American Research Group, Inc.

 

2013 New York City Primary Pollster Ratings

 

The following table lists the accuracy of pollsters in the September 10, 2013 New York City primary using the measure of polling accuracy proposed by Martin, Traugott, and Kennedy.

An accuracy measure of 0.00 indicates that the odds ratio of the survey results matches the odds ratio of the actual results (the closer to 0.00, the better).

 

2013 New York City Democratic Mayoral Primary
de Blasio Thompson Odds
Ratio
Accuracy
Quinnipiac 39% 25% 1.01 0.01 
NYT/Siena 32% 18% 1.15 0.14 
WSJ/NBC4/Marist 36% 20% 1.16 0.15 
AMNY-News12 29% 24% 0.78 -0.25 
PPP 38% 19% 1.29 0.26 
Actual 40.2% 26.0%

The accuracy measures do not differ at the 95% confidence interval.

 

2013 New York City Democratic Comptroller Primary
Stringer Spitzer Odds
Ratio
Accuracy
Quinnipiac 50% 43% 1.06 0.02 
WSJ/NBC4/Marist 45% 47% 0.88 -0.13 
AMNY-News12 43% 46% 0.86 -0.16 
PPP 41% 45% 0.83 -0.18*
NYT/Siena 35% 50% 0.64 -0.44*
Actual 52.2% 47.8%

*The accuracy measures for the Public Policy Polling poll and the New York Times/Siena College poll fall outside the 95% confidence interval.

The accuracy measures for the Quinnipiac poll and the New York Times/Siena College poll differ at the 95% confidence interval.

The Quinnipiac poll was the only poll to correctly predict Stringer as the winner.

 

2013 New York City Republican Mayoral Primary
Lhota Catsimatidis Odds
Ratio
Accuracy
AMNY-News12 50% 28% 1.38 0.32*
Quinnipiac 48% 24% 1.54 0.43  
Actual 52.6% 40.6%

*The accuracy measure for the AMNY-News 12 poll falls outside the 95% confidence interval. The accuarcy measure for the Quinnipiac poll does not fall outside the 95% confidence interval because it had a sample size of just 101 likely Republican primary voters.

ARG Home