A Statistical Analysis of Transfer and Student Mobility in Ontario: What the University/College Applicant Survey™️ Tells Us

Authors
Roger Pizarro Milian
Rod Missaghian
David Zarifa
Henrique Hon
Reference Number
R2269
Date
Status
Abstract

Foreword 

Rod Missaghian, ONCAT 

Postsecondary transfer research in Ontario – despite making significant strides in recent decades – continues to suffer from a lack of data sources that systematically capture patterns in student mobility. For this reason, ONCAT has been diligently working to find innovative data sources, potential new data-linkages, and other opportunities that allow us to extend our understanding of transfer and student mobility in Ontario.  

In the spring of 2020, Academica Group graciously provided ONCAT with access to one of the richest and largest educational datasets in Canadian postsecondary education (PSE): The University/College Applicant Survey™ (UCAS). This proprietary data source provides impressive coverage of hundreds of data fields capturing postsecondary applicants’ demographic characteristics, educational background and aspirations, usage of various information sources, decision-making, and other relevant topics. The UCAS™ has been conducted annually by Academica since the mid-2000s and has been fine-tuned over the years in consultation with PSE stakeholders to capture emerging topics of interest. During this period, the UCAS™ has been completed by hundreds of thousands of applicants to 100+ Canadian colleges, polytechnics, and universities. To date, the UCAS™ remains one of the most trusted data sources for institutional decision-makers across Canada. 

ONCAT is now releasing a series of briefs and papers that outline the initial statistical analysis of transfer and student mobility in Ontario based on this UCAS™ dataset. The analysis presented in this series was developed by the ONCAT research team in partnership with researchers from across the sector and a cross-sector panel of external reviewers. This work builds on previous ONCAT-funded research (Henderson & McCloy, 2017) that also used UCAS™ data. This series contains an introductory paper followed by three briefs: 

It is our hope that this statistical research will advance transfer research and instigate useful discussions at multiple levels within policy and administrative circles.  

To introduce this set of briefs, Situating the UCAS™ Dataset within the Ontario PSE Data Landscape provides a short overview of the education data landscape in Ontario, and how the UCAS™ helps address some notable data limitations. While ONCAT was developing this overview, McMaster professor of sociology, Dr. Karen Robson published a much more detailed overview of the data landscape in Canada. While our overview focuses on the Ontario data context there are overlapping themes, such as the lack of K-12 to postsecondary longitudinal data linkage infrastructure, as well as strict data sharing protocols making available data (i.e., at the Ministry of Colleges and Universities) difficult to access. 

Brief 1 in this series focuses on regional disparities in transfer intent among Ontario college applicants. It begins with a look at the often-understudied concept of ‘transfer intent’ and how degree aspirations among college applicants vary by region of residence in Ontario. Previous research by (Zarifa, Hango & Pizarro Milian, 2019) shows that individuals in remote Northern regions possess lower levels of educational attainment, while other research has explored how post-secondary enrolment is influenced by both proximity and the number of available institutions in a given geographic area (Hango et al., 2021). We explore regional variation for potential vertical transfer applicants (college to university transfer) and develop models to assess the effects of region as well as other demographic variables on transfer ‘intent,’ using the proxy of degree aspirations. The beauty of the UCAS is that it contains applicant variables that can help researchers measure ‘intent’ towards degrees, which are often missing in administrative datasets. However, more Ontario based qualitative research is required to understand the nuances behind postsecondary decision-making processes over time (Missaghian, 2020). ONCAT is currently developing our own in-house transfer intent survey and interviews with students who are interested in transferring credits, which we hope to build into a longitudinal project. 

Brief 2 in the series shifts the emphasis away from transfer ‘intent’ and looks at the effect of socio-economic status (SES) on pathways into college. Using the proxies of parental education and income to operationalize socio-economic status we ask: Are parental education or household income associated with the pathways students take into college? This is an important question to ask as researchers in Canada have often connected student pathways to socio-economic status (Robson et al. 2019; Walters et al. 2021). It is important to note that we are limited to applicant information with the UCAS™ and acknowledge that inequalities can play out differently when looking at enrolment versus applicant data. We find that all other things being equal – knowing an applicant’s SES background is not a particularly useful piece of information when trying to predict what pathway they are taking into the college sector, even though our initial models showed a small but significant positive relationship between student socio-economic status and the probability of traveling direct entry pathways.  

Brief 3 explores another integral variable in assessing student pathways, and that is prior student educational performance, or grades. This is an often-missing piece of data in administrative datasets that do not directly link with high school records, although ONCAT has made strides in developing K-12 with PSE data linkage projects. In this brief we assess the influence of self-reported high school grades on pathways into university and find that as grades increase, the likelihood of college to university pathways decreases. The opposite is true with direct entry and lateral pathways into university, thus establishing the positive effect of higher grades. It is important to note however, that the distribution of potential student pathways suggests that students with higher and lower grades travel all of the various pathways into university.  

Future briefs in this series will include an exploration of Ontario college applicants first-choice institutions, an examination of self-reported disability status and transfer pathways, and how region of origin influences transfer pathways.  

We would like to acknowledge the individuals who donated their time and expertise to help produce and revise the exploratory analyses presented through this series. We would also like to acknowledge Academica Group for granting ONCAT access to the UCAS™ data, which enabled us to perform the statistical analyses that inform the briefs in this series. In doing so, they have performed a great service towards the advancement of transfer research in Ontario.