A Regression Discontinuity Design Framework for Controlling Selection Bias in Evaluations of Differential Item Functioning

Differential item functioning (DIF) is often used to examine validity evidence of alternate form test accommodations. Unfortunately, traditional approaches for evaluating DIF are prone to selection bias. This article proposes a novel DIF framework that capitalizes on regression discontinuity design analysis to control for selection bias. A simulation study was performed to compare the new framework with traditional logistic regression, with respect to Type I error and power rates of the uniform DIF test statistics and bias and root mean square error of the corresponding effect size estimators. The new framework better controlled the Type I error rate and demonstrated minimal bias but suffered from low power and lack of precision. Implications for practice are discussed.

Labor Market Returns to Vocational Secondary Education

We study labor market returns to vocational versus general secondary education using a regression discontinuity design created by the centralized admissions process in Finland. Admission to the vocational track increases initial annual income, and this benefit persists at least through the mid-thirties, and present discount value calculations suggest that it is unlikely that life cycle returns will turn negative through retirement. Moreover, admission to the vocational track does not increase the likelihood of working in jobs at risk of replacement by automation or offshoring. Consistent with comparative advantage, we observe larger returns for people who express a preference for vocational education. (JEL D15, I21, I26, J24, J31, O33)

The Origins of Common Identity: Evidence from Alsace-Lorraine

We study how more negative historical exposure to the actions of nation-states—like war, occupation, and repression—affects the formation of regional identity. The quasi-exogenous division of the French regions Alsace and Lorraine allows us to implement a geographical regression discontinuity design at the municipal level. Using measures of stated and revealed preferences, we find that more negative experiences with nation-states are associated with a stronger regional identity in the short, medium, and long run. This is linked to preferences for more regional decision-making. Establishing regional organizations seems to be a key mechanism to maintaining and strengthening regional identity. (JEL H77, N43, N44, N93, N94, Z13)

Impact of the Economic Impact Payments on Consumer Spending:\\ Analysis on a Granular Level of American Counties

This paper draws on weekly consumer spending data collected by American private companies to analyze the impact of the Economic Impact Payments on consumer spending in American counties. We use regression discontinuity design to quantify the causal effect of the Stimulus Checks on spending and use heterogeneity in economic and demographic factors to determine which groups of counties increased their spending the most, to see what factors affected the Stimulus Checks’ effectiveness. We then use the observed difference in impact across groups of counties to discuss whether the Stimulus Checks were the optimal governmental policy in the crisis and discuss the effectiveness of one-time transfer payments in future recessions.

How Sticky Is Retirement Behavior in the U.S.?

We study how increases in the U.S. Social Security full retirement age (FRA) affect benefit claiming behavior and retirement behavior separately. Using long panels of Social Security administrative data, we implement complementary research designs of a traditional cohort analysis and a regression-discontinuity design. We find that while claiming ages strongly and immediately shift in response to increases in the FRA, retirement ages exhibit persistent “stickiness” at the old FRA of 65. We use several strategies to explore the likely mechanisms behind the stickiness in retirement and find suggestive evidence that employers play a role in workers' responses to the FRA.

Do Academically Struggling Students Benefit From Continued Student Loan Access? Evidence From University and Beyond

We estimate the effects of student loan access on educational attainment and labor market returns in New Zealand. We exploit the introduction of a national policy mandating a 50% pass rate for student loan renewals using a regression discontinuity design. Retaining loan access increases re-enrollment for students around the threshold, and a majority eventually graduate with a bachelor's degree within seven years. We find that retaining student loan access leads to large labor market returns for struggling students. The additional debt from further borrowing is both small relative to the earnings returns and declines quickly due to faster repayment.

Commodity Shocks and Incumbency Effects

Why do incumbents enjoy an electoral advantage in some political settings but suffer from a disadvantage in others? We propose a novel explanation linking variation in incumbency effects with exogenous commodity shocks. While voters attempt to sanction incumbents for economic performance, changes in commodity prices affect their evaluations and condition the electoral fortunes of incumbents vis-à-vis challengers. We test our argument in Brazilian municipalities, combining a plausibly exogenous measure of variation in commodity prices with a close election regression discontinuity design. Our results show that increases in the price of agricultural commodities greatly enhance the prospects of incumbents, while negative shocks exacerbate their incumbency disadvantage, especially in rural municipalities. Further investigation suggests that commodity shocks do not operate via voter learning about candidate quality, changes in the pool of candidates, shifts in voter preferences, or strategic elite investments. Instead, we find suggestive evidence that commodity shocks affect voters' evaluations through their effect on local economic growth.