Predictors and Trajectories of Mothers' Nonviolent Discipline Use: A Longitudinal Multilevel Examination

Background and Purpose: Researchers and policymakers are increasingly promoting nonviolent discipline use across countries and cultures. In 2011, the United Nations Convention on the Rights of the Child (UNCRC) stated that caregivers need to embrace nonviolent forms of discipline, and that programs should be put in place to widely promote nonviolent discipline techniques. While some research has examined the effectiveness of nonviolent discipline, scant research has explored the predictors of nonviolent discipline or the trajectory of nonviolent discipline use across time. In order to promote nonviolent discipline, policymakers need to understand the factors that predict nonviolent discipline, as well as when mothers are most and least likely to use nonviolent discipline. This study examined mothers’ socioeconomic predictors of nonviolent discipline use, as well as mothers’ trajectory of nonviolent discipline across early childhood.

Methods: Data came from three waves of the Fragile Families and Child Wellbeing Study (FFCWS). The sample was restricted to mothers living with the child at least half the time (N = 2,898). Nonviolent discipline was measured using the Conflict Tactics Scales at child ages 3, 5, and 9. Use of nonviolent discipline in the past year consisted of four ordinal items (0=none, 1=once, 2=twice, 3=3-5 times, 4=6-10 times, 5=11-20 times, 6=more than 20 times): verbal reasoning (i.e., “explained why the behavior was wrong”), taking away privileges (i.e., “took away privileges''), time-out (i.e., “put child in ‘time out’”) and distraction (i.e., “gave [child] something else to do instead of what [child] was doing”). Time-varying socioeconomic status variables included educational attainment (1=less than high school [comparison], 2=high school, 3=some college, 4=college or greater), maternal-reported household income (continuous), and employment status (0=not employed, 1=employed). Longitudinal multilevel ordinal logistic regression analyses were conducted in Stata. Trajectories were compared using AIC and BIC.

Results: Maternal educational attainment was associated with an increased odds in all forms of nonviolent discipline (college or greater, verbal reasoning: OR = 3.10, p < .001; took away privileges: OR = 1.99, p < .001, time out: OR = 1.66, p < .001, distraction: OR = 3.00, p < .001). Household income was weakly associated with verbal reasoning (OR = 1.02, p = .020), time out (OR = 1.02, p = .043), and distraction use (OR = 1.02, p = .003). Employment status was only associated with verbal reasoning use (OR = 1.16, p = .016). All models were best fit as quadratic models, with mothers, on average, frequently using all forms of nonviolent discipline at child ages 3 and 5, and decreasing their use at child age 9.

Conclusions and Implications: Results suggest maternal education may be a stronger predictor of nonviolent discipline use compared to other socioeconomic measures. Future research should explore the mechanisms linking maternal education and nonviolent discipline use. Findings provide preliminary evidence of the important role of maternal education in the charge of ending violence against children. Interventions aiming to increase nonviolent discipline may need to target mothers of younger children, when mothers are needing to use these techniques more often.