Loading...

Archive for the 'Methodology' Category

28 Feb

About Those Pollster Report Cards, Part III (Response to Blumenthal ‘National Journal’ Column)

Mark Blumenthal, writing in today’s 02/28/08 National Journal, “Telling Good Polls From Bad,” wonders for the third time about which polls and which pollsters ought to be included in any “Pollster Report Card.” I have for two weeks owed Blumenthal a response to his initial question about whether pollsters who release polls closer to Election […]

20 Feb

A Tool To Help Study the Intersection of Poll Methdology and Election Poll Accuracy

How can you tell the difference between one election poll and another? Here’s a start … (keep clicking on the graphic until the size is legible).

Your help filling-in the handful of empty cells (flagged with “?”), is appreciated.
The goal of this undertaking is to better understand the correlation between the methodological choices an election pollster makes and the […]

16 Feb

About Those SurveyUSA Pollster Report Cards: Part I

Mark Blumenthal, the “Mystery Pollster,” writing at Pollster.com, implies several things about SurveyUSA’s Pollster Report Cards that I will address here:
Blumenthal wonders if SurveyUSA uses in its 2008 Pollster Report Cards a measure of poll error that may be cherry-picked to favor SurveyUSA. Blumenthal wonders if SurveyUSA results might be less accurate, compared to other […]

06 Feb

2008 Pollster Report Card Through Super Tuesday, **ACTIVE** Pollsters Only

38 polling firms have polled one or more of the 36 Presidential Primaries and Caucuses to date in 2008.
Of them, 14 have polled 4 or more contests, and are designated as “active” election pollsters. Here’s a report card on the 14 most active 2008 election pollsters, ranked from smallest average error to largest average error […]

06 Feb

2008 Pollster Report Card Through Super Tuesday — Includes All Pollsters

38 polling firms have polled one or more of the 36 Presidential Primaries and Caucuses to date in 2008. Here’s how the pollsters stack-up, ranked from smallest average error to largest average error (where smaller error is better, and zero is perfect).

Complete supporting detail, and caveats, on the jump …

06 Feb

SurveyUSA Super Tuesday Highlights

SurveyUSA polled in 18 separate election contests on Super Tuesday. In 8 of the contests, SurveyUSA was the most accurate pollster. In 2 more contests, SurveyUSA was tied for most accurate. Contests where SurveyUSA polling was best:

1
 
California Democratic Primary
KABC-TV, KPIX-TV, KGTV-TV, KFSN-TV

2
 
Connecticut Democratic Primary
WTNH-TV, WABC-TV

3
 
New York Democratic Primary (tie)
WABC-TV, WGRZ-TV, WHEC-TV, WNYT-TV

4
 
New Jersey GOP Primary […]

25 Jan

UPDATE 2: So It’s All Settled Then: Obama Will Win Tomorrow’s SC Primary. Unless …

 * * Updated 10:15 am ET 01/26/08 with final numbers from Zogby and ARG; list re-sorted.
 =
Last time 14 pollsters all told you that Barack Obama would win a Primary, all 14 were wrong. That was the day before New Hampshire, 18 days ago.
History has an opportunity to repeat itself this weekend: 7 8 9 pollsters […]

24 Jan

What Are South Carolina Independents Trying to Tell Us?

There are enough whitecaps in the SC water to tip a pollster canoe. Consider:

The number of self-described Independents in the SurveyUSA SC Democratic Primary poll released today is up from 7% in December to 16% today.
The number of self-described Republicans in the SC Democratic Primary poll released today is up from 4% in December to […]

29 Oct

Why are Some SurveyUSA Election Polls conducted of ‘Registered’ Voters and Some of ‘Likely’ Voters?

When the candidates and the date of an election are known, SurveyUSA will attempt to seek out and interview “likely voters” for a particular contest.  For ballot measures, SurveyUSA will interview “likely voters” once the ballot language is known, and stable.
However, in contests where the candidates are not yet known, either because the primary (or […]

14 Oct

Does SurveyUSA use RDD or RBS Sampling?

Choosing what phone numbers to dial for an opinion survey is known as “sampling.” The two most common approaches to sampling are:

RDD (Random Digit Dial)
RBS (Registration Based Sampling)

SurveyUSA has completed a number of side-by-side, parallel tests using RDD vs RBS, and to date is the only research company to publish the results of a side-by-side […]

Copyright 2008 SurveyUSA®, Verona NJ, all rights reserved. Terms & Conditions.