Examining 9 Forms of Parental Discipline across 60 Low- and Middle-Income Countries: A Bayesian Multilevel Analysis

Background and Purpose: Parental discipline can serve as a mechanism for social change, as parents across cultures rely on different forms of discipline to raise children, which can have long-term consequences for children’s wellbeing. While research suggests that physical and psychological punishment are harmful for children, non-aggressive discipline has been less studied, especially in low- and middle-income countries (LMICs). Additionally, although parents tend to utilize numerous forms of discipline, few studies examine multiple forms of discipline simultaneously. This study explores the association of 9 forms of discipline on child aggression across 60 LMICs.

Methods: Data came from the 4th and 5th rounds of UNICEF Multiple Indicator Cluster Surveys (N=235,064). Parental discipline was measured using the UNICEF-modified Parent-Child Conflict Tactics Scale and asked whether the respondent or anyone in the household used each form of discipline in the past month (0 = no, 1 = yes). Psychological punishment included 1) shouting and 2) calling the child names. Physical punishment included 3) shaking, 4) spanking on the bottom, 5) hitting or slapping on the arm or leg, 6) hitting with an object, 7) hitting or slapping on the face, head, or ears. Non-aggressive discipline included 8) taking privileges away and 9) explaining why the behavior was wrong (i.e., verbal reasoning). Child aggression was measured with one question from the Early Childhood Development Index, which asked respondents whether the child kicks, bites, or hits other children or adults (0 = no, 1 = yes). Control variables included child age, child sex, household wealth score (standardized within each country), number of household members, mothers’ and fathers’ education (0 = none [comparison], 1 = primary, 2 = secondary-plus), whether the respondent believes children need physical punishment to be raised properly (0 = no, 1 = yes), and community (0 = urban, 1 = rural). Multilevel logistic regression was run in R using Bayesian estimation, providing odds ratio (OR) coefficients.

Results: All psychological and physical forms of punishment were associated with child aggression. Shouting (OR = 1.20, 95% CI [1.13, 1.28]); calling the child names (OR = 1.30, 95% CI = [1.21, 1.39]); shaking (OR = 1.21, 95% CI [1.14, 1.28]); spanking (OR =1.12, 95% CI = [1.06, 1.19]); hitting on the arm/leg (OR = 1.12, 95% CI = [1.06, 1.20]); and hitting on the face, head, or ears (OR = 1.25, 95% CI = [1.15, 1.35]) were associated with an increased odds of child aggression. Regarding non-aggressive discipline, taking away privileges was not associated with child aggression (OR = 1.03, 95% CI = 0.99, 1.08), while verbal reasoning was associated with a decreased odds of child aggression (OR = 0.92, 95% CI = [0.85, 0.99]).

Conclusions and Implications: Results suggest that physical and psychological punishment are associated with increased child aggression. Accounting for all forms of discipline simultaneously, calling the child names was associated with the greatest odds of child aggression. Parental interventions conducted in LMICs may benefit from advising parents against physical and psychological punishment, and promote non-aggressive forms of discipline, particularly verbal reasoning.