Modelling GTHA Post-Secondary School Location Choice

The purpose of this paper is to develop a school location choice model for post-secondary (PS) students in the Greater Toronto-Hamilton Area (GTHA). This analysis differs from previous PS school choice modelling in three respects. Firstly, the model is not representing the college choice process directly. Instead, this analysis is an exercise in matching students who have already made PS school choice decisions to their selected institutions. While there are many areas of overlap, an important difference is that household information reflects where students reside after having selected a college, and possibly, moving out from their parental homes. Secondly, this study primarily analyzes geographical patterns in school location choice for applications in travel demand modelling. An emphasis is placed on modelling the accessibility of each school location to each student, rather than predicting school selectivity or institution type. Thirdly, an RF classifier is implemented for the location choice problem, a novel approach in the field, and its utility is compared to that of the classic econometric approaches. Section 2 presents a brief literature review of relevant works in PS school location choice modelling in general, and in the GTHA specifically. Section 3 introduces the two modelling methods used in this study: random utility models and random forest models. Section 4 describes the two datasets used: the 2015 and 2019 StudentMoveTO (SMTO) surveys. Section 5 presents a logit mode choice model for the 2015 dataset, and Section 6 then presents the development of a school location choice for this dataset. Section 7 presents the development of a random forest model for the school location choice problem and Section 8 summarizes and discusses the main results for the 2015 modelling. Building on the 2015 analysis, Section 9 describes the development of location choice models for the 2019 dataset, and section 10 summarizing the key findings from this analysis. Finally, Section 11 concludes the paper with a brief discussion of possible directions for future work.

• Record URL:
  https://www.tac-atc.ca/sites/default/files/conf_papers/barone_modellinggthapost-secondaryschoollocationchoice.pdf
• Corporate Authors:

  Transportation Association of Canada

  ,    
• Authors:
  • Baron, E
  • Santos, GM
  • MILLER, EJ
• Conference:
  • Transportation Association of Canada 2021 Conference and Exhibition - Recovery and Resilience: Transportation after COVID-19
• Publication Date: 2021

Language

• English

Media Info

• Media Type: Web
• Pagination: 1 PDF file, 507 KB, 20p.
• Monograph Title: Transportation Association of Canada 2021 Conference and Exhibition - Recovery and Resilience: Transportation after COVID-19

Subject/Index Terms

• TRT Terms: Conferences; Demand; Forecasting; Information processing; Location; Mathematical models; Origin and destination; Regional planning; Schools
• Geographic Terms: Canada
• ITRD Terms: 8018: Canada; 8525: Conference; 8655: Data processing; 285: Demand (econ); 132: Forecast; 9061: Location; 6473: Mathematical model; 687: Origin destination traffic; 359: Regional planning; 101: School
• Subject Areas: Education and Training; Planning and Forecasting; I72: Traffic and Transport Planning;

Filing Info

• Accession Number: 01785328
• Record Type: Publication
• Source Agency: Transportation Association of Canada
• Files: ITRD, TAC
• Created Date: Oct 22 2021 3:20PM