Journal of Psychosociological Research in Family and Culture

Predicting Suicidal Ideation Among Youth Using Family, Cultural, School, and Digital-Ecology Variables: A Stacked Ensemble Model with Calibration, Explainability, and Subgroup Fairness Assessment

Elpida Christou iD , Hsin-Yi Chang ^* iD , Andrés Benítez iD

Objective: This study aimed to develop and independently evaluate a calibrated and explainable stacked ensemble model for predicting clinically significant suicidal ideation among youth in Taiwan using family, cultural, school, and digital-ecology variables.

Methods and Materials: A cross-sectional predictive modeling study was conducted with 4,812 Taiwanese students aged 15–24 years. Participants were recruited from secondary schools, vocational schools, junior colleges, and universities through multistage stratified cluster sampling. Suicidal ideation was assessed using the Suicidal Ideation Attributes Scale. Candidate predictors included family functioning, family support, family conflict, filial-piety orientations, loss-of-face concerns, mental-health stigma, school belonging, academic stress, bullying, social-media disorder symptoms, cyberbullying, nighttime digital use, exposure to self-harm content, and digital resilience. Data were divided into development, calibration, and independent holdout samples. Elastic-net regression, support vector machines, random forests, gradient boosting, and extreme gradient boosting were combined using a regularized logistic meta-learner. Model performance was examined through discrimination, calibration, explainability, decision-curve analysis, and subgroup fairness assessment.

Findings: The calibrated stacked ensemble outperformed the conventional logistic-regression benchmark and all individual learners. In the independent holdout sample, the model achieved an area under the receiver operating characteristic curve of .889, an area under the precision–recall curve of .682, sensitivity of .861, specificity of .812, positive predictive value of .549, negative predictive value of .957, and Brier score of .115. Calibration was strong, with an intercept of 0.02 and slope of .98. School belonging, family functioning, cyberbullying, academic stress, family conflict, and exposure to self-harm-related digital content were the most influential predictors. Subgroup analyses showed generally comparable sensitivity, although specificity and calibration were weaker among several disadvantaged and smaller subgroups.

Conclusion: A calibrated stacked ensemble integrating multiple ecological domains can provide accurate and interpretable estimates of youth suicidal ideation, although its predictions should support rather than replace direct professional assessment and must be accompanied by ongoing fairness monitoring.

Preliminary Evidence on Moral Underpinnings of Rape Victim Blaming in Iran: Replication and Extensions of Milesi et al. (2020)

Mazyar Bagherian ^* iD , Homa Sheibani iD , Dr. Zohreh Hashemi iD , Bahare Bahmani iD

Objective: According to Moral Foundation Theory, people rely on five intuitive-based foundations in their moral decisions. These foundations are divided into Individualizing foundations (Care and Fairness) and Binding foundations (Loyalty, Authority, and Sanctity).

Methods and Materials: In an attempt to replicate Milesi at al’s (2020) findings in a non-WEIRD and honored context, across three studies, we explored the relationship between Moral Foundation and Acceptance of Modern Myths About Sexual Aggression (AMMSA) with rape victim-blaming.

Findings: Study 1 showed that Fairness, Authority, and Sanctity foundations were predictive of rape victim-blaming. Study 2 showed that binding foundations predict rape victim-blaming in a preregistered analysis. Our cross-country analyses (Iran, Germany, Spain, and Italy) showed that the effects of binding moral foundations were stronger among Iranian participants than those with less-honored cultures (i.e., Italy, Spain, and Germany). Exploratory analyses revealed no differences in rape victim-blaming among men and women. However, men had higher levels of endorsement of AMMSA than women. Religious individuals had higher levels of rape victim-blaming and AMMSA. Findings are discussed concerning the honored context of Iran’s society.

Conclusion: The findings indicate that rape victim-blaming in Iran is strongly associated with acceptance of modern myths about sexual aggression and binding moral foundations, particularly Authority and Sanctity. These relationships appear especially salient within honor-based cultural contexts, highlighting the importance of culturally sensitive approaches to understanding and reducing victim-blaming attitudes.

