Casey Breen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Casey Breen has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
Dennis Feehan
3 shared publications- Joshua Robert Goldstein
Department of Demography, University of California, Berkeley, CA 94720
3 shared publications - Ridhi Kashyap
Department of Sociology, Nuffield College, and Leverhulme Centre for Demographic Science, University of Oxford, Oxford, UK
3 shared publications Abraham Azar
IMPACT Initiatives , Geneva
2 shared publicationsChristina Kay
IMPACT Initiatives , Goma
2 shared publicationsDennis M. Feehan
2 shared publications
Latest Publications
- Open Access
- 18 November 2026
- Published by Center for Open Science in SocArXiv
Digital Diffusion: Mobile Internet and the Spread of Cultural Scripts about Family Life
Global cultural diffusion theories have long emphasized how institutional actors such as intergovernmental organizations, international non-governmental organizations, and state agencies spread cultural scripts. We theorize that mobile internet constitutes a qualitatively distinct pathway for the diffusion of these scripts. In contrast to traditional mass media diffusion channels, the internet is more continuous, diverse, and individually curated. This reconfiguration changes both the cultural models individuals encounter and the frequency with which individuals are exposed to them. We test this argument in two ways using fertility in Nigeria as an empirical case. First, we exploit the staggered rollout of mobile internet, finding that expanding coverage reduces fertility, corresponding to a 9% relative decline in the annual probability of birth. This decline is accompanied by shifts in contraceptive use, ideal family size, and women’s autonomy in financial and healthcare decision-making. Second, we analyze16,282 posts from fertility-related threads on a major Nigerian social media platform to characterize the cultural scripts about family life circulating online during the study period. The online discussions overwhelmingly endorse liberal scripts of planned fertility and modern contraception. These findings suggest that digital infrastructure can reshape existing diffusion processes with meaningful consequences for family life.
- 18 November 2026
- Published by Center for Open Science in SocArXiv
Open AccessDigital Diffusion: Mobile Internet and the Spread of Cultural Scripts about Family Life
Global cultural diffusion theories have long emphasized how institutional actors such as intergovernmental organizations, international non-governmental organizations, and state agencies spread cultural scripts. We theorize that mobile internet constitutes a qualitatively distinct pathway for the diffusion of these scripts. In contrast to traditional mass media diffusion channels, the internet is more continuous, diverse, and individually curated. This reconfiguration changes both the cultural models individuals encounter and the frequency with which individuals are exposed to them. We test this argument in two ways using fertility in Nigeria as an empirical case. First, we exploit the staggered rollout of mobile internet, finding that expanding coverage reduces fertility, corresponding to a 9% relative decline in the annual probability of birth. This decline is accompanied by shifts in contraceptive use, ideal family size, and women’s autonomy in financial and healthcare decision-making. Second, we analyze16,282 posts from fertility-related threads on a major Nigerian social media platform to characterize the cultural scripts about family life circulating online during the study period. The online discussions overwhelmingly endorse liberal scripts of planned fertility and modern contraception. These findings suggest that digital infrastructure can reshape existing diffusion processes with meaningful consequences for family life.
- Open Access
- 28 August 2026
- Published by Center for Open Science in SocArXiv
Record Linkage Error and Inference in Population Research
The explosion in digitization of large-scale administrative data and advances in record linkage have opened up new possibilities for studying core topics in population research, including social mobility, early-life determinants of longevity, and shifts in ethnoracial identity. Much attention has been paid to improving record linkage algorithms, but methodological guidance for researchers analyzing linked data is limited. We propose a general framework to explain how different types of linkage errors—false matches and missed matches—impact inference with linked data and introduce a bias-correction method. We then conduct three empirical case studies on canonical topics in population research: social mobility, the education–longevity gradient, and shifts in ethnoracial identification. These case studies demonstrate the settings in which linkage errors do and do not meaningfully impact substantive research findings. We conclude with practical recommendations for researchers performing inference with linked data and provide an accompanying R package implementing the methods introduced in this paper.
