Parenting Style Profiles and Child Quality of Life: A Latent Profile and Chain Mediation Analysis In Urban China

Background and Purpose: Child quality of life (QoL) has emerged as a critical public health concern in China, where rising rates of childhood obesity, type 2 diabetes, and mental health disorders signal converging threats to child well-being. While parenting style is widely recognized as a key determinant of child outcomes, prior research has relied predominantly on variable-centered approaches that obscure meaningful heterogeneity among families. Furthermore, the mechanisms through which distinct parenting profiles affect child QoL remain underexplored. Drawing on family systems theory and social learning theory, this study used a person-centered approach to identify latent parenting style profiles and examined the serial mediating roles of parent-child relationship quality and child self-esteem in linking parenting profiles to child QoL.

Methods: A school-based survey was conducted in November 2020 in a peri-urban area of Shanghai with roughly equal proportions of local and migrant residents. Data were collected from Grade 3–5 students via paper-based classroom questionnaires and from their parents via matched online surveys. After excluding unmatched dyads, 1,401 parent–child pairs were retained (response rate = 98.8%; mean child age = 10.74 years; 47.7% female). Parenting style was measured using the Alabama Parenting Questionnaire (APQ; three subscales: positive involvement, negative/ineffective discipline, and deficient monitoring). Child QoL was assessed with the PedsQL 4.0 Generic Core Scales; parent-child relationship with the Closeness to Parents Scale; and child self-esteem with the Rosenberg Self-Esteem Scale. Latent profile analysis (LPA) was conducted in Mplus 8.3 using the MLR estimator, with model selection guided by AIC, BIC, aBIC, entropy, and LMR/BLRT tests. Profile differences in child QoL were tested via the BCH method. Chain mediation was examined using structural equation modeling with 5,000 bootstrap resamples, controlling for child sex, age, household income, hukou status, and parental education.

Results: LPA identified four distinct parenting profiles: High Positive Involvement (53.08%; high positive engagement, low negative discipline, low monitoring deficiency), Mid-Balanced (29.47%; moderate on all dimensions), Low-Engagement (12.62%; low involvement across all dimensions), and Harsh-Neglectful (4.82%; low positive involvement, high negative discipline, high monitoring deficiency). Profile membership was significantly predicted by child sex and household income. Compared to the High Positive Involvement group, all other profiles were associated with significantly lower child QoL. Total effects on QoL were: High Positive Involvement (+9.419), Harsh-Neglectful (−10.148), Mid-Balanced (−7.471), and Low-Engagement (−3.063). Chain mediation analyses confirmed that parent-child relationship quality and child self-esteem serially mediated the parenting–QoL association across all profiles. Notably, the Low-Engagement profile showed a paradoxical pattern: a positive direct effect (+4.052) was substantially offset by a strong negative indirect effect (−7.116) operating primarily through damaged parent-child relationships.

Conclusions and Implications: Parenting style heterogeneity matters: even "moderate" or seemingly benign low-engagement parenting poses meaningful risks to child QoL through relational and self-esteem pathways. Social workers should prioritize family interventions that strengthen positive parental involvement and parent-child emotional connection, with targeted outreach to Harsh-Neglectful families and psychoeducation for Low-Engagement families who may misread disengagement as benign. The Mid-Balanced group — nearly one-third of families — represents the highest-yield population for preventive parenting programs.