preteenIn an escalating process across the first decade of life, children who are spanked go on to have more aggressive behavior, which in turn predicts increased spanking, a team of Columbia University researchers has found.

The study, which appears this week in the Journal of Youth and Adolescence, was co-authored by:

  • Dr. Michael MacKenzie, Associate Professor at Columbia School of Social Work (CSSW)
  • Dr. Eric Nicklas, Adjunct Assistant Professor (CSSW)
  • Dr. Jeanne Brooks-Gunn, Virginia and Leonard Marx Professor at Teacher’s College and the College of Physicians and Surgeons
  • Dr. Jane Waldfogel, Compton Foundation Centennial Professor (CSSW)

Go to article (with links to full text and pdf).

The researchers found at each age assessed that children who exhibited more behavioral problems went on to experience higher levels of spanking at the next assessment age, indicating that more difficult children may elicit more coercive forms of punishment from parents. Completing the other side of this bidirectional cycle, the researchers also found that spanking at each age predicted increased child aggression at later ages. These reciprocal effects build from the first year throughout the first decade of life. Lead author Michael MacKenzie said: “This is an important finding because amongst the many contributors to the decision to spank, including family and cultural values, we know that parental stress is an important factor for many, and parents with more challenging children need support to avoid this escalating pattern that can take hold.”  Results were consistent for boys and girls and across family racial/ethnic background.

The results highlight a bi-directional pattern between parent and child that starts early and amplifies across the first decade, suggesting that children would benefit if parents adopted more positive parenting behaviors. The findings are particularly important in the U.S. context given the relatively high rates of spanking in this country. Despite a growing research literature linking physical discipline with later child aggression, spanking remains a typical experience for American children. As MacKenzie noted:

These bidirectional findings are consistent with what has been found in the literature, but are of added importance given the detailed nature of the data we were using which allowed us to look across the entire first decade of life while controlling for a host of other factors that also affect children’s behavior, including their behavior at younger ages.

The findings were based on data on 1,874 families from the Fragile Families and Child Wellbeing Study (FFS), a population-based, birth cohort study conducted by researchers at Princeton and Columbia of children born in 20 large American cities between 1998 and 2000. Families were assessed shortly after giving birth and when the children were approximately 1, 3, 5, and 9 years old.

In their analysis, the researchers controlled for an extensive set of factors that might also influence spanking and children’s aggression, including:

  • other risk factors: e.g., late prenatal care, risky health behavior and substance use, intimate partner violence, paternal incarceration history, maternal IQ, parenting stress, and maternal depression and anxiety.
  • factors related to the child and family background: e.g., immigration status, educational attainment, employment status, family structure and income and child gender, child birth weight, and family size.

Taking advantage of the longitudinal data in the FFS, the researchers were able to control for children’s prior levels of aggression when looking at the influence of spanking on later aggression. And they were able to control for earlier spanking when looking at the influence of child aggression on later levels of spanking.

The Fragile Families Study is funded through grants from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), and a consortium of private foundations and other government agencies.

Related links:

Press coverage received:

Leave a Reply