Predicting Intergenerational Cultural Conflict in Ecuadorian Immigrant Families Based on Acculturation Gaps, Language Brokering, Perceived Discrimination, Parenting Styles, and Family Communication Patterns

Carla Quispe iD , Camila Espinoza ^* iD

Objective: This study aimed to predict intergenerational cultural conflict in Ecuadorian immigrant families based on parent–adolescent acculturation gaps, adolescent language-brokering burden, perceived discrimination, parenting styles, and family communication patterns.

Methods and Materials: This cross-sectional correlational study included 312 Ecuadorian immigrant parent–adolescent dyads residing in the New York metropolitan area. Participants were recruited through schools, community centers, cultural associations, religious organizations, and immigrant-support programs. Data were collected using the Intergenerational Conflict Inventory, Vancouver Index of Acculturation, Adolescent Language Brokering Questionnaire, Everyday Discrimination Scale, Parental Authority Questionnaire, Revised Family Communication Patterns Questionnaire, and a demographic form. Parent and adolescent questionnaires were completed separately in Spanish or English. Data were analyzed using Pearson correlation coefficients, hierarchical multiple regression, and moderation analysis in SPSS version 29.

Findings: The final hierarchical regression model significantly predicted intergenerational cultural conflict and explained 70.9% of its variance, F(17, 294) = 42.13, p < .001. Heritage-culture acculturation gaps, mainstream-culture acculturation gaps, language-brokering burden, perceived discrimination, authoritarian parenting, permissive parenting, and conformity orientation significantly predicted greater conflict. Authoritative parenting and conversation orientation significantly predicted lower conflict. Conversation orientation was the strongest protective predictor, whereas the heritage-culture acculturation gap was the strongest positive predictor. Conversation orientation significantly weakened the association between acculturation gaps and conflict, while authoritarian parenting significantly strengthened the relationship between perceived discrimination and conflict. Interactions involving adolescent birthplace and gender were not statistically significant.

Conclusion: Intergenerational cultural conflict in Ecuadorian immigrant families is shaped by the combined effects of cultural distance, adolescent translation responsibilities, discriminatory experiences, parenting practices, and communication processes. Open family communication and authoritative parenting may protect families from conflict, whereas rigid parental control, conformity-oriented communication, and burdensome language brokering may increase vulnerability. Culturally responsive family interventions should therefore emphasize dialogue, flexible parental authority, recognition of adolescents’ bicultural experiences, and reduced dependence on children for complex interpretation responsibilities.

Parental Phubbing and Adolescent Loneliness: A Longitudinal Cross-Lagged Panel Model of Emotional Availability, Parent–Child Attachment, and Problematic Social Media Use

Angela Villanueva iD , Samantha Clarke ^* iD , Elena Stoyanova iD

Objective: This study aimed to examine the reciprocal longitudinal relationship between parental phubbing and adolescent loneliness and to test parental emotional availability, parent–child attachment, and problematic social media use as mediating mechanisms.

Methods and Materials: A three-wave longitudinal study was conducted with 600 Canadian adolescents aged 12–17 years, recruited from public secondary schools in Ontario, British Columbia, Alberta, and Quebec. Data were collected at baseline, six months, and 12 months using standardized measures of parental phubbing, adolescent loneliness, parental emotional availability, parent–child attachment, and problematic social media use. A cross-lagged panel model was estimated using robust maximum-likelihood procedures and full-information maximum likelihood for missing data. Age, gender, family structure, socioeconomic status, daily social media use, and baseline mental health treatment were controlled.

Findings: The final model demonstrated good fit, χ²(1249) = 2024.51, CFI = .951, TLI = .946, RMSEA = .032, and SRMR = .046. Parental phubbing significantly predicted subsequent loneliness, β = .17, p < .001, whereas loneliness predicted later parental phubbing more weakly, β = .06, p = .032. Parental phubbing predicted lower emotional availability, β = −.22, weaker attachment, β = −.19, and greater problematic social media use, β = .15, all p < .001. Emotional availability, attachment, and problematic social media use significantly predicted later loneliness. The total indirect effect of parental phubbing on loneliness was significant, β = .101, p < .001, with relational pathways stronger than the behavioral pathway.

