Question:
SEN1: "Lets turn to the November election for U.S. Senate from New Hampshire ... will you vote for John Sununu the Republican ... Jeanne Shaheen the
Democrat ... someone else ... or haven't you decided yet?" IF UNDECIDED: "Would you say you are leaning toward Republican John Sununu or Democrat
Jeanne Shaheen?"
Sununu Shaheen Other Undecided (N=)
October 24-26 38% 49% 2% 12% 687
October 25-27 36% 50% 2% 13% 641
October 26-28 40% 48% 1% 11% 652
Registered Democrat 14% 81% 0% 6% 192
Registered Undeclared 32% 49% 3% 16% 274
Registered Republican 78% 12% 0% 9% 184
Democrat 8% 83% 1% 8% 291
Independent 35% 38% 3% 23% 111
Republican 80% 8% 1% 10% 230
Core Republican 95% 1% 1% 4% 191
Swing Voter 38% 30% 4% 28% 168
Core Democrat 5% 88% 0% 6% 294
Male 45% 43% 1% 11% 326
Female 34% 53% 2% 12% 326
High school or less 37% 43% 4% 16% 106
Some college 44% 44% 3% 9% 143
College graduate 41% 46% 0% 13% 227
Post-graduate 36% 57% 0% 7% 169
18 to 34 33% 52% 6% 9% 56
35 to 49 46% 40% 2% 12% 194
50 to 64 38% 51% 0% 11% 263
65 and over 39% 50% 1% 11% 120
Less than $30K 20% 64% 4% 12% 36
$30K to $60K 38% 50% 3% 10% 116
$60K to $75K 27% 63% 1% 9% 53
$75K to $100K 41% 51% 3% 5% 85
$100K or more 51% 42% 1% 5% 162
Attend services 1 or more/week 47% 39% 1% 12% 181
1-2 times a month 32% 55% 1% 12% 77
Less often 39% 52% 2% 8% 221
Never 37% 48% 1% 14% 162
10 yrs or less in NH 37% 44% 1% 18% 107
11 to 20 years 43% 44% 2% 12% 111
More than 20 years 40% 49% 2% 10% 426
North Country 51% 42% 2% 6% 68
Central / Lakes 33% 58% 0% 9% 106
Connecticut Valley 27% 62% 1% 11% 88
Mass Border 39% 43% 2% 16% 180
Seacoast 37% 50% 3% 10% 106
Manchester Area 55% 34% 1% 10% 105
First Cong. Dist 46% 41% 2% 11% 334
Second Cong. Dist 33% 55% 1% 11% 318

Poll Demographics

About this Poll
These findings are based on the 2008 WMUR / UNH New Hampshire Tracking Poll, conducted by the University of New
Hampshire Survey Center. This survey is sponsored by WMUR-TV Manchester and the University of New Hampshire.* Six
hundred sixty-one (661) likely New Hampshire voters were interviewed between October 26 and October 28, 2008. The
potential sampling error for the statewide survey is +/-3.8%.