piece with this final paragraph:
 
Finally, in one bit of presumably good news for a party that is looking for it, Americans now have a better opinion of the Democratic Party than of the Republican Party:  54 percent said they had a favorable view of Democrats, compared with 39 percent with an unfavorable view.  By contrast, 49 percent have a favorable view of Republicans, compared with 46 percent holding an unfavorable one.
 
 
I've emphasized "now" for a reason.  How do you define now?  To me, it suggests that this a new occurrence.  But looking at the polling data they provide for past polls on these topics, there's not a great deal of difference in the favorable opinion of Democrats now as opposed to elsewhere in the last four years.  The all time low for "favorable" for Democrats was 45% in the poll taken "11/20-24/02" and the all time high was 58% 1/21-24/02.  In these five polls listed for this year we see:
 
54%     1/12-15/04
51%     3/10-14/04
55%     7/11-15/04
52%   10/14-17/04
54%   11/18-21/04
 
"The nationwide telephone poll of 855 adults has a margin of sampling error of plus or minus three percentage points."
 
 
Are the above figures statistically significant?  No.    From the lowest figure (51%) to the highest (55%) we're looking at a difference of four percent.  There's no "now" to trumpet in this poll as best I can determine.
 
According to the data, in "2/00" the Dems spiked to 61% then fell back to 54% in "5/00".  Immediately prior to this spike, they were at 52% in "11/99."  There's a "now" here?  A "continue" absolutely, the trend continues that's what the poll demonstrates.  It doesn't demonstrate "Finally, in one bit of presumably good news for a party that is looking for it, Americans now have a better opinion of the Democratic Party than of the Republican Party . . ."
 
Write,  "Finally, in one bit of presumably good news for a party that is looking for it, Americans continue to have a better opinion of the Democratic Party than of the Republican Party . . ."  That's accurate.  Or focus on the fluctuations in "not favorable" responses for Democratic which show small shifts from time to time in the surveys of the last four years.  (The high is 45% of the respondents saying they have an unfavorable opinion of the Democrats in "5/9-12/03"; the low is 31% in "2/00.")
 
 
Oh but maybe the "now" means that this survey, for the first time, found the Democratic party scoring higher on favorables than the Republican party?
 
Okay, let's focus on the last four years by looking at both parties for a second -- maybe that's where the "now" comes from? 
 
 
There are eighteen polling periods in the last four years.  In 12 of them (which would be in 2/3 of the polls for the last four years), respondents ranked the Democratic Party more favorably than the Republican Party; in 3 polling periods, respondents ranked the Republican Party more favorably; and in 3 polling periods, respondents ranked both parties as equally favorable beginning with the first poll after Sept. 11, 2001 -- "01/21-24/02" -- and continuing for the polls covering the periods "7/13-16/02" and "10/3-5/02."
So for the first three polls listed after Sept. 11, we see each party ranked the exact same (that's right, EXACT; first they both get 58%, then they both drop down to 53% and then again at 53%). 
 
 
The three where the Republicans get higher rankings than Democrats are:
 
51%     11/20-24/02  (Democrats have 45%)
53%       5/9-12/03    (Democrats have 46%)
53%     12/14-16/03  (Democrats have 51%)
 
In the five polling periods since, the Democratic Party has been ranked more favorable than the Republican party, but Nagourney & Elder tell us "now"?
 
Let's remember the 3% margin of error.  Look at the "12/14-16/03" numbers.  They are well within the plus or minus 3%.  So not since May of 2003 has the Republican Party been ranked more favorable by respondents in these polls.  And this is "now"?
 
 
I don't see polling data from this poll that provides a "now."  I see the indication of a continuing trend in the polling data.  But there's no now there.
 
Look at the data, maybe I'm missing something.  Feel free to post or to e-mail at
common_ills@yahoo.com any comments. 
 


Do you Yahoo!?
Meet the all-new My Yahoo! – Try it today!