Caring practices of marriage migrant women in Korea

Abstract

Korea has experienced tremendous changes to virtually all aspects of its family structure, including an influx of international marriage migrants from developing countries in Asia. Because marriage migrant females are biological and cultural reproducers, they play a significant role as caregivers for their husband’s family. Despite the importance, our understanding about its burden is significantly limited. To fill this gap, this paper examined caregiving practices of marriage migrants. Results indicate that there are large country-of-origin variations in caring for both disabled family members and young children. An interaction between education and the country of origin showed more nuanced pattern.

Keywords:

• International marriage
• care burden
• disability and child care
• South Korea

Notes

1. Contrary to the 2009 NSMF, which pursued complete enumeration of marriage migrants in Korea, the 2012 NSMF collected data from sample of them (Jeon, Kim, Yi, Joo, & Choi, 2011). Consequently, the number of respondents in the 2012 NSMF decreased significantly compared to the previous round of the NSMF. As of writing this paper, the 2015 NSMF data had been released to the public. However, one of the dependent variables in this study, the presence of a disabled family member, was unavailable for unknown reasons. Due to this data limitation, the current study did not use the most recent round of the NSMF.

2. Preliminary analysis indicated that educational attainment is negatively and linearly associated with two measures of care burden. In other words, the evidence of non-linear effect of education on care burden was not found, and hence it is treated as a continuous variable in subsequent analysis.

3. It should be noted that there were differences in types of respondents between the two waves of NSMF. That is, respondent of NSMF 2008 was migrant wives whereas NSMF 2012 respondent was Korean husbands. Thus, it is possible that a discrepancy in the definition of the disabled between migrant wives and their husbands may exist. For instance, Korean husbands may consider a disabled as holding a government-issued disability registration card while migrant wives perceive any family member having needs for help in daily living. To the extent that Korean husbands and their migrant wives define the disabled differently, the proportion of living with disabled family members might be decreased between the two waves of NSMF.

4. Approximately 1 out of 7 working age population (20–64) in OECD member countries (14.3%) had a chronic illness or disability that caused difficulties in daily life. The discrepancy between Korea and other OECD member states is largely due to the methods of data collection. That is, most OECD countries count disabled people through censuses while Korean data relies on the number of registered disabled individuals, which applies strict screening criteria (Korea Institute for Health and Social Affairs, 2010).

5. This discrepancy may also be associated with different type of survey respondent between the two waves of NSMF. A large drop in the percentage in ‘not working, previously worked’ and an increase in percentage of ‘never worked’ may be attributable to that Korean husbands do not know their wives’ career when they were in their country of origin or may not recognise the work that they were engaged in as ‘work’.

6. If information about marriage migrant’s religious affiliations is available, it would be helpful to understand the association between meeting through religious institutions and likelihood of living with disabled family members more thoroughly. Unfortunately, the NSMF in both waves did not ask about any religious preferences not only for marriage migrants but also for family members.

7. It is possible that there are variations in the mean age of migrant women by their origin country and those are interacting with educational attainment. In other words, there might be a 3-way interaction between age of migrant women, country of origin, and educational attainment. In order to examine the effect, 3-way interaction (i.e. age × country of origin × education) entered the equation. However, all three variables were categorical, and, thus, logistic regression was not estimated because there were some empty cells.