Ron Diris

Don't Hold Back? The Effect of Grade Retention on Student Achievement

This study analyzes the effect of age-based retention on school achievement at different stages of education. I estimate an instrumental variable model, using the predicted probability of retention given month of birth as an instrument, while simultaneously accounting for the effect of month of birth on maturity at the time of testing. The analysis further assesses heterogeneity in retention effects by achievement, by background characteristics, and by type of skill. Using international data from multiple waves of the PISA international assessment test, I find that grade retention in primary school harms student achievement across the distribution, while delayed school entry can produce positive results for those at the lower end. The identified local average treatment effect indicates that letting students retain in primary school because of a low relative age is harmful for their future school achievement.

How does early deprivation relate to later-life outcomes? A longitudinal analysis

Measures of material deprivation are increasingly used as alternatives to traditional poverty indicators. While there exists extensive literature focusing on the impact that growing up in a (financially) poor household has on future success, little is known about how material deprivation relates to long-run outcomes. This study uses longitudinal data from the 1970 British Cohort Study to assess the relationship between material deprivation and outcomes in adult life. We control for an extensive set of observable characteristics, and further employ value-added analysis and generalized sensitivity analysis to assess the nature of this relationship. We find that deprivation relates to a diverse set of outcome variables, but the magnitude of the conditional relationships are generally small. Immaterial indicators of family quality show relatively stronger ties to future outcomes, especially with respect to non-cognitive skills.

An Economic Analysis of the Optimal School Starting Age

This study develops and empirically applies an approach to identify the optimal school starting age. Postponing school entry involves a tradeoff between better learning through higher maturity and higher opportunity costs through foregone time. We model a Skill Technology Function, where learning becomes more effective as children mature, and use variation in month of birth for multiple countries from multiple data sources to empirically estimate the slope and the curvature of this function. To compare student performance at different exact ages, we express test scores in age equivalents. This allows us to separate starting age effects from age of testing effects, and shows that birth month effects stabilize relatively early. Our main conclusion is that higher starting ages improve achievement in both the short and the long run, but, for most conventional starting ages, this effect does not compensate for incurred opportunity costs.

Child poverty: what can social spending explain in Europe?

This study assesses the role of social spending in relation to child poverty in European welfare states. Using macro-level panel data from EU SILC 2005-2012, we analyze the effect of the size of social spending and the effect of how those benefits are targeted. We separately estimate the effect of pension benefits on child poverty, as the prevalence of multigenerational families makes them a relevant income source for families with children, especially in Southern and Eastern European welfare states. Estimating a GLS model including time and country fixed effects, we find that both cash transfers and pensions substantially reduce child poverty. Increased targeting also leads to lower poverty rates, but the effect sizes are more modest by comparison, and strongly depend on how targeting is defined. The estimates for social spending change little across various model specifications and we also obtain similar estimates when we use regional variation within countries to assess the same effects. Where social spending explains a large share of variation in poverty within countries over time, the explanatory power with respect to cross-sectional variation in poverty rates is limited. The complete model does explain a large share of cross-sectional disparities in poverty across European welfare states, but a sizable unexplained variation remains. This unexplained disparity likely relates to factors that are more invariable over time.

Allocating Instruction Time: How Language Instruction Can Affect Multiple Skills

There exists substantial variation in how schools allocate instruction time to school subjects. The effectiveness of that allocation depends on the immediate effect of instruction in one subject on achievement in the same subject, on how skills further develop over time, and on possible spillover effects on achievement in other subjects. Exploiting a policy intervention in Dutch primary education, we find that effects of language instruction on language skills fade away quickly, while effects of (early) language instruction on several other skills are long-lasting. The results illustrate that spillover effects can arise in the context of skill acquisition.

Estimating the Impact of Grade Retention: A Heterogenous Analysis

While the practice of grade retention is very widespread, previous research has consistently provided negative results regarding its impact. In this study, we execute a widespread analysis of the effects of retention, thereby specifically taking into account its endogenous and heterogenous nature. We execute an IV-estimation, using the strong correlation between month of birth and retention as an instrument, to assess the former and conduct a stepwise estimation method to deal with the latter. We thereby make a clear distinction between retention in kindergarten (Late Starts) and retention thereafter (Grade Retention). By looking at test scores at age 15, we find that the practice of Grade Retention is consistently bad for the whole distribution, while Late Starts do produce positive results for those at the lower end.