Health and educational outcomes for children in single-parent families

Children From Single Parent Homes and the Effect on the Child

Single parent families comprise a significant portion of the American population, and so it is important to identify the specific needs of this group. Studies which have been conducted indicate that children from single parent families are at an increased risk of mental and physical ill health, and are at increased risk of some specific problems including schizophrenia and psychosis. In addition, children from single parent families are also at increased risk of substance abuse and some other educational and behavioral problems. The large number of studies, across different countries, using different data sets and statistical methods, which give similar conclusions indicates the validity of the data. This therefore supports the theory that policy makers need to take account of there being health and educational needs which are specific to single parent families.

Introduction

Single parent families comprise a significant proportion of the U.S. population, and as such it is important to identify specific needs for this population when forming health and social policies. There is a large volume of evidence available on the adverse outcomes associated with raising children in single parent families (e.g. Weitoft et al., 2003), and this area was therefore selected as it offered a wealth of information for discussion of quantitative statistical methods. This essay examines the literature which is currently available in relation to adverse outcomes for children from single parent families. In particular, three quantitative studies are discussed, with examination of the statistical methods used. The discussion includes consideration of how the methodologies may support the findings. The overall findings in relation to the study hypothesis are then presented in the conclusion to the study.

Statement of the Problem

It is currently debated whether the number of single parent families in the U.S. is on the rise, or whether it has stabilized (O'Hare, n.d.). It is clear from census data that regardless of whether the number is on the increase, single parent families currently comprise a substantial portion of the American population. It is therefore important that there is substantial information available for policy-makers to allow for consideration of the specific needs of this group. A particularly important element of this consideration is in both health and education policy, as it is possible that the needs of children from single parent families may be somewhat different to the needs of children from two parent families.

If policy makers are to take account of the specific needs of children from single parent families, then there must be information available to allow for identification of these needs. The literature available currently discusses many issues of health and education with the supposition that adverse outcomes in these areas may be associated with being raised in a single parent home. It is therefore important to use this research to identify the specific adverse outcomes for which those from single parent homes may be at increased risk. It is anticipated that these will include multiple health and educational outcomes.

Findings

The Swedish study by Weitoft et al. (2003) found that in particular children from single parent families were at increased risk of psychiatric disease, attempted or actual suicide, injury and addiction. Specifically, the study found that the relative risk for psychiatric disease was 2.1 for girls and 2.5 for boys. Also the relative risk for attempted suicide was 2 for girls and 2.3 for boys. The relative risk for alcohol-related disease was 2.4 for girls and 2.2 for boys, while that for narcotic-related disease was 3.2 for girls and 4.0 for boys. Overall, it was also found that boys from single parent families had a higher raised risk of mortality than girls.

Wicks et al. (2005) also found that children from single parent families in Sweden were at increased risk of psychiatric disease, specifically schizophrenia and psychosis. This study found that the overall relative risk was 1.9 for those from single parent families. Kuntsche and Silbereisen (2004) also found that there was an increased risk of substance abuse in those children from single parent families. In particular, the study showed that there was some evidence of a difference in the level of drunkenness between single parent and two-parent families, but only in girls not boys (girls: t = 3.04, p .01). In contrast to this, there was some evidence of a difference in the level of tobacco use for boys and string evidence for girls (girls: t = 4.12, p

Discussion

The Swedish study by Weitoft et al. (2003) used a very large sample size, with over 65-000 children from one parent families and over 920-000 children from two parent families included as a control. This large sample size was possible as the data used was taken from the 1990 census. This would therefore indicate that the results of the statistical analysis would have more power than those using a small sample. This would then therefore indicate that these results were more accurate. In addition, the results of this study also had confounding factors removed, including socioeconomic status and parental history of mental illness and addiction. This would also increase the accuracy of the association between observed increase in risk and number of parents in the household. An important aspect of this study was the separation of the analysis by gender. As it may be seen through the difference in relative risk for each outcome between genders, this is important in determining the exact extent of the impact. The results which were obtained in this study were achieved through multivariate regression analysis, with logistical regression used where categorical variables were included. Different models were tested and the most appropriate selected based on the best exclusion of confounding variables. This method was the most appropriate given the presence of confounding variables, although it would possibly have been beneficial to then confirm these results using non-parametric methods such as the chi-squared test.

The study by Wicks et al. (2005) supports the work by Weitoft et al. (2003) as it too shows that children from single parent families are at increased risk of mental illness. The relative risk ratio was slightly lower than that found by Weitoft et al. (2003) although this could be due to methodological differences. The study also used census data from Sweden, although a much larger sample was used with data from 1970-1990 used instead of only one census, resulting in over 2.1 million participants. There were also different confounding variables, with factors such as foreign-born parents included alongside the other variables considered by Weitoft et al. (2003). Aside from this, similar statistical methods were used, with logistic regression to account for the presence of both numerical and categorical variables. It would have been interesting to see if this study also showed a gender-based difference, but this was not considered.

The study by Kuntsche and Silbereisen (2004) also supports the work of Weitoft et al. (2003) as it shows that there is evidence of an increased level of substance abuse in those from single parent families. The data which was used in this study was taken from the Health Behaviour in School-Aged Children (HBSC) which is a survey conducted in 30 European countries. Unlike the study by Weitoft et al. (2003) this risk is not quantifies however, as t-tests and X2 tests were applied rather than regression analysis. This method presents only evidence of whether there is a difference in the mean of two groups, so in this case showed there to be a difference in the mean level of substance abuse in children from one and two parent families. This type of analysis is a simpler method than regression, and may be ample to answer the research question of whether there is increased risk for those in single parent families. It would also appear to be able to offer some insight as to whether there was a gender-based difference in substance abuse. This would however require confirmation using a one-tailed t-test between the mean for boys and girls from single parent families. The main drawback from this study's approach to the statistical analysis is that it does not allow for removal of confounding variables. This means that as there may be an association between single parent families and other socioeconomic variables, if there was an association between these and substance abuse then this may interfere with the results.

A further aspect which has not been considered in any of the three studies discussed is the reason behind the single parent status of the family. For example Biblarz and Gottainer (2004) found that adverse outcomes in terms of educational attainment may be more prevalent in families where the mother was divorced in contrast to those where she was widowed. It is possible that there may be similar effects observed across all adverse outcomes, although this is not clear from the three studies discussed.

Conclusion

Given the wealth of information available in the area, there would appear to be strong support that children raised in single parent families are at increased risk of a range of adverse outcomes. The fact that there are numerous different statistical methods and data sets used in these different studies lends further support to this hypothesis. In particular, there appear to be a significantly higher risk of mental and physical ill health, substance abuse, educational and behavioral problems. In particular, there also appear to be gender-based differences for most of these outcomes. It is important that these risks are accounted for when policy is made, as it may allow this group to be targeted by specific policies to lower the risk.