AEFP 45th Annual Conference

Toward a Meaningful Impact through Research, Policy & Practice

March 19-21, 2020

A Rising Tide that Lifted All Boats? The Effects of Competition on Early Childhood Education Quality

Presenter: 
Qing Zhang, University of California, Irvine, qingz8@uci.edu

The public demand and funding for early childhood education (ECE) have increased substantially in the past few decades, considerably expanding both the public and private sectors in the child care market (Bassok, Fitzpatrick, & Loeb, 2014; Bassok, Miller, & Galdo, 2016). While the influx of suppliers provides diverse choices for children and families, it also increases the level of competition in the market and brings about wide variations in program quality (Bassok, 2012; Bassok et al., 2014). Understanding how competition might affect the quality of providers is particularly important in the context of ECE, where the delivery of high-quality ECE relies on a mixed-delivery system. Because the ECE market involves myriad, relatively small, third-party providers, it is typically more competitive than the K-12 market. Therefore, ECE providers might be more sensitive than K-12 schools to competition and may be more likely to respond to market-based policies that utilize competition (such as providing clear information of quality to the market) to improve quality and child outcomes (Bassok, Dee, & Latham, 2019).
Theoretically, competition can serve as an incentive for schools to improve quality and productivity because they are forced to compete with each other for students (Friedman, 1962; Manski, 1992; Hoxby, 2003; MacLeod & Urquiola, 2012). However, to date, no study has directly examined the relationship between competition and program quality in the context of ECE. Despite the large literature on the competitive effects of the K-12 school choice programs on student outcomes, it is not clear to what extent the estimated competitive effects can be attributed to school quality improvement. This is because isolating the effects of competition on school quality is notoriously challenging. When using student outcomes as a proxy for school quality improvements, the estimated effects of competition can be confounded with other effects such as peer effects and compositional changes in the student body, which simultaneously affect the student outcomes. As a result, most studies focus on the net (or aggregate) effects of competition on student outcomes (i.e., the effects of school quality improvement combined with the effects of compositional changes and peer effects) (Urquiola, 2016). Henry and Gordon (2006) are the only scholars who have examined the effects of competition in the ECE market on overall student performance in grade 3. However, they also focus on the net effects of competition, with the inability to differentiate the effects of competition on program quality from other simultaneous effects on student outcomes.
Using a cross section of the North Carolina Quality Rating and Improvement System (NC QRIS) data in 2019, the current study will provide the first estimation of the effects of competition on program quality in ECE. North Carolina has one of the oldest and most mature QRIS in the country, thus providing a compelling context to study the effects of competition on program quality (measured by the quality ratings in QRIS). The goal of this study is to provide a rich description of the relationship between the level of competition in a local ECE market (defined as a school district) and the quality of the ECE providers in that district. Because quality ratings are a direct measure of program quality, they are arguably less noisy than student outcomes and also free from the confounding effects of compositional changes and peer effects. We will thus use OLS regressions and control for an extensive set of school district characteristics. We use four measures of competition to capture different dimensions of competition (i.e., Herfindahl index, diversity, density, and distance). Because the NC data include the detailed quality indicators used to determine the final quality rating, we will be able to examine the mechanisms by which competition affects quality. Specifically, we will test the following hypotheses:
Hypothesis 1: A higher level of competition in a school district is associated with a higher average quality of ECE programs in the district.
Hypothesis 2: Providers in highly competitive school districts are more likely to have better structural quality (i.e., staff qualification, sanitation, staff-child ratio, and violations) and process quality (i.e., classroom interaction).
Together, this study will provide a starting point for future work to understand the association between competition and program quality. The exploration of the mechanisms will provide hypothesis-generating evidence on the strategies programs might adopt in response to competition.

Poster: 

Comments

This is a really interesting poster! I'm curious about some of the measurement decisions and robustness--for instance, how many centers are unrated? Do we know anything about them in terms of composition of students, etc.? Are your findings sensitive to excluding these providers? I know you said the diversity piece is also TBD--I wish the session were in person, because I'd love to hear more about what types of diversity you're thinking about there. For instance, are you thinking about programs with particular academic focuses (like Montessori or Waldorf)? Designations around who runs the programs (e.g., Head Start centers vs. private vs. public pre-K)? I look forward to seeing future iterations and extensions of this work, possibly at future in-person AEFPs?

I agree with Cassie -this is super interesting and we need more research on this topic. A small detail point that you would have been able to explain in person: what are you graphing, precisely, in your figures? Are those the coefficients on the relationship between competition and quality rating or outcome? More substantively, how should we think of these measures of competition relative to each other? Why would we see different relationships? What are the mechanisms that might lead to different associations between, say, a distance measure and a density measure of competition? And how might you push the diversity of provider question further? i.e., can you classify competition as between similar kinds of providers? How about the customers they serve? Thanks for this great poster! I hope to see the next iteration and next year's AEFP!

I agree that longitudinal data will help, both because you can explore using district-specific effects and because you can use lagged competition measures. I would also think about incorporating American Community Survey data into your analysis to get a richer description of the environment in which these providers operate. Adding those additional controls will reduce the possibility of omitted variable bias. Finally, take a glance at some of the work on supply of private and charter schools. Those papers can help you think about how to measure competition and how to account for diversity.

Thank you for sharing your poster. My general takeaway is that competition doesn’t appear to be consistent associated with higher levels of quality – if anything, the effect looks to be negative as the “distance to the nearest competitor” measure has a positive effect (i.e., negative association of competitiveness) with most measures of outcomes. I wonder if it would makes sense to combine the competitiveness measures using principal component analysis and perhaps do the same with the outcome measures, so as to capture the overall relation between competition and outcomes. marklong@uw.edu

Add new comment