Essays in the Application of Quasi-Experimental Methods to Linked Administrative Data
Bibliographic record
Abstract
Each of the three papers in this thesis applies quasi-experimental methods to administrative data in order to make causal inferences about the effects of a particular event or policy. Paper 1: Does citizenship improve migrant parents’ integration and children’s outcomes?: Evidence from a natural experiment Does obtaining citizenship improve the integration of immigrants into the host country’s society? Does it improve the outcomes of immigrants’ children? Attempts to answer these questions are thwarted by potential self-selection: those who are more motivated to integrate are also more likely to naturalise. In order to make causal inferences, this paper exploits a natural experiment of New Zealand’s removal of birthright citizenship. It finds that this policy change had no effect on family outmigration behaviour, parents’ fertility and labour market outcomes, nor on children’s health outcomes. These results contrast to existing evidence on Germany’s introduction of birthright citizenship, which was found to have positive effects on parents’ and children’s integration outcomes. These different findings may reflect differences in the two countries’ broader immigration policy settings, and the resulting differences in the characteristics of the countries’ migrants. These differences also highlight that the New Zealand case is of potentially more relevance to Canada and the US, the only two western countries that retain unrestricted birthright citizenship, given greater similarities in these countries’ immigration policy settings. Paper 2: Workforce vaccine mandates: The effect on vaccine uptake and healthcare workers’ labour market outcomes As part of its COVID-19 policy response, the New Zealand government implemented vaccination mandates as a condition of ongoing employment for certain workers. This paper examines the effect of these mandates on vaccination uptake among mandated healthcare, education and corrections workers and on healthcare workers’ labour market outcomes. This is enabled by New Zealand’s linked population-wide administrative data, which includes a comprehensive national vaccination register linked to tax records to identify employment outcomes. Overall, the results suggest that in the context of already-high vaccination rates, workforce vaccine mandates provided limited benefit in terms of increasing vaccination rates among mandated workers. Moreover, they negatively impacted healthcare workers’ labour market outcomes, which may have had wider consequences in terms of exacerbating existing health workforce skills shortages. Paper 3: The effect of a minor health shock on labour market outcomes: The case of concussions The literature on health shocks finds that minor injuries have only short-term labour market impacts. However, mild traumatic brain injuries (mTBIs, commonly referred to as concussions) may be different as the medical literature highlights that they can have longer-term health and cognitive effects. Moreover, TBIs are one of the most common causes of disability globally, with the vast majority being mild. Thus, it is important to understand the impact of mTBIs on labour market outcomes. We use administrative data on all medically-diagnosed mTBIs in New Zealand linked to monthly tax records to examine the labour market effects of a mTBI. We use a comparison group of those who suffer a mTBI at a later date to overcome potential endogeneity issues, and employ a doubly-robust difference-in-differences method. We find that suffering a mTBI has negative labour market effects. Rather than dissipating over time, these negative effects grow, representing a decrease in employment of 20 percentage points and earning losses of about a third after 48 months. Our results highlight the need for timely diagnosis and treatment to mitigate the effect of mTBIs to reduce economic and social costs.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.519 | 0.711 |
| Meta-epidemiology (narrow) | 0.002 | 0.003 |
| Meta-epidemiology (broad) | 0.004 | 0.005 |
| Bibliometrics | 0.004 | 0.006 |
| Science and technology studies | 0.006 | 0.044 |
| Scholarly communication | 0.012 | 0.017 |
| Open science | 0.007 | 0.010 |
| Research integrity | 0.012 | 0.018 |
| Insufficient payload (model declined to judge) | 0.014 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".