Wise men have always understood the power that could come from
having critical information in the right place at the right time.
The old Nursery Rhyme, For Want of a Nail, expresses our
understanding of the importance of the smallest detail in the
critical information process.
The ancient Druids who built Stonehenge understood the fine details
of predicting eclipses, and the power that came from that knowledge.
We didn.t understand what they were doing until Gerald S Hawkins
used a computer to analyze the ancient heavens and the Stonehenge
geometry. Their mastery of the details is staggering.
By the early 1800.s the Rothschild fortunes were already
considerable, and their international intelligence organization was
unmatched. Napoleon was sweeping across Europe and British war
bonds were dropping. Nathan Rothschild realized the stakes and had
his agents ready . he knew the outcome at Waterloo several hours
before anyone else in London. He started shorting British bonds,
driving them to 5% of their face value, then he quietly started
buying every bond available. When word of Wellington.s victory
reached London the bonds surged back to more than face value.
Rothschild.s already vast fortune was multiplied twenty times over
in a few hours.
But wise men can forget, especially when they are on a roll. In
1942, as their invasion fleet sailed towards Midway, the Japanese
Admirals decided they were doing so well they could call their
search planes back early, saving the gas. If they had searched the
full pattern they would have seen the three American carriers
hiding behind Midway and the war would have gone quite differently.
When the Americans dropped out of the clouds the Japanese flight
decks were covered with refueling planes, and the world was changed
forever. For lack of a nail.
Today there are many wise men involved in political campaigns, and
they understand that the critical information is provided by
polling. The very wise ones also understand that the nail that a
great poll rests on is the sample of voters to contact.
Be sure your great poll rests on the best sample available, the one from VCS.
In the political wars, polls play a pivotal role. The polls depend
on a good sample. The sample is the nail on which rests the poll,
and perhaps the election.
BEST VOTER FILE
VCS has the most complete and reliable national database of
registered voters in existence. The file is constantly refreshed
with new data; all data is standardized; addresses are vetted by
CAS software. VCS has been collecting these files for years; many
of our files have data which is no longer available through the
election officials; this applies particularly to vote history, but
also in some cases birth dates and registration dates. VCS also
enhances the files with ethnic coding (based on VCS.s proprietary
ethnic dictionary) and with demographic data from the census. VCS
addresses the major problems that polling and voter outreach
campaigns face today: 1) Having good phone numbers for the largest
possible number of voters to maximize your ability to talk with
voters and 2) Having up to date data on who votes by absentee
ballot so that you can reach those who vote absentee as well as
those who go to the polls on election day. VCS captures phone
numbers that voters indicate when they register and supplements
that information quarterly with commercial phone number lists. VCS
files also include data on whether the voter has asked to be a
"permanent absentee" and how frequently and when the voter has
actually voted absentee.
VCS is proud of the quality of our database and the fact that we
have supplied thousands of samples over the years with virtually
no errors. With VCS you can be sure you are calling the right
voters.
FASTEST DELIVERY
Generally the results of polls are needed quickly, and a pollster
is under constant pressure to deliver fast. Many are startled
when we produce their required sample is less than an hour. Even
the more complex samples are generally available in a few hours.
VCS has an extensive set of database counts online, which you can
review immediately in considering your universe; then you can use
the VCS online ordering software, weborder, to order your sample
online. VCS personnel have extensive experience with polling and
polling samples and are always available to help.
ANALYSIS OF YOUR SAMPLE
With the VCS techniques, you will obtain a sample that is
generally representative of your selected voters. Even with all
these efforts, you may wish to do some weighting. In order to
determine weights, it is necessary to understand how your sample
compares to your universe on a number of key characteristics. VCS
has developed the best set of cross tabs available for sample
analysis. Each VCS sample is accompanied by two of these
profiles, one of the sample and one of the universe, for side by
side comparison. These profiles show characteristics (gender,
party, age ranges, etc) cross-tabbed with vote history. Check a
pair out at
SampleProfile
. The profiles can be
modified, so if you are interested in different cross-tabs we can
provide them.
VOTER SELECTIONS
Many times you want to sample only a portion of the voters. VCS
provides extensive selection capability, so if you want to poll
married Democratic Jewish women in the Buffalo Media Market who
have voted in the last two elections -- no problem. Our online
counts give you the size of many voter selections immediate.
SOLID SAMPLING METHODOLOGY
VCS can produce simple nth stratified samples or clustered
samples from our superb voter database. Our voter files are
maintained in geographic sequence, and most samples are extracted
in this sequence. More complex stratifications can be
accomplished by sorting the voter file into a different sequence
before collecting the sample points or by taking multiple samples
based on specific demographics, then merging these samples into a
single sample. Clustered samples can be produced in many
different ways.
YALE TURNOUT MODELS
You can also obtain a sample based on voter turnout models such
as those documented in the work of Dr Donald Green and Dr. Alan
Gerber of Yale University. These samples are more complex and
require that the model be previously created. VCS creates
standard models for some jurisdictions. If you are interested in
obtaining a sample based on these models, some assistance from
VCS personnel may be necessary.
OVERSAMPLING AND COMPLEX CLUSTERING
Sometimes oversampling is best accomplished by taking larger
samples in specific areas that are known to have a population
that is difficult to contact. VCS can do this; the oversamples
can be presented separately or integrated with the larger sample.
More frequently, clustering is employed to control the
oversample. However, due to the distribution of phone numbers in
the database, homogeneity may be compromised. People are
increasingly difficult to contact by phone. There are many
reasons for this including an increasing tendency to unlist phone
numbers and the increased use of cell phones. Acquiring and using
cell numbers is hampered by laws making the use of such numbers
illegal without the permission of the owner. This problem means
that the number of voters with good phones on the file is
decreasing. This can result in a skewed sample if only those with
numbers are considered when skipping through the file, since more
voters will be chosen from stable areas or areas that have
cleaner address structures.
In order to mitigate this problem, VCS can provide .pure phone
clusters.. This means that all voters who meet the other
selection criteria are counted when skipping through the file,
whether they have a phone on file or not. However, when the
clusters are filled, only those with phones are included in the
cluster. This preserves the geographic homogeneity of the sample.
The idea of .pure clusters. can be extended to gender and party
affiliation also. A sample with pure phone, gender and party
clusters insures not only geographic homogeneity, but also
insures certain numbers of men, women, and Democrats,
Republicans, and Others.
