Skip to main content

Table 5 Series of logistic regression models explaining/predicting ‘violence victimisation’

From: Family violence in a sample of treatment-seeking gamblers: the effect of having dependent children

Covariates from individual block sub-models Covariate category Any violence victimisation
 %
For covariate category
Model 1
(“Children” covariate only)
n = 164
Model 2
(+ Socio-demographics)
n = 162
Model 3
(+ Socio-demographics + Psycho-social factors)
n = 159
Odds ratio (95% CI) p-value Odds ratio (95% CI) p-value Odds ratio (95% CI) p-value
Children aged < 18 years usually living in household by Gender of Gambler Male, no children (n = 65) 38.46 1 (reference) 0.005 1 (reference) 0.002 1 (reference) 0.005
Female, no children (n = 39) 41.03 1.13 (0.50–2.50) 1.01 (0.43–2.36) 0.95 (0.39–2.33)
Female, with children (n = 31) 77.42 5.49 (2.06–14.60) 6.77 (2.39–19.22) 5.97 (2.06–17.34)
Male, with children (n = 29) 55.17 1.97 (0.81–4.78) 2.41 (0.92–6.34) 2.16 (0.78–6.01)
Socio-demographics (block 1)
 Asian No (n = 133) 54.89   1 (reference) 0.005 1 (reference) 0.02
Yes (n = 29) 27.59 0.25 (0.09–0.66) 0.30 (0.10–0.85)
 Income support No benefit (n = 99) 53.54   1 (reference) 0.009 1 (reference) 0.01
Support/NZ Super/Student (n = 65) 43.08 0.37 (0.18–0.78) 0.38 (0.17–0.81)
Psycho-social factors (block 2)
 Symptom rating scale—Inadequacy ≤ 3 (n = 61) 36.07    1 (reference) 0.02
> 3, ≤ 5 (n = 22) 31.82 0.86 (0.28–2.69)
> 5, ≤ 8 (n = 39) 64.10 3.00 (1.17–7.66)
> 8 (n = 39) 64.10 2.99 (1.20–7.44)
Adjusted R-squared    0.12 0.22 0.28
  1. Note, for the model predicting violence victimisation, no gambling block variables were able to be added
  2. The gambling “block” model for violence victimisation (which just had covariates “Children aged < 18 years usually living in household by Gender of Gambler” and “Pub or club EGMs main type”) had an odds ratio for violence victimisation comparing “Pub or club EGMs is the main type” to “Pub or club EGMs is NOT the main type” of 2.37 (1.21–4.65) with a significant p-value of 0.01. Had we added “Pub or club EGMs main type” to Model 3 for violence victimisation which takes into account socio-demographic and psychosocial factors we would have reduced the odds ratio above to an insignificant (with p = 0.30) 1.51 (0.70–3.29)