Skip to main content

Table 2 Results of maximum likelihood random effects regression

From: Validity of self-reported criminal justice system involvement in substance abusing women at five-year follow-up

  Model 1   Model 2  
  Odds ratio/Rho Z Odds ratio/Rho Z
Prison
   Official record of prison 70.60 ***5.58 47.26 ***4.81
   Official record of theft NA   0.91 -0.15
   Official record of drugs offences NA   2.35 1.27
   Official record of violence NA   2.23 1.10
Theft
   Official record of theft 3.30 *2.34 1.96 1.31
   Official record of drugs offences NA   5.45 **3.35
   Official record of violence NA   1.28 0.40
   Official record of prison NA   4.80 *2.44
Drugs
   Official record of drugs offences 6.66 ***3.76 5.17 **3.33
   Official record of theft NA   1.84 1.21
   Official record of violence NA   1.70 0.92
   Official record of prison NA   4.24 *2.65
Violence
   Official record of violence 6.28 *2.45 5.59 *2.00
   Official record of drugs offences NA   1.20 0.20
   Official record of theft NA   0.50 -0.82
   Official record of prison NA   2.49 0.92
  1. Notes: Model 1: Logistic regression with a random effect for intra-subject variation and the corresponding official record as predictor. Model 2:Logistic regression with a random effect for intra-subject variation and official record for all categories as predictors. † p < 0.10. * p < 0.05. ** p < 0.01 *** p < 0.001. NA: Not applicable as this effect is not included in present model.