23456785 of 13
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Life Course Prediction for Labor Market Outcomes: Lessons from Register-Based Machine Learning
Stockholm University, Faculty of Social Sciences, Department of Sociology.ORCID iD: 0009-0006-0117-9120
2026 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Machine learning has begun making its way into the methodological toolkit for the social sciences, but only very recently, with many questions still unanswered. Combining this toolkit with population register data offers a promising perspective into researching the human life course and the complexities within. This dissertation explores how Swedish and Finnish population registers combined with decision tree-based machine learning algorithms can be leveraged for labor market research, in particular for studying the predictability of long-term unemployment in early adulthood and lifetime disposable income. The theoretical foundations on this work lie at the intersection of the life course framework and how different forms of capital accumulate and transfer intergenerationally. Machine learning offers a useful tool to jointly model multiple domains in life, and how they together form trajectories towards these labor market outcomes.

Studies 1 and 2 concern long-term unemployment in early adulthood in the context of Finland and leverage data from the full birth cohort of 1987. The predictors for both studies are chosen based on theories and previous empirical literature on unemployment, and operationalized longitudinally where possible, some following the ages related to school transitions. Study 1 focuses on the extent to which long-term unemployment at ages 25–30 can be predicted, as well as the differences in predictability across the outcome groups and socioeconomic background. The results show that when compared to logistic regression, all the tree-based models fare better overall, but their performance is mostly due to correctly predicting the outcome class employed, while incorrectly predicting most of the unemployed. Regarding the socioeconomic origin, the low socioeconomic status (SES) group is more difficult to predict overall than the high-SES group, but within the low-SES group, the correct predictions for the unemployed increase. The findings highlight our limited understanding of the processes leading to unemployment, as well as the importance of learning more about sub-group predictability. Study 2 unpacks these results for the best performing model, Gradient Booster, by exploring the combinations that predict someone as: A) employed or unemployed, and B) correctly or incorrectly employed or unemployed. The aim is to further evaluate which processes and theories are the model’s blind spots, and the method used is Principal Component Analysis. For the employed group, parental unemployment was the prominent component steering the model wrong, while for the unemployed it was the combination of family cohesion, mental wellbeing and intergenerational transmission of education.

Study 3 focuses on lifetime disposable income in the context of Sweden, leveraging population registers for the birth cohorts of 1967–1971. This study explores the extent to which it is possible to predict a person’s income class based on their early life conditions and further unpacks this to answer which characteristics each income class shares, and which sub-groups end up in the same income class, through different combinations of characteristics. The predictors are based on theories and empirical literature, not only focusing on socioeconomic origins, but also geographic origins, family structure, and gender. The analysis of predictability compares logistic regression to two tree-based algorithms, as well as a composite model where a different model for each income group is chosen based on highest performance. The results find the machine learning models to outperform multinomial logistic regression, with the composite model being the best. The middle-income class has the highest predictive accuracy, while the lowest income class has the lowest. These predictions are then unpacked using Shapley Additive Explanations, which are further analyzed using Hierarchical Agglomerative Clustering. The results reveal heterogeneity both across and within income classes, with complex realities at the bottom of the income distribution, and a particularly streamlined path at the top.

This dissertation contributes to the field of life course prediction by exploring the predictability of the two labor market outcomes and highlights the importance of exploring the predictability of sub-groups. Furthermore, it presents ways in which population register data can be analyzed with machine learning methods to reveal more details about the human life course.

Place, publisher, year, edition, pages
Stockholm: Department of Sociology, Stockholm University , 2026. , p. 47
Series
Dissertation series / Stockholm University Demography Unit, ISSN 1404-2304 ; 34
Keywords [en]
life course, machine learning, labor market, unemployment, income, population registers, register-based, life course prediction
National Category
Sociology (Excluding Social Work, Social Anthropology, Demography and Criminology) Demography
Research subject
Sociological Demography
Identifiers
URN: urn:nbn:se:su:diva-260353ISBN: 978-91-8107-753-7 (print)ISBN: 978-91-8107-754-4 (electronic)OAI: oai:DiVA.org:su-260353DiVA, id: diva2:2106243
Public defence
2026-11-19, hörsal 5, hus B, Universitetsvägen 10 B and online via Zoom, public link is available at the department website, Stockholm, 13:00 (English)
Opponent
Supervisors
Available from: 2026-10-27 Created: 2026-09-29 Last updated: 2026-10-15Bibliographically approved
List of papers
1. The (Un)Predictability of Early (Un)Employment: A Machine Learning Approach
Open this publication in new window or tab >>The (Un)Predictability of Early (Un)Employment: A Machine Learning Approach
2024 (English)In: Socius: Sociological Research for a Dynamic World, ISSN 2378-0231, Vol. 10Article in journal (Refereed) Published
Abstract [en]

This article integrates scholarship on determinants of labor market transitions with the flexible modeling strategies available through machine learning. Leveraging population registers from Finland, this study analyzes how well we can predict (un)employment and for whom we can (not) predict. The study finds that the predictability of long-term (un)employment improves when using tree-based nonparametric models compared to logistic regression but that predictability varies substantially across outcome groups. None of the models predict very accurately for the unemployed net of baseline likelihood to be unemployed—a group that is well researched and often of interest when designing policies and interventions. Overall accuracy is driven by the employed while incorrectly predicting most of the unemployed. Additionally, the outcomes for individuals with low parental education are overall more difficult to predict than the outcomes for individuals with mid/high parental education, whereas predicting unemployment slightly improves among the low parental education group.

