(Non-)Return of Ukrainian War Refugees: Modeling Repatriation Using Logistic Regression
The scale of forced migration of Ukrainians following the full-scale Russian invasion of Ukraine on February 24, 2022, population losses, destruction of infrastructure, and other consequences underscore the importance of addressing the issue of return migration. The return of forced migrants from Uk...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taras Shevchenko National University of Kyiv
2024-12-01
|
| Series: | Соціологічні студії |
| Subjects: | |
| Online Access: | https://sociostudios.vnu.edu.ua/index.php/socio/article/view/406 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850150946886123520 |
|---|---|
| author | Mykola Sydorov Svіtlana Salnikova Anna Severyn |
| author_facet | Mykola Sydorov Svіtlana Salnikova Anna Severyn |
| author_sort | Mykola Sydorov |
| collection | DOAJ |
| description | The scale of forced migration of Ukrainians following the full-scale Russian invasion of Ukraine on February 24, 2022, population losses, destruction of infrastructure, and other consequences underscore the importance of addressing the issue of return migration. The return of forced migrants from Ukraine is a significant phenomenon that arises in socio-demographic and economic contexts and requires theoretical revision and conceptual clarification. The purpose of this study is to develop predictive models for the repatriation of Ukrainian war refugees and to evaluate the factors that decrease the likelihood of their return. The authors used the Factum Group dataset, focusing on Ukrainian women who left Ukraine after the full-scale invasion and did not return from emigration, to obtain a sample of forcibly displaced Ukrainian migrant women. Repatriation models were constructed using logistic regression, applied to the overall dataset, and to subsamples from Germany and Poland – countries that received the largest number of Ukrainian war refugees. The assumption that return factors depend on the host country was confirmed. Repatriation models for Germany and Poland demonstrate higher accuracy than the general model and reveal distinct sets of factors influencing the (non-)return of war refugees. In Germany, key factors include the standard of living, availability of work abroad, overall life satisfaction, and prior residence in areas of Ukraine unaffected by active hostilities. For Poland, the significant factors are overall life satisfaction and the intention to return. It was found that not all factors influencing the decision to leave the country – such as region, availability of financial resources, knowledge of a foreign language, and family status – are significant for the decision to return. The repatriation models demonstrate a high level of accuracy in predicting negative cases and moderate accuracy in predicting positive cases, with cases of uncertainty excluded from the regression analysis. These findings confirm that factors related to economic stability, social comfort, and personal intentions are fundamental for predicting the likelihood of (non)return among Ukrainian female migrants. |
| format | Article |
| id | doaj-art-ef38c5c17ff841a0ba09388f8eafd550 |
| institution | OA Journals |
| issn | 2306-3971 2521-1056 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Taras Shevchenko National University of Kyiv |
| record_format | Article |
| series | Соціологічні студії |
| spelling | doaj-art-ef38c5c17ff841a0ba09388f8eafd5502025-08-20T02:26:24ZengTaras Shevchenko National University of KyivСоціологічні студії2306-39712521-10562024-12-012(25)819310.29038/2306-3971-2024-02-33-33324(Non-)Return of Ukrainian War Refugees: Modeling Repatriation Using Logistic RegressionMykola Sydorov0https://orcid.org/0000-0002-5333-8393Svіtlana Salnikova1https://orcid.org/0000-0001-6368-9480Anna Severyn2https://orcid.org/0009-0006-4610-5432Taras Shevchenko National University of KyivTaras Shevchenko National University of KyivTaras Shevchenko National University of KyivThe scale of forced migration of Ukrainians following the full-scale Russian invasion of Ukraine on February 24, 2022, population losses, destruction of infrastructure, and other consequences underscore the importance of addressing the issue of return migration. The return of forced migrants from Ukraine is a significant phenomenon that arises in socio-demographic and economic contexts and requires theoretical revision and conceptual clarification. The purpose of this study is to develop predictive models for the repatriation of Ukrainian war refugees and to evaluate the factors that decrease the likelihood of their return. The authors used the Factum Group dataset, focusing on Ukrainian women who left Ukraine after the full-scale invasion and did not return from emigration, to obtain a sample of forcibly displaced Ukrainian migrant women. Repatriation models were constructed using logistic regression, applied to the overall dataset, and to subsamples from Germany and Poland – countries that received the largest number of Ukrainian war refugees. The assumption that return factors depend on the host country was confirmed. Repatriation models for Germany and Poland demonstrate higher accuracy than the general model and reveal distinct sets of factors influencing the (non-)return of war refugees. In Germany, key factors include the standard of living, availability of work abroad, overall life satisfaction, and prior residence in areas of Ukraine unaffected by active hostilities. For Poland, the significant factors are overall life satisfaction and the intention to return. It was found that not all factors influencing the decision to leave the country – such as region, availability of financial resources, knowledge of a foreign language, and family status – are significant for the decision to return. The repatriation models demonstrate a high level of accuracy in predicting negative cases and moderate accuracy in predicting positive cases, with cases of uncertainty excluded from the regression analysis. These findings confirm that factors related to economic stability, social comfort, and personal intentions are fundamental for predicting the likelihood of (non)return among Ukrainian female migrants.https://sociostudios.vnu.edu.ua/index.php/socio/article/view/406war refugeesforced migrationrepatriation modellogistic regressionmigration factorsrepatriation factors |
| spellingShingle | Mykola Sydorov Svіtlana Salnikova Anna Severyn (Non-)Return of Ukrainian War Refugees: Modeling Repatriation Using Logistic Regression Соціологічні студії war refugees forced migration repatriation model logistic regression migration factors repatriation factors |
| title | (Non-)Return of Ukrainian War Refugees: Modeling Repatriation Using Logistic Regression |
| title_full | (Non-)Return of Ukrainian War Refugees: Modeling Repatriation Using Logistic Regression |
| title_fullStr | (Non-)Return of Ukrainian War Refugees: Modeling Repatriation Using Logistic Regression |
| title_full_unstemmed | (Non-)Return of Ukrainian War Refugees: Modeling Repatriation Using Logistic Regression |
| title_short | (Non-)Return of Ukrainian War Refugees: Modeling Repatriation Using Logistic Regression |
| title_sort | non return of ukrainian war refugees modeling repatriation using logistic regression |
| topic | war refugees forced migration repatriation model logistic regression migration factors repatriation factors |
| url | https://sociostudios.vnu.edu.ua/index.php/socio/article/view/406 |
| work_keys_str_mv | AT mykolasydorov nonreturnofukrainianwarrefugeesmodelingrepatriationusinglogisticregression AT svítlanasalnikova nonreturnofukrainianwarrefugeesmodelingrepatriationusinglogisticregression AT annaseveryn nonreturnofukrainianwarrefugeesmodelingrepatriationusinglogisticregression |