- 28 August 2026
- Published by Center for Open Science in SocArXiv
Open AccessRecord Linkage Error and Inference in Population Research
The explosion in digitization of large-scale administrative data and advances in record linkage have opened up new possibilities for studying core topics in population research, including social mobility, early-life determinants of longevity, and shifts in ethnoracial identity. Much attention has been paid to improving record linkage algorithms, but methodological guidance for researchers analyzing linked data is limited. We propose a general framework to explain how different types of linkage errors—false matches and missed matches—impact inference with linked data and introduce a bias-correction method. We then conduct three empirical case studies on canonical topics in population research: social mobility, the education–longevity gradient, and shifts in ethnoracial identification. These case studies demonstrate the settings in which linkage errors do and do not meaningfully impact substantive research findings. We conclude with practical recommendations for researchers performing inference with linked data and provide an accompanying R package implementing the methods introduced in this paper.
- Open Access
- 13 April 2026
- Published by Wiley in Population and Development Review
- Vol. 52 (2) , 299-318
Structured Inequality, Uncertain Lifespans: Demographic Perspectives on Predicting Individual‐Level Longevity
There are striking disparities in life expectancy across sociodemographic groups in the United States, shaped by structural forces such as racism, class inequality, and policy environments. To what extent do sociodemographic characteristics structure—or fail to structure—individual lifespans? Using U.S. Census data linked to administrative death records, we assess how well early‐adulthood social, economic, and demographic characteristics predict individual lifespan in a cohort of men born in 1910 and observed through their deaths between 1975 and 2005 ( N = 121,000). Despite large group‐level disparities, we find that sociodemographic characteristics measured in early adulthood explain less than two percent of the overall variation in individual lifespan. These findings reaffirm a central demographic regularity: variance in life expectancy between groups is small compared to variation in lifespan within groups. This highlights the fundamentally nondeterministic nature of how structural inequality shapes individual mortality.
- 13 April 2026
- Published by Wiley in Population and Development Review
- Vol. 52 (2) , 299-318
Open AccessStructured Inequality, Uncertain Lifespans: Demographic Perspectives on Predicting Individual‐Level Longevity
There are striking disparities in life expectancy across sociodemographic groups in the United States, shaped by structural forces such as racism, class inequality, and policy environments. To what extent do sociodemographic characteristics structure—or fail to structure—individual lifespans? Using U.S. Census data linked to administrative death records, we assess how well early‐adulthood social, economic, and demographic characteristics predict individual lifespan in a cohort of men born in 1910 and observed through their deaths between 1975 and 2005 ( N = 121,000). Despite large group‐level disparities, we find that sociodemographic characteristics measured in early adulthood explain less than two percent of the overall variation in individual lifespan. These findings reaffirm a central demographic regularity: variance in life expectancy between groups is small compared to variation in lifespan within groups. This highlights the fundamentally nondeterministic nature of how structural inequality shapes individual mortality.
- Open Access
- 24 March 2026
- Published by Center for Open Science in SocArXiv
Migrant Social Integration Before Arrival: Evidence from 3.9 Billion Facebook Friendships
Social media platforms increasingly enable migrants to build ties and socially integrate in their destination country well before they physically relocate, yet conventional research on migration treats social integration as a post-arrival process. Using de-identified Facebook data from 2.1 million recent international migrants from 163 sending countries to five major destination countries – Canada, Germany, Spain, the United Kingdom, and the United States – we track 3.9 billion friendships formed before and after migration. We identify a distinct six-month migration preparatory phase preceding migration, during which migrants sharply increase the number of new online social ties formed with individuals in the destination country. In this same window, approximately half of migrants joined online community groups oriented toward integration; qualitative analysis shows these groups are largely organized around community-building, employment, and housing. Migrants who joined such groups formed roughly 63% more destination-country Facebook ties than comparable migrants who did not, and a substantial share of these ties later developed into close friendships. These findings show that migrant social integration begins before arrival, with social media platforms functioning as core infrastructure through which social integration itself is organized.
- 24 March 2026
- Published by Center for Open Science in SocArXiv
Open AccessMigrant Social Integration Before Arrival: Evidence from 3.9 Billion Facebook Friendships
Social media platforms increasingly enable migrants to build ties and socially integrate in their destination country well before they physically relocate, yet conventional research on migration treats social integration as a post-arrival process. Using de-identified Facebook data from 2.1 million recent international migrants from 163 sending countries to five major destination countries – Canada, Germany, Spain, the United Kingdom, and the United States – we track 3.9 billion friendships formed before and after migration. We identify a distinct six-month migration preparatory phase preceding migration, during which migrants sharply increase the number of new online social ties formed with individuals in the destination country. In this same window, approximately half of migrants joined online community groups oriented toward integration; qualitative analysis shows these groups are largely organized around community-building, employment, and housing. Migrants who joined such groups formed roughly 63% more destination-country Facebook ties than comparable migrants who did not, and a substantial share of these ties later developed into close friendships. These findings show that migrant social integration begins before arrival, with social media platforms functioning as core infrastructure through which social integration itself is organized.
