1. Objectifs

Intentions :
« The study described here examines the specifcs risk factors linked to homelessness among adolescents, using a sample of adolescents of both sexes under the responsibility of child protection agencies (Youth Protection Centres). » (p. 216)

2. Méthode

Échantillon/Matériau :
« Our final sample consists of 408 families with adolescents between 12 and 17 and represents 41.1% of the initial sample. » (p. 219)

Instruments :
- Diagnostic Interview Schedule for Children version 2.3 (Schaffer, Fisher, Piacentini, Schwab-Stone, & Wicks, 1992);
- Family Assessment Device (Epstein, Baldwin, & Bishop, 1983);
- Alabama Parenting Questionnaire (Shelton, Frick, & Wootton, 1996);
- Index of Parental Attitude (Hudson, 1982);
- Conflict Tactics Scales (Straus, 1979);
- Indice de gravité de toxicomanie (Pauzé et al., 2000). (p. 220)

Type de traitement des données :
Analyse statistique

3. Résumé

« There are two factors that limit our knowledge of the risk factors associated with homelessness among runaway adolescents, namely (1) the samples used are often composed of youth homeless service users and/or youths living on the streets (visible homelessness), whereas most adolescents in fact use ’private’ resources (hidden homelessness), and (2) failure to use an adequate control group to identify risk factors associated specifically with homelessness. Our study compares the characteristics of two groups of youths under the supervision of the youth protection system, according to the presence or absence of periods of homelessness. The results throw light on the factors underlying the shift from ’at risk’ to ’homeless’, showing that youths with experience of homelessness are more likely to have been placed in substitute home environments, have experienced significant relationship difficulties with one of their parents (deterioration of the parent/youth relationship and parental abuse) and to have been diagnosed with behavioural disorders. The findings suggest that the decision to place young people under supervision is based more on the dynamic between risk factors rather than on the existence of behavioural problems. » (p. 215)