Mental health assessments are often time-consuming and costly (Charlson et al., 2019), but questionnaires can offer a cost-effective way to assess mental health. While these questionnaires can be completed by different assessors (e.g. children [‘self-report’], parents, teachers), most surveys collect data from only one respondent.
To the best of our knowledge, no studies in Australia have evaluated whether the assessment of children’s mental health does actually vary depending on the type of assessor answering the questionnaire. The only study that investigated this issue was conducted by Johnston and colleagues (2014) in the United Kingdom. Using data from Australia, this study investigated whether (1) children’s mental health and its relationship to income vary depending on who reports children’s mental health and (2) whether the reporting variation depends on income.
Exploring the issue of assessor variation will increase understandings of mental health measures in survey data. This knowledge, combined with improved understandings of the impact of parental income on children’s mental health, could inform policies to improve the wellbeing of Australian children.
What did the study do?
The study analysed data from the Longitudinal Study of Australian Children (LSAC) to investigate whether family income affects children’s mental health – and whether the effect, if any, varies across different kinds of informants.
In the Longitudinal Study of Australian Children (LSAC), children’s mental health was evaluated by:
- parents
- teachers; and
- children (i.e. self-report).
The current study focused on data for children aged 10-15 years as younger children were unable to answer the relevant questionnaire.
The study assessed children’s mental health using the Strength and Difficulties Questionnaire (SDQ) (Goodman & Goodman, 2009), which is widely used in the literature (Khanam and Nghiem, 2016; Khanam and Nghiem, 2018). To ensure that results were comparable with the only known previous study on this topic (i.e. Johnston et al., 2014), the current study focused on the following three SDQ categories:
- Hyperactivity (e.g. restlessness, being over-active, constantly fidgeting or squirming, being easily distracted).
- Emotional symptoms (e.g. having many worries; often seeming worried; often appearing unhappy, down-hearted or tearful; acting nervous or clingy in new situations; complaining frequently of headaches, stomach aches or sickness).
- Conduct problems (e.g. frequent temper tantrums or displays of a hot temper; often fighting with other children or bullying them; often acting argumentative with adults).
Family income was represented by weekly parental income. To help isolate the relationship between family income and child mental health, the authors controlled for various covariates (i.e. confounding variables), including child demographics (i.e. age, sex, ethnicity, number of siblings, language spoken at home), parents’ education level, and parental health.
What were the main findings?
The study found that mental health measures varied across assessors. The main findings included:
- Mental health problems were rated as more severe when assessed by children themselves rather than parents or teachers.
- Parents tended to report more severe behaviour problems than teachers.
- Children rated themselves more negatively for internalising behaviours (e.g. emotional problems) compared to their teachers and parents.
Family income was found to have protective effects on children’s mental health. However, the effects of family income on child mental health in Australia was only about half of that in the UK.
In addition, the differences across categories of assessors reduced as family income increased. This finding is consistent with the aforementioned UK study (Johnson et al., 2014), but the effect of income on these reporting differences was smaller in Australia.
What are the implications for practice, research and policy?
For researchers, relying solely on parents’ assessments of children’s mental health (which often occurs in the literature) may lead to under-reporting of mental health problems.
Similarly, practitioners may wish to consider whether solely focusing on a parent’s report might not be as representative of the child’s mental health problems, and whether obtaining information from other informants (e.g. child self-report, teacher report) may help develop a more-comprehensive understanding of a child’s mental health.
Regarding policy implications, the findings suggest that financial supports for low-income families might help improve the mental health of Australian children. “Policy initiatives that improve affordability and accessibility of services in poorer communities can [also] contribute to breaking the cycle of hardship experienced by children growing up in disadvantaged financial circumstances” (Khanam, Nghiem, & Rahman, 2019, p. 11).
Increased child mental health services and supports should also be made available and accessible to teachers and parents in low socio-economic areas (e.g. mental health services in schools; student-level and classroom-level interventions). Educational initiatives that aim to improve mental health literacy may also promote better identification, understanding, and management of children’s mental health issues.
Finally, additional research is needed to replicate and ensure the generalisability of these findings. Similar studies should be conducted using data from other countries, and results should be compared between developed and developing countries.