The making of sound decisions about the allocation of resources for provincial parks requires a significant amount of information, in particular attendance data. The purpose of this project is to develop a statistical model to be used for policy analysis and decision making.

In this project, the modeling building process is reviewed and then applied to the development of a logistic regression model to predict day-use park attendance and the effect of policy changes on attendance. The data used consists of characteristics of both the visitor origins and facilities or features of the parks. This spatial interaction model establishes a relationship between the number of day-use visits to parks, the attractiveness of the parks, and the distance between origins and parks. The attractiveness of a park is based on the number of recreation opportunities and facilities offered. The "distance" parameter is most significant as it bears the greatest weight on park attendance forecasting. The model is evaluated in terms of its accuracy, ease of use, and data requirements.

An analysis is made of the spatial re-allocation of a fixed total number of day-use park visits resulting from four policy changes: increasing the attractiveness of a park, decreasing the attractiveness of a park, adding a park, and closing a park. Following the interpretation of the results of the policy analysis, model limitations are assessed and potential model refinements are recommended. Model refinements could include changes to: data sampling, model structure, and market segmentation.