**University of British Columbia**

Faculty of Commerce and Business Administration

# British Columbia 2001

Provincial Election Forecaster

using the new electoral district boundaries

*developed by Prof. Werner Antweiler*

UBC-ESM Home Page |
UBC-ESM Election Forecasting

### Voter Migration Matrix

You must guess the probability with which a 1996
voter will vote for a party in 2001. The sum of
each row must equal one. Once you have completed
your guess of the voter migration matrix, press
the FORECAST button. The forecaster program will
apply your voter migration matrix to the 1996
election results to find out which party will
win in each riding. Leaving the matrix unchanged
will display the 1996 election results.

**Methodology:**

Because the B.C. election system
is based on "first-past-the-post", prediction of the election
result in terms of the seats distribution in the Legislative
Assembly must be based on a prediction of the election result
in each constituency.
The use of a voter migration matrix reflects the notion that
voters change their opinion about candidates and parties in
a similar way across the entire province. That is, a party that
loses in Vancouver tends to lose in the Okanagan valley and
vice versa. In the simplest form, this "swing" can be applied
across all constituencies.

However, applying this simple voter migration matrix
to the 1996 election results
is a crude way of forecasting the outcome of the 2001 election.
It is crude in three ways. First, it does not reflect the change
in population. Some children have reached voting age, some (mostly older)
people have died, and other people have moved to B.C. or away from
B.C. Second, it does not reflect the possibility that 1996 voters
opt to abstain in 2001, or abstainers in 1996 opt to vote in 2001.
Third, the transition matrix is applied identically across all
constituencies, thus ignoring important local factors.
A more sophisticated approach would take care of these three
problems.

A further problem arises from the redrawing of
the boundaries of electoral districts. There will be 79 seats instead
of 75 as in 1996. This means that some voters
who voted in one district in 1996 will vote in a different district
in 2001. Dr Julian West has kindly
provided a data set which reflects these changes. Thus, the
figures provided for each riding are based on these redrawn
boundaries and not on the original 1996 boundaries. When you
apply the identity voter migration matrix (with all ones in the
diagonal), you will see the result of the 1996 election as if
the 2001 electoral district boundaries had been in effect in 1996. These
numbers will therefore differ from the actual 1996 election outcome in
each riding.

This hypertext page and the election forecasting
software are © 1995/1999 by
Werner Antweiler.