- Open Access
- 4 March 2026
- Published by Center for Open Science in SocArXiv
Structured Inequality, Uncertain Lifespans: Demographic Perspectives on Predicting Individual-level Longevity
There are striking disparities in life expectancy across sociodemographic groups in the United States, shaped by structural forces such as racism, class inequality, and policy environments. To what extent do sociodemographic characteristics structure – or fail to structure – individual lifespans? Using U.S. Census data linked to administrative death records, we assess how well early-adulthood social, economic, and demographic characteristics predict individual lifespan in a cohort of men born in 1910 and observed through their deaths between 1975 and 2005 (N = 121,000). Despite large group-level disparities, we find that sociodemographic characteristics measured in early adulthood explain less than two percent of the overall variation in individual lifespan. These findings reaffirm a central demographic regularity: variance in life expectancy between groups is small compared to variation in lifespan within groups. This highlights the fundamentally non-deterministic nature of how structural inequality shapes individual mortality.
- 4 March 2026
- Published by Center for Open Science in SocArXiv
Open AccessStructured Inequality, Uncertain Lifespans: Demographic Perspectives on Predicting Individual-level Longevity
There are striking disparities in life expectancy across sociodemographic groups in the United States, shaped by structural forces such as racism, class inequality, and policy environments. To what extent do sociodemographic characteristics structure – or fail to structure – individual lifespans? Using U.S. Census data linked to administrative death records, we assess how well early-adulthood social, economic, and demographic characteristics predict individual lifespan in a cohort of men born in 1910 and observed through their deaths between 1975 and 2005 (N = 121,000). Despite large group-level disparities, we find that sociodemographic characteristics measured in early adulthood explain less than two percent of the overall variation in individual lifespan. These findings reaffirm a central demographic regularity: variance in life expectancy between groups is small compared to variation in lifespan within groups. This highlights the fundamentally non-deterministic nature of how structural inequality shapes individual mortality.
- Open Access
- 16 January 2026
- Published by Center for Open Science in SocArXiv
Unpacking the Hispanic Mortality Paradox: A Research Note on Country-of-Origin Variation
Hispanic immigrants in the United States face persistent socioeconomic disadvantage but paradoxically experience lower mortality throughout the life course. We document this Hispanic mortality paradox across 22 national-origin groups using Social Security administrative mortality records from 1988--2005. The paradox extends to migrants from every country in South America, Central America, and the Hispanic Caribbean. Yet the magnitude of the mortality advantage varies substantially by country of origin, state of arrival, and period of arrival, demonstrating the importance of examining subgroup-specific outcomes rather than treating Hispanics as a monolithic population. Taken together, these findings highlight the need to move beyond the aggregated Hispanic category and recognize how diverse migration histories and incorporation contexts shape the health trajectories of Hispanic immigrants.
- 16 January 2026
- Published by Center for Open Science in SocArXiv
Open AccessUnpacking the Hispanic Mortality Paradox: A Research Note on Country-of-Origin Variation
Hispanic immigrants in the United States face persistent socioeconomic disadvantage but paradoxically experience lower mortality throughout the life course. We document this Hispanic mortality paradox across 22 national-origin groups using Social Security administrative mortality records from 1988--2005. The paradox extends to migrants from every country in South America, Central America, and the Hispanic Caribbean. Yet the magnitude of the mortality advantage varies substantially by country of origin, state of arrival, and period of arrival, demonstrating the importance of examining subgroup-specific outcomes rather than treating Hispanics as a monolithic population. Taken together, these findings highlight the need to move beyond the aggregated Hispanic category and recognize how diverse migration histories and incorporation contexts shape the health trajectories of Hispanic immigrants.
