Why did we do this study?

We did this study to improve our understanding of how different factors in a mother’s life—like her mental health, financial resources, relationship status, education, and parenting style—affect her child’s development.

What did we do?

  • We used data from 634, low-income families in Toronto.
  • Mothers reported on their mental health (i.e., depression, anxiety, stress), marital conflict, parenting styles, annual income, parent(s) employment status and their level of educational when their infant was approximately 12 months of age.
  • Mothers also reported on their infant’s temperament (i.e., anger and fear) when the infant was about one year old and the same child’s mental health (i.e., conduct problems, hyperactivity) and temperament when the child was about three years old.
  • We used the Patient Health Questionnaire to assess maternal depressive symptoms, the Generalized Anxiety Disorder-2 Item to assess anxiety symptoms, and the Perceived Stress Scale to assess maternal stress. Infant temperament was assessed using the Infant Behavior Questionnaire, and child mental health was assessed using the Strengths and Difficulties Questionnaire.

Patient Health Questionnaire: This questionnaire has 2 questions assessing symptoms of depression. Thinking of the last two weeks, questions are rated from 0 (not at all) to 3 (nearly every day).

Generalized Anxiety Disorder-2 Item: This questionnaire has 2 questions assessing symptoms of anxiety. Thinking of the last two weeks, questions are rated from 0 (not at all) to 3 (nearly every day). Total scores range from zero to six, with scores of three or higher indicating the presence of anxiety.

The Perceived Stress Scale: This questionnaire has 4 questions assessing the perception of stress in one’s life in the last two months. Questions are rated from 0 (never) to 4 (very often). Total scores range from 0 to 16, with scores of three or higher indicating the presence of stress.

The Parental Cognitions and Conduct Toward the Infant Scale: This questionnaire has 44 questions assessing minor and major marital conflict, including the frequency, severity, resolution, and efficacy of marital conflict. For this study, we only asked questions that assess how often partners have minor and major disagreements. 

The Conflicts and Problems Scale: This questionnaire has 28 questions assessing parental self-efficacy, perceived parental impact, parental hostile-reactive behaviors, and parental overprotection. For this study, we asked 2 questions assessing parental impact and 2 items assessing perceived parental hostile-reactivity

Infant Behavior Questionnaire: This questionnaire assesses temperament in children between the ages of one to three years. In this study, we included questions that assess fear and anger. Mothers rated the frequency of reactions over the last two weeks from 1 (never) to 7 (always). Higher scores indicate more difficult temperaments.

Strengths and Difficulties Questionnaire: This questionnaire has 25 questions assessing children’s mental health and behavioral problems, including emotional problems, conduct problems, hyperactivity, and peer problems. Mothers rated each question from 0 not true) to 2 (certainly true). Higher scores indicate higher problem behavior.

  • We analysed the data using Latent Class Analysis to identify groups of mothers who are similar in how they responded to questions about their reported mental health, marital conflict, parenting styles, annual income, parent(s) employment status and educational attainment, and infants’ temperament. We then used an Analysis of Variance to see if children of mothers in the different groups differed in their temperament and mental health at three years of age.
More information about Latent Class Analysis

Latent class analysis is a way of finding hidden groups or patterns in data. It sorts people into groups based on their answers to questions or behaviors, so people who are more similar are grouped together.

More information about Analysis of Variance

An Analysis of Variance (ANOVA) is a way of seeing if there are meaningul differences between the averages (means) of three or more groups. Essentially, it’s a way to check if the groups are truly different from each other.

What did we find?

From the Latent Class Analysis, we identified four groups of mothers.

Four groups of mothers we identified from the Latent Class Analysis.
Child mental health outcomes at 3 years.

What does this mean?

  • The method we used revealed different groups of moms and children
  • Even though this sample was of low-income families, it is still a spectrum, with some doing better financially than others
  • We saw that those on the lower end of the financial spectrum had the poorest mental health and more challenging children (the first group)
  • We also saw that some moms had more stable mental health and less difficult children (the fourth group), despite still facing low-income like the rest of the sample à something resilient about these moms!
    • Maybe it’s their higher level of education, of having a partner, but understanding what makes some people more resilient is important to understand
  • Parent mental health matters! Helping moms with mental health struggles could directly benefit their children’s development and behavior.