The Relationship Between Geographic Relocation and Childhood Problem Behaviour - October 1998

Several theories have been offered to explain the potentially harmful impact of frequent moving on child and adolescent behaviour. Our choice of a guiding theoretical framework in the study of childhood relocation is Travis Hirschi's (1969) Social Control Theory (also known as Social Bonding Theory).

Social Control Theory posits that social constraints (e.g., child attachment or bonds to family, school and church) function to prevent or inhibit the occurrence of problem behaviour. Problem behaviour occurs when emotional attachments to important agents of socialization such as the family, church, and school, attenuate or disappear. These weakened social bonds signify a loss of commitment by the young person to the conventional social order (society) and a lack of response to the sanctions of significant others (e.g., parents, teachers) on inhibiting problem behaviour. Within this framework, geographic relocation is viewed as a force that threatens to weaken bonds to the conventional social order.

Social Control Theory identifies three elements that bond individuals to society, thereby preventing the occurrence of problem behaviour. They are: 1) attachment to significant others (e.g., parents, teachers, and religious leaders); 2) commitment to and involvement in conventional goals and activities (e.g., school and school-related activities); and, 3) belief in shared norms and standards of conduct.

Proponents of Social Control Theory (e.g., Agnew, 1991; Krohn, Massey, Skinner & Lauer, 1983; Paternoster & Iovanni, 1986) argue that "attached" young people are hesitant to engage in problem behaviour because such activities risk invoking a negative response from significant others whose opinions are valued. Commitment to and involvement in conventional activities (e.g., school work, athletics, church, and youth clubs) lessens the likelihood of problem behaviour because the young person is well-integrated into the conventional social order (i.e., the committed person has less time to engage in problem behaviour) and is committed to the pursuit of socially-approved goals (e.g., educational attainment) (i.e., the committed person has a stake in conformity). Finally, the occurrence of problem behaviour is expected to be low among young people who espouse strong beliefs in the conventional values, norms, and expectations of society.

Traditional Social Control Theory placed heavy emphasis on attachment to agents of socialization as "mechanisms of control". Borrowing from Patterson's coercion model of deviant behaviour, current reformulations of the theory (see Gottfredson & Hirschi, 1990; Sampson & Laub, 1993) have broadened the concept of control to include measures of parental disciplinary practices and supervision. These include: 1) frequent monitoring or supervision of children's activities; 2) parental acceptance of children; and 3) the use of punitive measures to correct misbehaviour in a consistent and loving manner. Together, these categories capture Stack's (1994) distinction between demand measures of social control (the idea that the demand to control children may increase with moving) and the supply of social control (the idea that moving may erode parental emotional resources).

2.1 Study Objectives

Based on a review of the literature and Hirschi's Social Control Theory, a test of the following hypothesis is proposed: geographic relocation (frequent and recent moves) increases the probability of problem behaviour1 among children by weakening parental supervisory capacity and disciplinary practices and child emotional attachments to family, school, church, and community. Two primary study objectives include: 1) to advance our understanding of the underlying mechanisms linking the number and recency of childhood geographic moves with problem behaviour; and (2) to identify the conditions under which frequent and recent moving is associated with problem behaviour. Previous studies have shown that environmental change for children is sufficiently potent to generate formal and informal referrals to special services (Elias, Gara, & Ubriaco, 1985). Identifying the mechanisms through which relocation may influence childhood behaviour (i.e., mediators) and the circumstances under which harmful effects occur (i.e., moderating influences) will facilitate the development of effective school and community programming for relocated youth.

2.2 Models Tested

Based on the work of Baron and Kenny (1986), illustrations of the hypothesized models are presented in Figure 1. For the mediator model, the left-hand construct is the independent variable, which consists of total moves and recency of the last move. The middle construct consists of the proposed mediators which include measures of social control. The right-hand construct (i.e., the dependent variable) represents childhood problem behaviour.

Mediation is demonstrated when a statistically significant relationship between geographic relocation and the dependent variable is substantially reduced or eliminated after the addition of mediators to the model. The independence model assumes that geographic relocation and social control each have a directand unique influenceon childhood problem behaviour. This is demonstrated when geographic relocation and social control measures yield statistically significant relationships with study outcomes when the effects of each set of variables are partialled out. A moderator or interaction effect is shown when the relationship between the independent and dependent variables differs significantly across levels of social control/bonding (i.e., hypothesized mediators).