Kwangman Ko

This study examined the applicability of regularized regression in family science by focusing on fathers’ psychological well-being as an applied example. Using cross-sectional survey data from 277 U.S. fathers, we compared Ordinary Least Squares (OLS) regression with Ridge, LASSO (Least Absolute Shrinkage and Selection Operator), and Elastic Net in regularized regression to evaluate how diverse familial factors—including father involvement, father role identity, parenting competency, parent–child relationships, coparenting relationships, and work–family conflict—concurrently predict fathers’ well-being. Results indicated that while the three regularized approaches produced similar overall performance, they provided advantages over OLS by reducing overfitting, improving generalizability, and enhancing interpretability. We also compared the analysis results of the three approaches in terms of the prediction accuracy and interpretability. The findings reveal how regularized regression can highlight key factors, such as father involvement, coparenting, and parenting competency. This study also illustrates how regularized regression complements traditional regression in family science by addressing multicollinearity, reducing overfitting, and supporting feature selection. These results highlight the potential of computational approaches such as machine learning to strengthen both theory and practice in family research.

Using a person-centered approach grounded in family systems theory, we identified subgroups of fathers based on their involvement types and examined how these profiles differed regarding intergenerational transmission, familial factors, and fathers’ sociodemographic characteristics. Data were collected from 277 U.S. fathers, revealing two distinct profiles: High-involved and Low-involved fathers. High-involved fathers demonstrated significantly greater engagement, such as praising and affection, than their Low-involved counterparts. Regarding intergenerational transmission, our findings support the modeling hypothesis, indicating that positive paternal emotional experiences in childhood are associated with increased paternal involvement. However, there was limited evidence for compensatory hypothesis behaviors. High-involved fathers reported better psychological well-being, stronger father role identities, better parenting competence, closer parent–child relationships, better coparenting relationships, and greater work and family conflicts. Low-involved fathers were more likely to be younger and single-not-married. These findings can be referred to by practitioners and policymakers for developing intervention programs or policies supporting paternal involvement.

Objective: Guided by a family systems theoretical framework, the study reported herein explores the utility of using supervised machine learning random forest regression for understanding fathers' psychological well‐being.Background: Although fathers' psychological well‐being has not received much attention, understanding how familial factors contribute to fathers' mental health will benefit fathers themselves as well as their families, given the interdependence of the family system. Supervised machine learning using a random forest regression can be useful for identifying the hierarchical relationships between factors that shape fathers' psychological well‐being.Method: The study includes 277 U.S. fathers with at least one preschool‐aged child as participants. Study variables include fathers' psychological well‐being, father involvement, parental competency, parent–child relationships, coparenting relationship quality, work and family conflict, and fathers' demographic information.Results: The supervised machine learning model was trained using a random forest regression. After tuning, the random forest regression with the training data identified parent–child relationship conflict as the most important predictor, followed by father involvement, coparenting relationships, work and family conflict, and parental competency (R² = .62).Conclusion: This research shows the benefits of taking a supervised machine learning random forest statistical approach to increasing understanding of the complexity of factors related to fathers' psychological well‐being.Implications: To support fathers' psychological well‐being, practitioners and family educators may consider addressing familial factors such as parent–child relationship conflict, father involvement, and coparenting relationship quality within a family.

With the emphasis on children’s responsibility for the care of ageing parents, this study examined how Chinese adult children’s support provided to parents was associated with filial piety, support from parents and parent-child contact frequency. With the 2006 Chinese General Social Survey, we used structural equation modelling with 1,452 adults with two living parents and tested the model for sons and daughters separately. For both groups, the results showed that (1) filial piety was positively associated with emotional support provided to parents; (2) support received from parents was positively related to instrumental and emotional support to parents; and (3) parent-child contact frequency was linked to instrumental support. For adult daughters, financial support was positively associated with the support received from parents and negatively related to parent-child contact frequency. This study suggests that the traditional norm of filial piety may be less influential than other factors for adult children’s support behaviour.

First-generation college students (FGCSs) are known to experience more difficulty in the college education environment compared to continuing-generation college students. The current study used a person-centered approach to investigate subgroups of FGCSs that have had similar experiences of social, financial, academic, and cultural barriers in their college lives. A total of 382 undergraduate students in the U.S. participated in the study. A latent profile analysis revealed there may be two particular profiles for FGCSs including these four barriers. FGCSs in Profile 2 reported more social, financial, academic, and cultural barriers. In addition, these students also reported lower levels of self-esteem and higher levels of stress, anxiety, depression, and drop-out intention than those in Profile 1. Students’ gender, their parent's marital status, and the estimated debt after graduation significantly predicted profile membership. Interventions to support FGCSs will also be discussed.

We compared attitudes about child support between the United States and South Korean respondents, using a multiple segment factorial vignette to gather quantitative and qualitative data from 132 Americans and 132 South Koreans. South Koreans were more likely than U.S. respondents to think child support should be paid. Respondents in both countries thought fathers should pay more child support when mothers had sole custody than when both parents shared custody. Remarriage, combined with custody, affected respondents’ attitudes. The primary rationale of U.S. respondents focused on issues of fairness, while Koreans highlighted fathers’ responsibility and obligation. Cross-cultural differences in attitudes regarding child support payments, and their implications for support compliance rates in the United States and South Korea, are discussed.

This study examined how the longitudinal associations among children’s negative emotionality, mothers’ depressive symptoms, parental warmth, and children’s school readiness and whether the associations vary as a function of fathers’ positive involvement in low-income South Korean families. Participants were 399 families including mothers (M_age = 32.54 years at Time 1), fathers (M_age = 35.23 years at Time 1), and children (M_age = 38.92 months at Time 1; 50.5% boys) in the Panel Study on Korean Children. Results revealed that children’s negative emotionality was indirectly associated with their school readiness three years later, through its association with mothers’ depressive symptoms and warmth. Mothers’ warmth mediated the association between mothers’ depressive symptoms and children’s school readiness, and fathers’ warmth mediated the association between fathers’ positive involvement and children’s school readiness. Our findings revealed the family processes underlying children’s school readiness development in low-income Korean family contexts. Our findings also provide information useful for efforts to detect family risks and to establish family policies to promote low-income children’s school readiness.