Conclusion: Parental phubbing appears to contribute prospectively to adolescent loneliness primarily by reducing perceived emotional availability and attachment security and, secondarily, by increasing problematic social media use.

Cultural Values, Gender-Role Beliefs, and Marital Satisfaction Among Dual-Earner Couples: An Actor–Partner Interdependence Structural Equation Model of Work–Family Conflict and Perceived Partner Support

Patrick O’Sullivan iD , Allison Wong ^* iD

Objective: This study aimed to examine the actor, partner, direct, and indirect associations of cultural values and gender-role beliefs with marital satisfaction through work-to-family conflict, family-to-work conflict, and perceived partner support among Canadian dual-earner couples.

Methods and Materials: This cross-sectional correlational study included 250 different-gender dual-earner married couples, comprising 500 participants residing in Canada. Both spouses independently completed validated questionnaires assessing horizontal and vertical individualism and collectivism, gender-role beliefs, work-to-family conflict, family-to-work conflict, perceived partner support, and marital satisfaction. Data were analyzed using confirmatory factor analysis and Actor–Partner Interdependence Structural Equation Modeling. Measurement invariance across wives and husbands was examined, and indirect effects were tested using 5,000 bias-corrected bootstrap resamples.

Findings: The structural model demonstrated acceptable fit, χ²(872) = 1517.46, CFI = .943, TLI = .936, RMSEA = .038, and SRMR = .047. Horizontal collectivism significantly predicted greater perceived partner support through actor and partner effects. Vertical individualism predicted greater work-to-family conflict, lower perceived partner support, and lower marital satisfaction. Egalitarian gender-role beliefs predicted lower work-to-family conflict and greater perceived partner support for both spouses. Work-to-family conflict negatively predicted marital satisfaction through significant actor and partner pathways, whereas perceived partner support positively predicted marital satisfaction and produced the strongest direct effects. Significant bootstrapped indirect effects showed that work-to-family conflict and perceived partner support partially mediated the associations of cultural values and gender-role beliefs with marital satisfaction. The model explained 58.1% of the variance in wives’ marital satisfaction and 53.4% in husbands’ marital satisfaction.

Conclusion: Marital satisfaction among dual-earner couples is shaped by interdependent cultural, ideological, occupational, and relational processes. Egalitarian gender-role beliefs and horizontal collectivism appear to strengthen marital functioning by reducing work–family strain and increasing perceived partner support, whereas vertical individualism may undermine satisfaction through heightened conflict and reduced relational support.

Perceived Cultural Stigma and Professional Help-Seeking Intentions Among Families of Children with Mental Health Difficulties: A Multigroup Mediation Model Across Urban and Rural Communities

Álvaro Montoya iD , Mark Robertson ^* iD , Lukas Brandner iD

Objective: This study aimed to examine the direct and indirect association between perceived cultural stigma and professional help-seeking intention through family social support and determine whether this mediation model differed between urban and rural Canadian communities.

Methods and Materials: This cross-sectional study included 648 parents and primary caregivers of children aged 6–17 years with clinically significant mental health difficulties, comprising 356 urban and 292 rural participants from six Canadian provinces. Participants completed measures of perceived cultural stigma, family social support, professional help-seeking intention, child psychological difficulties, and demographic characteristics. Confirmatory factor analysis and measurement invariance testing preceded multigroup structural equation modeling. Indirect effects were evaluated using 5,000 bias-corrected bootstrap samples, with caregiver, child, clinical, and regional variables controlled.

Findings: Perceived cultural stigma significantly predicted lower family social support (β = −.41, p < .001) and weaker professional help-seeking intention directly (β = −.27, p < .001). Family social support significantly predicted stronger help-seeking intention (β = .39, p < .001). The indirect effect of stigma through family support was significant (β = −.16, 95% CI [−.33, −.19]), indicating partial mediation. The stigma-to-support pathway was significantly stronger among rural than urban caregivers (β = −.52 vs. −.33, p = .009). The indirect effect was also stronger in rural communities (β = −.18 vs. −.15, p = .036), whereas the support-to-help-seeking and direct stigma-to-help-seeking pathways did not differ significantly between groups.

