Skip to main content
. Author manuscript; available in PMC: 2010 Jan 12.
Published in final edited form as: Res Soc Work Pract. 2009 Jul 1;19(4):423–433. doi: 10.1177/1049731508329407

Table 4.

Multilevel Logit Model for Probability of Negative Placement Disruption (n = 337 Families, n = 59 Groups)

Negative Placement Odds Ratio
Model 1 Model 2
Foster Child Age 0.93 0.95
Foster Child Sex (male) 1.02 0.92
Number of Prior Placements 0.99 1.01
Foster Parent- African American 2.25 2.24
Foster Parent- Hispanic 1.53 0.96
Foster Parent Age 0.99 0.52
Ethnic Match Foster Parent-Child 0.65 0.63
Ethnic Match Foster Parent-Group Leader 0.67 0.81
Kinship Status 0.33** 0.42**
Log FP Mood 13.75** 13.95**
Group Engagement × Child Age 0.87
Group Engagement × Child Sex (male) 0.23
Group Engagement × Prior Placements 1.16
Group Engagement × FP African American 0.64
Group Engagement × FP Hispanic 0.03*
Group Engagement × FP Age 0.14
Group Engagement × Ethnic Match Child 3.68
Group Engagement × Ethnic Match Leader 0.79
Group Engagement × FP Kinship 9.06
Group Engagement × FP Mood 1.78
Incremental Change in Proportion of Variance Explained .05 .41

Note: FP = foster parent

**

p < .01;

*

p < .05