Mario Martinez-Jimenez

Abstract forthcoming.

Abstract forthcoming.

How far does the health-system burden of community violence extend beyond people who are physically injured? We study school shootings as sudden local shocks, linking 2021–2024 American Family Cohort primary-care records to 28 shootings in which at least one person was killed or wounded. A stacked event study compares children living 0–5 miles from the affected school with children living 10–15 miles away. Following a shooting, nearby children were 3.96 percentage points more likely to have a quarter with recorded mental-health or neurodevelopmental care, an increase equal to 37% of the pre-shooting comparison-group mean. The response appears in diagnoses, procedures, and clinical notes concerning suicidal ideation or ADHD, but we do not detect comparably clear changes in medication use or overall primary-care contact. This composition is consistent with trauma entering routine care partly through clinical recognition, assessment, and follow-up rather than prescriptions or additional contact alone. Estimates remain positive when any shooting is omitted, although they are smaller under equal-event weighting and alternative comparison rings. These findings show that casualty counts and prescription records can miss a diffuse outpatient response to local violence and that post-event planning should extend beyond direct victims and the affected school.

Do nearby climate mobilisations weaken young adults' attachment to work and education? We link European Social Survey interviews to the timing and location of five Fridays for Future mobilisations from 2019 to 2023. Our triple-difference design compares four-week near-versus-far changes for adults aged 18–40 with the same changes for adults aged 41–50. The events sharply increased movement-specific search attention. Yet nearby mobilisation did not reduce an index of education, employment, job search, and financial security: the estimate is 0.26 standard deviations (95 percent confidence interval [−0.09, 0.62]). Outcome-specific bounds exclude a 20 percent deterioration in education, active unemployment, and income difficulty, but not in paid work or NEET status. The well-being estimate is negative but imprecise (−0.23 standard deviations; [−0.55, 0.09]) and varies across younger age bins, so it does not support a common youth welfare effect. These estimates capture an age-differential response to regional proximity, not the effect of attending a protest or personally noticing it. The evidence rejects broad short-run economic withdrawal while leaving temporary affective costs unresolved.

Neonatal units care for clinically vulnerable newborns, including infants born premature or at low birthweight. Yet little is known about whether neonatal attendance signals later educational needs. We use data on 8,406 nine-year-olds from Growing Up in Ireland to study reading tests and parent- and teacher-reported difficulties. After adjusting for observed birth and family characteristics and applying the survey weight, children with a history of neonatal care scored about 0.05 standard deviations lower on two reading measures. Neither reading estimate was statistically different from zero, although the confidence intervals also allowed modest disadvantages. Parents reported a 2.8-percentage-point higher rate of learning difficulties; this estimate was also not statistically different from zero, while teachers reported no difference. Caregiver-reported dyspraxia was more common in one follow-up analysis, but direct tests and teacher reports did not show a similar pattern. Adjustments for missing outcomes and plausible reporting error did not change the main interpretation. Neonatal attendance overlapped with, but was not equivalent to, low birthweight or prematurity. The low-birthweight analysis revealed an important design constraint: nearly four in five low-birthweight children attended neonatal care, so the cohort cannot support a separate attendance comparison within this high-risk group. Overall, the findings support using neonatal attendance as a practical early-life marker to help identify children who may benefit from sustained developmental and educational follow-up. The study therefore shows that neonatal attendance can complement birthweight and prematurity when targeting follow-up, while the effects of admission and treatment require designs with stronger clinical data and exogenous variation.

Does schooling make people judge women's and men's employment more equally? This paper combines seven rounds of the European Social Survey with country- and cohort-specific compulsory-schooling reforms and a randomized split-ballot vignette. One additional year of schooling raises rejection of men's priority for scarce jobs by 1.6 percentage points and a two-item employment-equality index by 1.2 points. The job-priority estimate remains significant under country-level wild-bootstrap inference, and every round-specific employment coefficient is positive. Independently, assigning an otherwise identical full-time-work scenario to a mother rather than a father lowers approval by 18.5 percentage points, with penalties of 22.9 points in 2006 and 14.4 points in 2018. The instrumental-variables estimate of schooling's moderation of this penalty is 0.9 percentage points per year (standard error 0.7), so the data do not establish that schooling closes the gap. Schooling promotes more equal stated employment attitudes, while a large maternal-employment penalty remains and is not shown to narrow with education.