Conclusion: Perceived cultural stigma may discourage caregivers from seeking professional care for their children partly by weakening family support, particularly in rural communities, highlighting the need for culturally responsive and family-centered interventions.

Explainable Machine Learning Prediction of Adolescent Depression From Family Functioning, Parenting Practices, Peer Relationships, Cultural Values, and Digital Media Behaviors

Mathieu Schmit iD , Emily Chen ^* iD , Tanja Kralj iD

Objective: This study aimed to develop, validate, and explain a machine-learning model for predicting elevated depressive symptoms among adolescents using family functioning, parenting practices, peer relationships, cultural values, and digital-media behaviors.

Methods and Materials: A cross-sectional predictive study was conducted with 1,000 Canadian adolescents aged 13–18 years who were recruited through stratified multistage cluster sampling. Depression was assessed using the Patient Health Questionnaire for Adolescents, while validated instruments measured family functioning, parenting practices, peer attachment, cultural-value conflict, problematic social-media use, and related behavioral factors. The dataset was divided into development and independent holdout samples. Elastic-net logistic regression, random forest, gradient-boosted decision trees, extreme gradient boosting, support-vector machines, and multilayer perceptron models were compared. Model performance was evaluated using ROC-AUC, precision-recall AUC, sensitivity, specificity, balanced accuracy, F1 score, and Brier score. SHapley Additive exPlanations were used to determine global feature importance, directional effects, nonlinear thresholds, and predictor interactions.

Findings: Extreme gradient boosting demonstrated the strongest predictive performance, achieving a ROC-AUC of .879, precision-recall AUC of .756, sensitivity of .842, specificity of .762, balanced accuracy of .802, F1 score of .723, and Brier score of .132. It outperformed elastic-net logistic regression and all other candidate algorithms. Poor family functioning was the most influential predictor, followed by peer alienation, problematic social-media use, nighttime device use, cybervictimization, inconsistent discipline, and cultural-value conflict. Positive parenting, longer sleep duration, trusted-friend availability, and supportive online-community engagement reduced predicted risk. Important interactions were observed between family dysfunction and peer alienation, problematic social-media use and nighttime device use, and cybervictimization and trusted-friend availability. Model performance remained stable across alternative depression cutoffs and major demographic subgroups.

Conclusion: Explainable machine learning can identify adolescent depression risk with good discrimination and calibration by integrating relational, cultural, and digital-behavioral factors, while SHAP analysis provides clinically meaningful explanations of individual and global predictions.

Predicting Fertility Intentions Among Young Adults Using Machine Learning: The Relative Importance of Family Expectations, Housing Insecurity, Gender Ideology, Career Uncertainty, Relationship Quality, and Cultural Values

Katarína Horváthová iD , James O'Connor ^* iD

Objective: This study aimed to predict positive fertility intentions among young adults in Germany and determine the relative importance of family expectations, housing insecurity, gender ideology, career uncertainty, relationship quality, and cultural values.

Methods and Materials: A cross-sectional predictive study was conducted with 1,284 adults aged 18–35 years residing in Germany. Participants completed measures of fertility intentions, family expectations, housing insecurity, gender ideology, career uncertainty, relationship quality, and cultural values. The dataset was divided into an 80% training set and a 20% independent test set. Logistic regression, elastic-net regression, random forest, extreme gradient boosting, support vector machine, and multilayer perceptron models were developed using nested stratified cross-validation. Model performance was evaluated through the area under the receiver operating characteristic curve, precision–recall area, balanced accuracy, sensitivity, specificity, F1 score, and Brier score. Predictor importance was assessed using permutation importance and Shapley additive explanation values.

