Discussion
The current study provided a retrospective yet longitudinal look at identified risk factors of childhood problem behaviors. The study sought to test a conceptual model based on early childhood risk factors identified by an exhaustive review of the literature. In the study, the sample proved to resemble a middle class, not-at-risk population. Using this sample, the model was only moderately (62%) successful in identifying children who have problem behaviors at school age and had a propensity to miss children with significant behavior problems.
Even in this non-clinical population, significant predictors of later behavior problems were identified in very young children. Gender, temperament, and parenting ability in early childhood were significant predictors of behavior problems in later school age. In addition, the transaction of these predictors dramatically increased the risk of later behavior problems for a young toddler. A young toddler having all three predictors or risk factors, male gender, early difficult temperament and a less able parent was eight times more likely to have behavior problems at school age.
This research corroborates the growing body of evidence describing the transactional nature of child development in which the bidirectional affects of individual risk factors are identified (Feldman, Eidelman, & Rotenberg, 2004; Robison, Frick, & Sheffield, 2005; West & Newman, 2003). For example Barry and colleagues (2005) found maternal mental health was significantly related to behavior problems in boys even when controlling for SES and income. In contrast, Wake and colleagues (2006) found that negative infant behavior was significantly related to poor maternal mental health. Our finding can be better understood by examining the characteristics of the study sample in comparison to previous research and by examining current thinking in measuring socioeconomic factors, particularly poverty, in children.
The current study sample consisted of primarily middle income families that did not appear to have characteristics generally thought to be associated with poverty or low socioeconomic status. For example, almost 50% of the mothers in this sample reported some college education and 54% of families were two-income households with a median income of $38,777.00, well above the poverty line in 1994 (U.S. Census Bureau, 2000). Even though this sample appeared to have few of the well-documented risk factors, child gender, temperament, and parenting ability were still significant predictors of behavior problems, indicating that these variables cross socioeconomic boundaries.
Given the presence of significant predictors even in this middle class sample, more research is needed to identify and meet the needs of all families, not just families with known risk factors. Much attention has been given to young, single, urban, poor mothers, and virtually no attention has been given to low income or middle class mothers living in rural areas. Much attention has been given to the role of poverty and community violence in urban American, and little or no attention has been given to families who make just enough money to get by living in suburban America.
National data sets have inherent limitations as do secondary analyses. The major disadvantage of secondary analyses is that research questions are being asked of a data set that was not collected to specifically answer that question. The researcher has no control over original data collection and is dependent on the original researcher's decisions about data collection and storage (Magee, Lee, Giuliano, & Munro, 2006). In addition to the limitations associated with large data sets and with secondary analyses, this research has two major limitations. The first limitation is in the measurement of temperament. Studies of very young children require parental report of temperament and of the outcome; therefore, the measures are not independent of each other. In the current study, temperament was assessed for negativity or difficulty during the first 2 years of life, and behavior problems were assessed between 6 to 9 years of age. The length of time between each measure increases the chance that the measures are independent of each other, but given the bidirectional effects of parenting and temperament, independence is not assured.
A second limitation of this study is the under-representation of families from lower income levels. Previous research suggests that environment, particularly the aspect of the home environment, which is influenced by income, is significant to child development and to parent-child interactions. All income levels need to be well represented in research concerning children and parents
Given that previous research has highlighted the importance of environment on the developing child (Caughy & O'Campo, 2006) and on the development of a parent (Beeber & Shandor, 2003; Brooks-Gunn, Klebanov, Smith, & Lee, 2001), further research is needed to more fully understand how the home environment acts or interacts on both the child and the parent. For example, does the home environment mediate or moderate the effects of a negative community environment? What factors must be present in the home to promote child development and enhance parenting ability? Given the unique role of poverty on child behavior and development (Eamon, 2000, 2001), more research is needed to describe the role of family income on the child, the parent, and the environment. To suggest that only the child who lives in poverty are at risk for negative behavioral and developmental outcomes contradicts anecdotal evidence and clinical experience. Finally, while research points to the negative implications of poverty, children who live in poverty may have positive experiences associated with poverty.
Pediatr Nurs. 2008;34(1):37-44. © 2008 Jannetti Publications, Inc.
Cite this: Predicting School-Age Behavior Problems: The Role of Early Childhood Risk Factors - Medscape - Jan 01, 2008.
Comments