Do Adolescent Neighborhoods Predict Educational Partner Choice? Evidence from Sweden, 1990-2022
de Souza Maia, Vinicius LU
; Dribe, Martin
LU
and Hedefalk, Finn
LU
(2026)
In Lund Papers in Economic Demography
- Abstract
- Individual and family characteristics are important determinants of partner selection byeducation, but social environments during upbringing, such as neighborhoods, also shapepreferences and behavior with long-term consequences. This study examines how educationallevel in the neighborhood of upbringing relates to having a university-educated partner. Usingfull-count register data and logistic regressions, we analyze Swedish cohorts born 1974–1982,followed from 1990 to 2022. We measure neighborhood conditions by the share of highlyeducated among the parents of same-aged peers in the neighborhood at age 16, using a fine-grained k-nearest neighbor approach. Our findings show that a higher educational level in theneighborhood increases the... (More)
- Individual and family characteristics are important determinants of partner selection byeducation, but social environments during upbringing, such as neighborhoods, also shapepreferences and behavior with long-term consequences. This study examines how educationallevel in the neighborhood of upbringing relates to having a university-educated partner. Usingfull-count register data and logistic regressions, we analyze Swedish cohorts born 1974–1982,followed from 1990 to 2022. We measure neighborhood conditions by the share of highlyeducated among the parents of same-aged peers in the neighborhood at age 16, using a fine-grained k-nearest neighbor approach. Our findings show that a higher educational level in theneighborhood increases the likelihood of having a partner with a university degree later inlife. This holds for both genders, and for first and higher-order unions, with strongerassociations in larger cities and among individuals without tertiary education and from lower-educated origins. Neighborhood associations exist beyond parental background and owneducational attainment. Since the associations are strongest at the lower ends of both own andparental education, the overall tendency toward educational homogamy may be reduced. Thissuggests that partner-relevant preferences and networks are shaped early in life and remainquite stable over the life course. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/6c9adb54-b0f2-424a-975b-bd0cbad060a7
- author
- de Souza Maia, Vinicius
LU
; Dribe, Martin
LU
and Hedefalk, Finn
LU
- organization
- publishing date
- 2026-09-07
- type
- Working paper/Preprint
- publication status
- published
- subject
- keywords
- neighborhood effects, education, partner selection, bespoke neighborhoods, k-nearest neighbor, Sweden
- in
- Lund Papers in Economic Demography
- issue
- 2026:4
- pages
- 45 pages
- project
- Socioeconomic Segregation – The Impact of Neighborhoods, Schools and Policy Across the Life Course
- The long reach of the neighborhood: Health, education and earnings in Landskrona, Sweden, 1904-2015
- language
- English
- LU publication?
- yes
- id
- 6c9adb54-b0f2-424a-975b-bd0cbad060a7
- alternative location
- https://www.lusem.lu.se/sites/lusem.lu.se/files/2026-09/LPED%202026%204_0.pdf
- date added to LUP
- 2026-09-07 12:56:42
- date last changed
- 2026-09-11 07:18:43
@misc{6c9adb54-b0f2-424a-975b-bd0cbad060a7,
abstract = {{Individual and family characteristics are important determinants of partner selection byeducation, but social environments during upbringing, such as neighborhoods, also shapepreferences and behavior with long-term consequences. This study examines how educationallevel in the neighborhood of upbringing relates to having a university-educated partner. Usingfull-count register data and logistic regressions, we analyze Swedish cohorts born 1974–1982,followed from 1990 to 2022. We measure neighborhood conditions by the share of highlyeducated among the parents of same-aged peers in the neighborhood at age 16, using a fine-grained k-nearest neighbor approach. Our findings show that a higher educational level in theneighborhood increases the likelihood of having a partner with a university degree later inlife. This holds for both genders, and for first and higher-order unions, with strongerassociations in larger cities and among individuals without tertiary education and from lower-educated origins. Neighborhood associations exist beyond parental background and owneducational attainment. Since the associations are strongest at the lower ends of both own andparental education, the overall tendency toward educational homogamy may be reduced. Thissuggests that partner-relevant preferences and networks are shaped early in life and remainquite stable over the life course.}},
author = {{de Souza Maia, Vinicius and Dribe, Martin and Hedefalk, Finn}},
keywords = {{neighborhood effects; education; partner selection; bespoke neighborhoods; k-nearest neighbor; Sweden}},
language = {{eng}},
month = {{09}},
note = {{Working Paper}},
number = {{2026:4}},
series = {{Lund Papers in Economic Demography}},
title = {{Do Adolescent Neighborhoods Predict Educational Partner Choice? Evidence from Sweden, 1990-2022}},
url = {{https://www.lusem.lu.se/sites/lusem.lu.se/files/2026-09/LPED%202026%204_0.pdf}},
year = {{2026}},
}