- Open Access
- 9 January 2026
- Published by Center for Open Science in SocArXiv
The Black–White Mortality Crossover: Evidence from Linked U.S. Administrative Mortality Records
First documented in 1930, the Black–White mortality crossover remains one of the most enduring puzzles in Demography. Black Americans experience higher age-specific mortality rates than White Americans throughout most of the life course, yet this pattern reverses at advanced ages. A leading explanation is that there is no crossover at all. Rather, differential age misreporting or other data errors create a spurious crossover. We use a large-scale administrative cohort dataset of American men (N = 2.3 million deaths) to assess whether the Black–White mortality crossover is real or an artifact of data error. We find a mortality crossover for the male birth cohorts of 1890–1905 at age 86. At age 75, Black men have nearly a 10% higher probability of dying than White men, but by age 95, the probability of dying is 8% higher for White men. The quality of our mortality data, paired with a series of sensitivity analyses, provides new evidence that the observed crossover is not an artifact of age misreporting. Our findings address a long-standing empirical debate and emphasize the need for further investigation into alternative explanations of the crossover, including selective survival mechanisms and reversals of group-specific disadvantage at advanced ages.
- 9 January 2026
- Published by Center for Open Science in SocArXiv
Open AccessThe Black–White Mortality Crossover: Evidence from Linked U.S. Administrative Mortality Records
First documented in 1930, the Black–White mortality crossover remains one of the most enduring puzzles in Demography. Black Americans experience higher age-specific mortality rates than White Americans throughout most of the life course, yet this pattern reverses at advanced ages. A leading explanation is that there is no crossover at all. Rather, differential age misreporting or other data errors create a spurious crossover. We use a large-scale administrative cohort dataset of American men (N = 2.3 million deaths) to assess whether the Black–White mortality crossover is real or an artifact of data error. We find a mortality crossover for the male birth cohorts of 1890–1905 at age 86. At age 75, Black men have nearly a 10% higher probability of dying than White men, but by age 95, the probability of dying is 8% higher for White men. The quality of our mortality data, paired with a series of sensitivity analyses, provides new evidence that the observed crossover is not an artifact of age misreporting. Our findings address a long-standing empirical debate and emphasize the need for further investigation into alternative explanations of the crossover, including selective survival mechanisms and reversals of group-specific disadvantage at advanced ages.
- Open Access
- 14 October 2025
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 122 (42)
Mapping subnational gender gaps in internet and mobile adoption using social media data
The digital revolution has ushered in many societal and economic benefits. Yet access to digital technologies such as mobile phones and internet remains highly unequal, especially by gender in the context of low- and middle-income countries (LMICs). While national-level estimates are increasingly available for many countries, reliable, quantitative estimates of digital gender inequalities at the subnational level are lacking. These estimates, however, are essential for monitoring gaps within countries and implementing targeted interventions within the global sustainable development goals, which emphasize the need to close inequalities both between and within countries. We develop estimates of internet and mobile adoption by gender and digital gender gaps at the subnational level for 2,075 regions in 117 LMICs from 2015 through 2025, a context where digital penetration is low and national-level gender gaps disfavoring women are large. We construct these estimates by applying machine-learning algorithms to Facebook user counts, geospatial data, development indicators, and population composition data. We calibrate and assess the performance of these algorithms using ground-truth data from subnationally representative household survey data from 33 LMICs. Our results reveal striking disparities in access to mobile and internet technologies between and within LMICs. These disparities imply that as of 2025, women are 19% less likely to use the internet and 8% less likely to own a mobile phone in LMICs, corresponding to over 190 million fewer women owning a mobile phone and over 320 million fewer women using the internet.
- 14 October 2025
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 122 (42)
Open AccessMapping subnational gender gaps in internet and mobile adoption using social media data
The digital revolution has ushered in many societal and economic benefits. Yet access to digital technologies such as mobile phones and internet remains highly unequal, especially by gender in the context of low- and middle-income countries (LMICs). While national-level estimates are increasingly available for many countries, reliable, quantitative estimates of digital gender inequalities at the subnational level are lacking. These estimates, however, are essential for monitoring gaps within countries and implementing targeted interventions within the global sustainable development goals, which emphasize the need to close inequalities both between and within countries. We develop estimates of internet and mobile adoption by gender and digital gender gaps at the subnational level for 2,075 regions in 117 LMICs from 2015 through 2025, a context where digital penetration is low and national-level gender gaps disfavoring women are large. We construct these estimates by applying machine-learning algorithms to Facebook user counts, geospatial data, development indicators, and population composition data. We calibrate and assess the performance of these algorithms using ground-truth data from subnationally representative household survey data from 33 LMICs. Our results reveal striking disparities in access to mobile and internet technologies between and within LMICs. These disparities imply that as of 2025, women are 19% less likely to use the internet and 8% less likely to own a mobile phone in LMICs, corresponding to over 190 million fewer women owning a mobile phone and over 320 million fewer women using the internet.
