Who Were New Hampshire’s Likely Democratic Primary Voters?

One of the pre-election pollsters’ most daunting tasks is
to divine a model for the “likely voter.”  Since most respondents in a
pre-election telephone survey will report that they intend to vote, pollsters
use a series of questions that help them estimate which members of their
telephone samples are actually likely to show up at the polls, and which
respondents are likely to demur.  Like grandma’s favorite recipes, survey
organizations typically keep their “likely voter model” a closely held secret.

 In the pre-election polls for the recent New Hampshire
primary, the likely voter models for the Republican primary performed well,
though the likely voter model for the Democratic primary performed poorly.  The
pre-election estimates for the Republican candidates were derived from models
that were (presumably) based on the 2000 Republican presidential primary.  For
both Republican primaries, the voter turnout was  remarkably similar (238,606 in
2000 and 238,909 in 2008), so the pre-election polls’ likely voter model worked
well, and their predictions were reasonably
accurate.

 The voter turnout for the Democratic primary, on the
other hand, was unprecedented in at least two ways: the number of participants
and the proportion of registered Democrats who showed up to vote.  In 2004, a
(then) record of 219,787 voters turned out to vote–the previous record for the
Democratic primary was in 1992, when 167, 819 voters participated.  This year, a
record shattering 287,849 voters participated in the New Hampshire Democratic
primary–including nearly two thirds (66.3%) of the state’s registered Democrats
(up from 43.3% in 2004).  Simply stated, the 2008 New Hampshire Democratic
primary had a voter turnout rate that resembled a November presidential
election, not a usual party primary, and the likely voter models for the polling
organizations were focused on a primary–this time, that simply did not work.

Allan McCutcheon