Findings: Extreme gradient boosting produced the strongest predictive performance, with an area under the receiver operating characteristic curve of .846, precision–recall area of .799, balanced accuracy of .784, sensitivity of .764, specificity of .803, F1 score of .753, and Brier score of .159. Its discrimination was significantly greater than that of standard logistic regression and elastic-net regression. Relationship quality was the most influential predictor domain, accounting for 24.8% of total SHAP importance, followed by housing insecurity at 21.6%, cultural values at 18.9%, career uncertainty at 15.7%, family expectations at 11.6%, and gender ideology at 7.4%. Severe housing and career uncertainty reduced the probability of positive fertility intentions, whereas high relationship quality and family-centered cultural values increased it. Family expectations showed a nonlinear pattern, and gender ideology was more influential when combined with supportive relationships and expectations of shared caregiving.

Conclusion: Fertility intentions among young adults in Germany are best predicted through an integrated assessment of relational, residential, cultural, and occupational conditions, with relationship quality and housing security representing the most influential domains.

Journal of Psychosociological Research in Family and Culture

About the Journal

Current Issue

Articles

Parental Phubbing and Adolescent Loneliness: A Longitudinal Cross-Lagged Panel Model of Emotional Availability, Parent–Child Attachment, and Problematic Social Media Use

Cultural Values, Gender-Role Beliefs, and Marital Satisfaction Among Dual-Earner Couples: An Actor–Partner Interdependence Structural Equation Model of Work–Family Conflict and Perceived Partner Support

Perceived Cultural Stigma and Professional Help-Seeking Intentions Among Families of Children with Mental Health Difficulties: A Multigroup Mediation Model Across Urban and Rural Communities

Explainable Machine Learning Prediction of Adolescent Depression From Family Functioning, Parenting Practices, Peer Relationships, Cultural Values, and Digital Media Behaviors

Predicting Fertility Intentions Among Young Adults Using Machine Learning: The Relative Importance of Family Expectations, Housing Insecurity, Gender Ideology, Career Uncertainty, Relationship Quality, and Cultural Values

Silenced at Home, Visible Online: A Digital Narrative Study of How LGBTQ+ Young Adults Negotiate Family Belonging, Cultural Norms, Concealment, and Online Social Support

Reconstructing Fatherhood After International Migration: A Narrative Inquiry into Masculinity, Emotional Expression, Parenting Practices, and Transnational Family Obligations

Early Identification of Families at Risk of Child Maltreatment: A Fairness-Audited Machine-Learning Model Integrating Caregiver Stress, Social Isolation, Economic Hardship, Family History, and Neighborhood Disadvantage

Language

Issues

Information

Keywords

Latest publications

Make a Submission

Cover

Editorial Board

Journal Statistics Summary

Predicting Suicidal Ideation Among Youth Using Family, Cultural, School, and Digital-Ecology Variables: A Stacked Ensemble Model with Calibration, Explainability, and Subgroup Fairness Assessment

Preliminary Evidence on Moral Underpinnings of Rape Victim Blaming in Iran: Replication and Extensions of Milesi et al. (2020)

Predicting Intergenerational Cultural Conflict in Ecuadorian Immigrant Families Based on Acculturation Gaps, Language Brokering, Perceived Discrimination, Parenting Styles, and Family Communication Patterns

Parental Phubbing and Adolescent Loneliness: A Longitudinal Cross-Lagged Panel Model of Emotional Availability, Parent–Child Attachment, and Problematic Social Media Use

Cultural Values, Gender-Role Beliefs, and Marital Satisfaction Among Dual-Earner Couples: An Actor–Partner Interdependence Structural Equation Model of Work–Family Conflict and Perceived Partner Support

Perceived Cultural Stigma and Professional Help-Seeking Intentions Among Families of Children with Mental Health Difficulties: A Multigroup Mediation Model Across Urban and Rural Communities

Explainable Machine Learning Prediction of Adolescent Depression From Family Functioning, Parenting Practices, Peer Relationships, Cultural Values, and Digital Media Behaviors

Predicting Fertility Intentions Among Young Adults Using Machine Learning: The Relative Importance of Family Expectations, Housing Insecurity, Gender Ideology, Career Uncertainty, Relationship Quality, and Cultural Values