Research

Scholars have increasingly recognized how marginalized communities experience government through criminal legal system encounters, potentially shaping political engagement. A parallel state institution, Child Protective Services (CPS), intervenes with millions of U.S. families annually. The CPS case provides an opportunity to assess theoretical expectations about how punitive state systems affect political life. Analyzing administrative and voting data from a large Pennsylvania system, we first present descriptive estimates of adult voting in families with CPS contact, highlighting how low participation in this large fraction of the population matters for the electorate and overall participation. Then, using the nearly-random assignment of CPS caseworkers, we estimate the causal effect of having a case opened or a child removed on registration and voting. We do not find evidence of substantial participatory effects from these types of CPS contact. We consider interpretations of these null findings and suggest additional research on this vast, understudied government system.

Background: Studies of war-affected youth have not yet examined how trauma memories relate to prospective changes in mental health and to subjective or social experiences. Methods: We interviewed a gender-balanced, randomly selected sample of Afghan child-caregiver dyads (n = 331, two waves, 1 year apart). We assessed lifetime trauma with a Traumatic Event Checklist, past-year events with a checklist of risk and protective events, and several child mental health outcomes including posttraumatic distress (Child Revised Impact of Events Scale, CRIES) and depression. We examined the consistency of trauma recall over time, identified mental health trajectories with latent transition modeling, and assessed the predictors of posttraumatic distress and depression trajectories with multinomial logistic regressions. Results: From baseline to follow-up, reports of lifetime trauma significantly changed (p ≤ 0.01). A third of the cohort reported no trauma exposure; only 10% identified the same event as their most distressing experience. We identified four CRIES trajectories: low or no distress (52%), rising distress (15%), declining distress (21%), and sustained high distress (12%). Youth with chronic posttraumatic distress were more likely to be girls (OR = 5.78, p ≤ 0.01), report more trauma exposure at baseline (OR = 1.55, p ≤ 0.05) and follow-up (OR = 5.96, p ≤ 0.01), and experience ongoing domestic violence (OR = 4.84, p ≤ 0.01). The risks of rising distress and sustained distress showed a steady increase for youth recalling up to four traumatic experiences. Depression and CRIES trajectories showed weak comorbidity. Conclusions: Memories of violent events are malleable, embedded in social experiences, and present heterogeneous associations with posttraumatic distress. Our study provides insights on resilience and vulnerability to multiple adverse childhood experiences, highlighting research and clinical implications for understanding trauma in conflict-affected youth.

Background: In humanitarian settings, family-level drivers of mental health are insufficiently documented; we examined the strength of caregiver–child associations with two-wave, family-level Afghan data. Methods: We recruited a gender-balanced sample of 681 caregiver–child dyads (n = 1,362 respondents) using stratified random-sampling in government schools in Kabul (364 dyads) and refugee schools in Peshawar (317 dyads). One year after baseline, we re-interviewed 64% of Kabul and 31% of Peshawar cohorts (n = 331 dyads, 662 respondents), retaining fewer Peshawar families due to refugee repatriation. In multivariable analyses adjusted for baseline, we assessed the extent to which caregiver mental health (Self-Report Questionnaire, SRQ-20) was associated with child symptom scores of post-traumatic stress (Child Revised Impact of Events Scale, CRIES), depression (Depression Self-Rating Scale, DSRS), psychiatric difficulties, impact, and prosocial strength (Strength and Difficulties Questionnaire, SDQ). Results: Caregiver mental health was prospectively associated with all eight measures of child mental health at follow-up, adjusted for baseline. For post-traumatic stress, caregiver mental health had a predictive impact comparable to the child experiencing one or two lifetime trauma events. For depression, caregiver mental health approached the predictive impact of female gender. Thus a one SD change in caregiver SRQ-20 was associated with a 1.04 point change on CRIES and a 0.65 point change in DSRS. For multi-informant SDQ data, caregiver–child associations were strongest for caregiver ratings. For child-rated outcomes, associations were moderated by maternal literacy, a marker of family-level dynamics. Both adults and children identified domestic violence and quality of home life as independent risk and protective factors. Conclusions: In the context of violence and displacement, efforts to improve child mental health require a thoughtful consideration of the mental health cascade across generations and the cluster of adversities that impact family wellbeing. We identify culturally meaningful leverage points for building family-level resilience, relevant to the prevention and intervention agenda in global mental health.

Child protective services investigate a third of child-homes for maltreatment, yet repercussions for parents remain unclear. I show that parents face considerable adversity when referred for alleged child maltreatment. Using a novel strategy that combines event studies with variation in intervention tendencies across investigators, I find that child protective services interventions mostly affect mothers. Mothers increase enrollment in mental health and substance abuse treatment services for a year, even when not mandated to do so. Earnings and employment are unaffected, but social services support decrease. Long-run effects differ by whether the intervention involves a child removal. Removals increase repeat child welfare involvement and short-run incarceration for parents.

Algorithm tools have the potential to improve public service efficiency, but our understanding of how experts use algorithms is limited, and concerns about resulting bias are widespread. We randomize access to algorithm support for workers allocating Child Protective Services (CPS) investigations. Access to the algorithm reduced child harm, especially for disadvantaged groups, and reduced oversurveillance of Black children by CPS. Workers better processed complementary information that the algorithm made salient, and targeted investigations to children at greater risk of harm irrespective of algorithm-predicted risk. Algorithm-only counterfactuals confirm that humans and algorithms are complementary for both efficiency and equity.

Quasi-experimental designs have transformed the estimation of key economic parameters at play. Yet, studies and designs that were valid to identify effects in a first generation, are typically inappropriate for studying effects on the second generation. Estimating effects on the children of individuals affected by a policy requires estimating social returns as opposed to private returns. We provide an illustration of how current estimation strategies can be problematic, and propose a solution for how this problem can be addressed. 

60% of Americans eligible for unemployment benefits are not receiving them. Amid widespread low and differential take-up rates of social programs (Ko & Moffitt, 2022) and lack of new solutions to address this (DellaVigna & Linos, 2022; Linos et al., 2022), we design a tailored generative AI assistant to overcome the barriers workers face in applying for benefits. We assess its effectiveness in a randomized controlled trial.  

Three out of every thousand workers experienced a serious accident at work in Sweden (Afa Försäkring’s report, 2023). Despite seminal work by Daniel Sullivan and Till Von Wachter (2009) on the effects of unemployment on health, we know surprisingly little of how firms contribute to health and health disparities. Using decades of employer-employee data matched to health outcomes in Swedish register data, we estimate the contribution of firms to worker health. Our methodological approach takes into account the fact that health is dynamic and effects may only reveal themselves over time including after the end of the employment spell, as well as the fact that workers may decide to move firms when their current firm has detrimental effects on their health.

Despite prominent disparities in child mortality, the child protection system’s impact on child death disparities has received limited attention. Leveraging close to random assignments of workers to families in an American county, this paper provides the first causal estimates of the child protection system on child deaths reported to the medical examiner. I find that opening a child welfare case drastically reduces such child deaths. Extrapolating from these local average treatment effects, the black-white child mortality gap would be two to three times larger without child welfare involvement. Child welfare plays a key role in addressing mortality inequality among children.