British Columbia 2001
Provincial Election Forecaster
using the new electoral district boundaries

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.