Parental Divorce and Academic Performance in Preterm Children

Study Overview and Variables

In the study by Nusinovici et al. (2018), the researchers used a quantitative method to assess the impact of divorce or separation of parents on their preterm children's academic performance at 5 years of age. "Preterm" was defined as birth before 35 weeks of gestation. The main variables included were: (1) the subject had to be a preterm baby, (2) the subject's parents had to be separated or divorced at some point prior to the child's fifth year — the point at which children typically enroll in kindergarten — and (3) the child had to be in school and able to be assessed by a teacher. The purpose of the study was to determine whether any correlation existed between premature birth, parental divorce or separation, and the child's academic achievement at age 5.

The method used was purposive sampling, drawing on survey data of "clinical anomalies associated with preterm births" (Nusinovici et al., 2018, p. 2). Divorce and separation timing was classified into two groups: children whose parents separated before 36 months of the child's birth, and children whose parents separated after 36 months but before 60 months. Gestational periods further classified subjects into three groups, ranging from 24 to 27 weeks up to 32 to 34 weeks of gestation. To assess the children academically, the Global School Adaptation (GSA) questionnaire was administered to teachers, evaluating both behavior and academic ability among the subjects.

The main findings showed that among preterm children, those who experienced parental divorce or separation between the ages of 3 and 5 tended to score lower on the GSA and perform academically poorer overall than preterm children whose parents had separated before the child turned 3, or whose parents were not separated at all. The implications were that preterm children whose parents separate may require additional academic and social support during the transition to formal schooling.

Sample Size and Research Design

The sample size of this study was suitably large, with over 3,300 babies included. The longitudinal study monitored these children from 2003, when births began to be recorded, through 2011, when the final year of academic performance analysis was conducted. The study found that more than 7,000 preterm babies were born between 2003 and 2011; more than a thousand were excluded because they showed a lack of adequate neurodevelopment, which would have made them outliers and skewed the results. Of the remaining children, 70% were evaluated using the GSA. All subjects came from the LIFT network of maternity wards in the Pays-de-la-Loire region of France. The sample thus provides a suitable size for the study and is defined clearly enough that the study could be replicated to verify its validity and reliability.

No formal research model was used in this study, as it was exploratory research with an empirical aim — namely, to provide quantitative data on the academic abilities of preterm children at age 5 whose parents had separated or divorced. Because the study was exploratory in nature, it did not include a hypothesis or a research model; it was descriptive, which was appropriate for its stated purposes. Had a research model been used, a conceptual model would most likely have been appropriate, potentially based on findings from a case study using a similar subject population — albeit with a much smaller sample size, as case studies are generally not conducive to large samples — from which a degree of generalizability could be obtained to support the model's construction.

A more rigorous approach would have been to formulate a hypothesis and then conduct a randomized controlled trial to test it. This could have been feasible, as there was nothing particularly surprising about these findings; they could have been hypothesized based on an assessment of the relevant available literature. With a clear hypothesis in place, the researchers could have controlled for various factors and directly tested their predictions rather than conducting what might be viewed as unnecessary exploratory fieldwork.

Validity and Methodological Limitations

The research does demonstrate a degree of validity in that it measures what it sets out to measure. It describes its measurement instrument — the GSA — as a standard tool used in France by the French Ministry of Education to examine student achievement in public schools, meaning its validity has already been externally established. However, a significant limitation of this study's approach is that it takes into consideration only three variables: the preterm gestation period, the parental situation of the children (whether divorced, separated, or still together), and the behavior and academic ability of the 5-year-old child.

There are several problems with this limited scope. First, divorce is common in families today, with roughly half of all marriages ending in divorce. Second, most preterm babies do not suffer lasting neurodevelopmental issues. Third, many factors are likely to influence a child's behavior and academic ability at age 5. While it may be possible to identify a correlation between parental divorce and poor academic ability — as this study does — correlation does not imply causation. The study examines only these few variables and does not control for others. Because this study is exploratory rather than experimental, its validity is inherently limited.

The question of internal validity is especially relevant here. Without controlling for confounding variables such as household stress, financial instability, changes in residence, or parenting arrangements, it is difficult to attribute the observed academic differences specifically to the timing of parental separation. The study's design, while appropriate for initial exploration, leaves open the question of what is actually driving the outcomes observed.