Custody Arrangements and the Development of Emotional or Behavioural Problems in Children - October 1998

3. Methodology

Previous

Contents

Next

The data examined are derived from the first cycle of the National Longitudinal Survey of Children and Youth, which was undertaken between the Fall of 1994 and the Spring of 1995 by Human Resources Development Canada and Statistics Canada. The NLSCY represents the first nationally representative, longitudinal study of the development of children in Canada. The sample for this cycle consists of 22, 831 children aged 0 to 11 years; the analysis in this paper focuses on the sub-sample of 18, 174 children who were between the ages of 2 and 11 years. The operational definitions of the variables in the analysis are outlined below.

3.1 Dependent Variable: Emotional or Behavioural Problems

Emotional and behavioural problems⁶ include the following: anxiety, emotional disorder, hyperactivity, inattention, conduct disorder, physical aggression and unsocial behaviour.⁷ The presence or absence of these problems was measured by means of indices calculated from responses to scales measuring each emotional or behavioural problem. Latent class analysis was employed to categorize each index into the following categories: child has the problem, child has somewhat of a problem, and child has no problem. Those children categorized as "child has the problem" on any one or more of the seven scales measuring the various problems (anxiety, emotional disorder, hyperactivity, inattention, conduct disorder, physical aggression and unsocial behaviour) are considered to have "one or more problems" for the purpose of the logistic regression analysis.

3.2 Independent Variables

Living Arrangement

The response categories for this variable are: "live with both parents" and "live in post-divorce/separation custody." When entered into the model examining only children living in post-divorce/separation custody, the response categories for types of custody arrangements are detailed, including "mother only," "father only," and "shared."

Age of the Child

The age of the child was measured in years. For the purpose of the descriptive analysis, it was recorded into categories. When entered into the logistic regression models, it was entered as a continuous variable.

Sex of the Child

The sex of the child was reported by the Person Most Knowledgeable about the child.

Education of the PMK

This variable is measured at the nominal level, with the following response categories: "less than secondary," "secondary school," "some post secondary," and "college or university degree."

Age of the PMK

Was reported in years; for the purposes of the descriptive analysis, it was recoded into categories. Age was left as a continuous variable when entered into the logistic regression models.

Household Income

Was measured using the following response categories: "under $10, 000," "$10, 000 to $14, 999," "$15, 000 to $19, 999," "$20, 000 to $29, 999," "$30, 000 to $39, 999," "$40, 000 to $49, 999," "$50, 000 and over. When entered into the logistic regression models, each category was recoded to its midpoint.

Number of Siblings

This was measured as a continuous variable, in both the descriptive and logistic regression analyses.

• ⁶See National Longitudinal Survey of Children and Youth user's Handbook and Microdata Guide (Human Resources Development Canada and Statistics Canada, 1997) for the indicators employed in each scale.
• ⁷Low scores on prosocial behaviour are considered an indication of unsocial behaviour for this analysis.