- Open Access
- 7 May 2025
- Published by Oxford University Press (OUP) in American Journal of Epidemiology
- Vol. 195 (1) , 49-59
Estimating death rates in complex humanitarian emergencies using the network survival method
Reliable estimates of death rates in complex humanitarian emergencies are critical for assessing the severity of a crisis and for effectively allocating resources. However, in many humanitarian settings, logistical and security concerns make conventional methods for estimating death rates infeasible. We develop and test a new method for estimating death rates in humanitarian emergencies using reports of deaths in survey respondents’ social networks. To test our method, we collected original data in Tanganyika Province of the Democratic Republic of the Congo (n = 5311), a setting where reliable estimates of crude death rates (CDR) are in high demand. Qualitative fieldwork suggested testing 2 different types of personal networks as the basis for CDR estimates: deaths among immediate neighbors and deaths among kin. We compare our network-based estimates (0.44 deaths per 10 000 person-days) against a standard retrospective household mortality survey, which estimated a CDR nearly twice as high (0.81 deaths per 10 000 person-days). Given that both methods are equally plausible, our findings highlight the need for further validation and development of both methods.
- 7 May 2025
- Published by Oxford University Press (OUP) in American Journal of Epidemiology
- Vol. 195 (1) , 49-59
Open AccessEstimating death rates in complex humanitarian emergencies using the network survival method
Reliable estimates of death rates in complex humanitarian emergencies are critical for assessing the severity of a crisis and for effectively allocating resources. However, in many humanitarian settings, logistical and security concerns make conventional methods for estimating death rates infeasible. We develop and test a new method for estimating death rates in humanitarian emergencies using reports of deaths in survey respondents’ social networks. To test our method, we collected original data in Tanganyika Province of the Democratic Republic of the Congo (n = 5311), a setting where reliable estimates of crude death rates (CDR) are in high demand. Qualitative fieldwork suggested testing 2 different types of personal networks as the basis for CDR estimates: deaths among immediate neighbors and deaths among kin. We compare our network-based estimates (0.44 deaths per 10 000 person-days) against a standard retrospective household mortality survey, which estimated a CDR nearly twice as high (0.81 deaths per 10 000 person-days). Given that both methods are equally plausible, our findings highlight the need for further validation and development of both methods.
- Open Access
- 1 March 2025
- Published by Center for Open Science in SocArXiv
Estimating death rates in complex humanitarian emergencies using the network survival method
Reliable estimates of death rates in complex humanitarian emergencies are criticalfor assessing the severity of a crisis and for effectively allocating resources. However,in many humanitarian settings, logistical and security concerns make conventionalmethods for estimating death rates infeasible. We develop and test a new methodfor estimating death rates in humanitarian emergencies using reports of deaths insurvey respondents’ social networks. To test our method, we collected original data inTanganyika Province of the Democratic Republic of the Congo, a setting where reliableestimates of death rates are in high demand. Qualitative fieldwork suggested testingtwo different types of personal networks as the basis for death rate estimates: deathsamong immediate neighbors and deaths among kin. We benchmarked our networkestimates against a standard retrospective household mortality survey, which estimateda crude death rate nearly twice as high as our network-based methods. Given bothmethods are equally plausible, our findings underscore the need for further validationand development of both methods.
- 1 March 2025
- Published by Center for Open Science in SocArXiv
Open AccessEstimating death rates in complex humanitarian emergencies using the network survival method
Reliable estimates of death rates in complex humanitarian emergencies are criticalfor assessing the severity of a crisis and for effectively allocating resources. However,in many humanitarian settings, logistical and security concerns make conventionalmethods for estimating death rates infeasible. We develop and test a new methodfor estimating death rates in humanitarian emergencies using reports of deaths insurvey respondents’ social networks. To test our method, we collected original data inTanganyika Province of the Democratic Republic of the Congo, a setting where reliableestimates of death rates are in high demand. Qualitative fieldwork suggested testingtwo different types of personal networks as the basis for death rate estimates: deathsamong immediate neighbors and deaths among kin. We benchmarked our networkestimates against a standard retrospective household mortality survey, which estimateda crude death rate nearly twice as high as our network-based methods. Given bothmethods are equally plausible, our findings underscore the need for further validationand development of both methods.