A School Is Born: Correlates of Recent Births of Independent Schools in Ontario

ABSTRACT

Ontario’s population of independent schools does not receive any public subsidies while having to compete with a public system that is well-funded, offers a relative abundance of choice, and performs well from an international perspective. Yet that population has grown steadily despite encountering those unfavorable conditions. To understand this growth, this paper examines correlates of independent school openings between 2011–12 and 2020–21. It draws on organizational ecology frameworks that are typically used to study for-profit businesses but are rarely used to examine schools. Data come from official school registries, coding school websites, and merging census information, and are used to operationalize 3 key concepts: carrying capacity, resource partitioning, and density dependence. Descriptive analyses, logistic regression models, and sensitivity checks suggest the following: new independent schools tend to emerge in International and “Third” sectors, at the secondary level, near other independent schools, and offer additional educational services. We interpret these findings as illustrating that internationalization is extending capacities for independent schools and is helping to enhance their organizational legitimacy, and that school characteristics are likely products of both period effects and organizational changes. Policy implications are discussed.

KEYWORDS:

• Canada
• independent schools
• internationalization
• organizational ecology

Acknowledgement

The authors would like to thank Dan Hamlin, Deani Van Pelt, and Derek Allison for their comments and advice. They would also like to thank the research team for their assistance in data collection: Stephen Reich, Xuefan Li, Sana Abuleil, and Betsy Hu. Earlier versions of this paper were presented to the DEPE Speakers Series at the University of Toronto in March 2023, and to the International School Choice and Reform Conference, Ft. Lauderdale, in January 2023.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. We use the term “independent” for schools operating in Ontario that do not receive funds from the provincial Ministry of Education and are subject to far weaker regulations than public schools.

2. Independent schools’ share of enrollments in Ontario rose from 4% in the 1980s to 6.9% in 2020 (https://www150.statcan.gc.ca/n1/daily-quotidien/211014/t001c-eng.htm; Statistics Canada, 2022b). In 2020, 21.3% of all provincial schools were independent (https://www.ontario.ca/page/facts-about-elementary-and-secondary-education; https://data.ontario.ca/en/dataset/private-school-enrolment-by-gender/resource/00320d1f-9896-4f75-94ad-b76a6048784f). This growing share of provincial enrollments came in a context of considerable overall population growth over 50 years (from 10 million to 15 million) that was accompanied by shrinking proportion of the population that was school aged (from 28.5% aged 0–14 years in 1972 to 15.7% in 2021).

3. Registries do not suggest that the pandemic greatly impacted the population of independent schools. Counts just prior to (2018–19), during (2019–20 and 2020–21), and after the pandemic (2021–22 and 2022–23) reveal that net growth continued during and after the pandemic, following a 30-year pattern (see Figure 1). For instance, between September 2020 and June 2023, 143 schools closed while 270 new schools opened.

4. A small number of schools lacked websites; most were affiliated with Amish and Mennonite religious communities. An expert familiar with those schools helped us collect information on them. For other schools without websites, we searched other sources, including Facebook pages and school association websites like Our Kids.

5. Since this project had multiple coders, we ensured consistency by devising and sharing a codebook, conducting training sessions, and holding ongoing discussions of coding decisions.

6. The number of public schools may be a misleading indicator of the numbers of school-aged children in areas if boards elect to retain schools with declining numbers or accommodate rising numbers of children by adding portables to their schools.

7. We also ran t-tests and logistic regressions that examined only schools in the Greater Toronto Area. We did so to test whether previous results were driven largely by social forces within the GTA or were robust across the province. We found similar directions and patterns of significant coefficients in both sets of models. Tables are available upon request.

8. Substituting the designation of a “new” school opening from the previous 10 years to the previous 5 years shifted the proportions of cases in the “1” category of the outcome variable from 50.8% to 30.1%, leaving 276 fewer schools in that category. As a result, there were zero “new schools” in the Elite sector. That finding is consistent with our characterization of the Elite sector as the most stable in Ontario’s independent school market, but created problems for our logistic estimates. Logit models become unstable if some combinations of categorical predictors and the outcome have empty cells – a problem of perfect prediction in which the outcome does not vary across categories. Perfect prediction causes maximum likelihood estimates to tend toward infinity and thus become inestimable. Stata’s “logistic” and “logit” procedures automatically remove any category that does not vary across the outcome and thus cannot estimate the maximum likelihood for that category. Since there were no 5-year births among Elite schools, Stata automatically removed the Elite category from those logistic models. However, we retained Elite schools in these logits using Stata’s user-written “Firthlogit” program to fit models containing the sector variable. Firth logistic models implement “penalized maximum likelihood logistic regression” which allows them to converge to finite estimates by “penalizing” the log-likelihood, a process similar to “exact” logit estimations of rare events and/or small samples. Coefficients in columns 2 and 4 of Table 5 were generated using firth logistic regression.

9. Observers may interpret this internationalization as having been driven largely by Chinese immigrants. American research has examined the phenomenon of ethnic-based private schools driving growth of independent schools in some jurisdictions (e.g., Maranto et al., 2014). However, the number of Chinese-oriented schools is surprisingly small in Ontario, even among newly opened schools and among those in the Greater Toronto Area (GTA) only. Province-wide, 65 of 1423 schools (4.6%) were exclusively affiliated with Chinese ethnic populations. Among the 447 schools that had opened during the previous 5 years, 9.2% were Chinese-affiliated. Among schools that had opened over the previous 10 years, 6.6% were Chinese-affiliated province-wide; the corresponding figures for the GTA only were 9.5% (openings within 10 years) and 12.4% (openings within 5 years). Thus, while Chinese-affiliated schools represent a dynamic and growing niche, their numbers are too small to be major drivers of the entire independent school sector, province-wide or in the GTA only.

Further, federal data sources suggest that the share of all international students in Canada across all educational levels (K-12 and postsecondary) from China fell from 28% in 2017 to 17% in 2021 (Canadian Bureau for International Education, 2018; Wang & Myers, 2022). Similarly, at the K-12 level, the number of Chinese K-12 international students in Canada fell from 12,640 in 2016 to 6,472 in 2021 (Apply Board, 2023), nearly a 50% reduction. However, during the same time period, our data show that the number of international independent schools in Ontario continued to grow. Thus, we conclude that while Chinese-oriented schools are well-represented among Ontario’s independent schools, particularly among newly opened ones in the Greater Toronto Area, and Chinese students represent a sizable proportion of Canada’s international students, they do not appear to be that market’s major driver of growth. Rather, that market is a patchwork with multiple niches whose continual growth has been driven by an array of social forces. Chinese immigrants occupy a notable but relatively small niche within that broader ecology.