Connecticut Vote Senate 2006

Anderson Analytics is currently testing new audio mining technology on the Connecticut race for Senate 2006.

Over the next few days we will be posting limited findings in aggregate on this page. If you are a candidate or reporter and would like additional data please contact Anderson Analytics ( inquiries [at] andersonanalytics.com).

Example audio excerpts from the survey:

Audio 1 Audio 2 Audio 3 Audio 4

Anderson Analytics --- BigEars is a whole new way to listen to the voice of the customer.

Ideal for gaining a deeper understanding on issues that may be emotional, the real-time detail data allows you access to both transcripts as well as actually hearing the sound file/respondent tone of voice and inflection for each question.

The new technology allows you to quickly, easily and inexpensively collect short survey data over the telephone or VOIP. Anderson Analytics can also apply text mining to the results allowing for quantitative analysis as well as qualitative.

Survey Results

Q: Regardless of who you plan to vote for, who do you think is most likely to win the election?

Q: And why, specifically, do you think they will win? (Open-ended questions)


Because (of)...
Total %

% of people who think Lamont will win and provided explanation

% of people who think Lieberman will win and provided explanation
he is more qualified
35.2
31.1
36.4
the expereince
23.9
6.7
30.8
the war
18.2
26.7
3.7
his integrity
13.2
8.9
15.0
people need (or are afraid of ) change
8.2
22.2
1.9
voters will be crossing party line
8.2
2.2
11.2
he is good for Connecticut
7.5
4.4
8.4
questions about his (or his opponent's) source of money
6.9
4.4
6.5
what the press is already saying
6.3
6.7
5.6
he is (or is not) the incumbent
5.0
0.0
7.5
he cares about the people (particularly the poor)
5.0
8.9
3.7
his record and service
4.4
0.0
6.5
he is good for the country
3.8
0.0
4.7
his (or his opponent's) support for the president
3.1
4.4
2.8
ties to (or no ties to) Washington
2.5
6.7
0.9
domestic issues
1.9
6.7
0.0


*The base size of respondents who provided explanations for why they
think candidate Schlesinger will win is too small for comparison

 

 

Q: And why, specifically, do you think they will win? (Open-ended question)
Reasons by party affiliation


Because (of)...

Democrats

Republicans

Independents
he is more qualified
23.7
30.8
28.6
the expereince
18.4
28.2
17.9
the war
23.7
10.3
10.7
his integrity
6.6
20.5
12.5
people need (or are afraid of ) change
10.5
0.0
7.1
voters will be crossing party line
10.5
5.1
5.4
he is good for Connecticut
6.6
12.8
1.8
questions about his (or his opponent's) source of money
6.6
5.1
5.4
what the press is already saying
5.3
5.1
5.4
he is (or is not) the incumbent
2.6
7.7
3.6
he cares about the people (particularly the poor)
3.9
7.7
1.8
his record and service
3.9
2.6
3.6
he is good for the country
2.6
5.1
1.8
his (or his opponent's) support for the president
3.9
2.6
1.8
ties to (or no ties to) Washington
2.6
0.0
3.6
domestic issues
3.9
0.0
0.0

 

 


Other interesting analysis:

C&RT

Neural Net

Web Chart

Psychological Profiling - Text Analysis of Voter Feedback

Anderson Analytics found a significant difference in the feedback on the psychological attribute ‘Task Orientation’. Task Orientation is higher among those who think Lamont will win than among those who think Lieberman will win. Naturally we would expect a high correlation between those expecting a candidate to win and supporting/voting for that candidate in the election.

As Lamont predictors/supporters are much more task oriented than Lieberman supporters, they are viewing the campaign as a solvable problem and that with group cohesion and cooperation can succeed. This is akin to a "rally the troops and we can win" philosophy. Surprisingly, this was not as evident among those who believe Lieberman will win. This may be because Lieberman currently seems to have a broader support base, and one which is therefore not as devout. The lack of this mentality among the pro-Lieberman voters may also be due to the fact that Lieberman is reported to be ahead in recent polls.



 

 

 

 

 

 

 

About this study :

This independent poll was conducted by Anderson Analytics to test new BigEars automatic telephone interviewing technology. Respondents were Connecticut Residents selected by Random Digit Dialing (RDD) in area codes 203 & 860. Only respondents who said they planned to vote in the November 7 election were surveyed. Fielding was conducted on October 31 through November 1 between 6:00pm-8:00pm EST.

Sample: Total Calls Answered = 500; (five Questions asked, bases vary)
Q. Who do you think will Win (confidence interval = +/-5.2%)
Q. Why do you think this candidate will win (confidence interval = +/-6.58%)

 

Terms of Service

Disclaimer: Anderson Analytics does not endorse any candidate nor make any claims in regard election prediction accuracy. We are delighted to share this information with those who agree to the terms of use below.

Indemnity: You agree to indemnify, defend and hold harmless Anderson Analytics, LLC, its officers, employees, consultants and agents from any and all third party claims, liability, damages and/or costs (including, but not limited to, attorneys fees) arising from your use of our services.

Intellectual Property: Text mining methodology used here is intellectual property of Anderson Analytics, LLC. You acknowledge and agree that this website contains proprietary and confidential information that is protected by applicable intellectual property and other laws.

© 2006 Anderson Analytics, LLC