Keywords
unemployment, life course, machine learning, prediction, register-based
National Category
Sociology (Excluding Social Work, Social Anthropology, Demography and Criminology)
Identifiers
urn:nbn:se:su:diva-234803 (URN)10.1177/23780231241286655 (DOI)001335530700001 ()2-s2.0-85207014938 (Scopus ID)
Funder
Swedish Research Council, 2019-00245
Available from: 2024-10-22 Created: 2024-10-22 Last updated: 2026-09-29Bibliographically approved
2. Unpacking the Unpredictability of Early Unemployment: Exposing the Blind Spots of a Machine Learning Model
Open this publication in new window or tab >>Unpacking the Unpredictability of Early Unemployment: Exposing the Blind Spots of a Machine Learning Model
(English)Manuscript (preprint) (Other academic)
Abstract [en]

Machine learning (ML) powered AI tools are increasingly popular in public agencies designing interventions or tools that aid decision-making. Yet, we know little about the sociological mechanics behind the models, and for whom they make good predictions. This study unpacks a Gradient Booster model predicting long-term unemployment in early adulthood and critically evaluates how ML models can misunderstand labor market inequalities. The model evaluated is a Gradient Booster predicting (un)employment, using longitudinal and intergenerationally linked life course data from Finnish population registers. Principal Component Analysis was used to discover combinations of early life experiences behind correct vs. incorrect predictions. The incorrectly predicted combinations, as well as the correctly predicted, highlighted well-established mechanisms for (un)employment. The results underline the importance of sociologically unpacking ML models for different outcome groups, to expose the possible blind spots of the models.

Keywords
machine learning, unemployment, life course, population register, labor market, inequality
National Category
Sociology (Excluding Social Work, Social Anthropology, Demography and Criminology) Demography
Research subject
Sociological Demography; Sociology
Identifiers
urn:nbn:se:su:diva-260354 (URN)
Available from: 2026-09-29 Created: 2026-09-29 Last updated: 2026-09-29
3. Diverging Paths to Economic Destination: Uncovering Heterogeneity in Early Life Conditions Across and Within Income Classes
Open this publication in new window or tab >>Diverging Paths to Economic Destination: Uncovering Heterogeneity in Early Life Conditions Across and Within Income Classes
(English)Manuscript (preprint) (Other academic)
Abstract [en]

This study aims to understand how combinations of multi-domain early life conditions together form diverging paths to economic destinations. Parental socioeconomic status (SES), family structure, gender, geographic origin, and migration are viewed through the life course perspective and modelled together using decision tree-based machine learning tools on Swedish population register data. The results demonstrate that machine learning performs better than multinomial logistic regression when it comes to capturing the patterns that produce variance in lifetime disposable income. Furthermore, combinations of early life conditions form vastly varying paths to economic destinations at different parts of the income distribution: the pathways at the bottom of the distribution are complex, whereas the road to the top is more streamlined Moreover, there is heterogeneity within each income class: people ending up in an income class can consist of several sub-groups defined by different combinations of early life conditions. One major contribution of this study is revealing through its novel modelling strategy how theories of income accumulation may work at some parts of the distribution, but not at others: more attention should be paid for developing theories for different parts of the distribution. 

Keywords
machine learning, life course, income, labor market, inequality
National Category
Sociology (Excluding Social Work, Social Anthropology, Demography and Criminology) Demography
Research subject
Sociological Demography; Sociology
Identifiers
urn:nbn:se:su:diva-260355 (URN)
Available from: 2026-09-29 Created: 2026-09-29 Last updated: 2026-09-29

Open Access in DiVA

Life Course Prediction for Labor Market Outcomes: Lessons from Register-Based Machine Learning(3749 kB)63 downloads
File information
File name FULLTEXT01.pdfFile size 3749 kBChecksum SHA-512
24e2a3fba0c0fc645287a1c1a1c095f341fa42c73415d03e31a296e392cb2936acb409b3b5a3f50aad71959eca16cad77004618bdbe5e6c21dae32e44317375a
Type fulltextMimetype application/pdf

Authority records

Kuikka, Sanni

Search in DiVA

By author/editor
Kuikka, Sanni
By organisation
Department of Sociology
Sociology (Excluding Social Work, Social Anthropology, Demography and Criminology)Demography

Search outside of DiVA

GoogleGoogle Scholar
Total: 63 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 3209 hits
23456785 of 13
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf