Table 2.
Odds ratios from key logit regressions of the likelihood of being at “very high” risk of psychological distress.
| Model A Simple | Model BFull multivariate model | Model C Participating in at least 50% of waves | Model DPermanent sample members only | |
|---|---|---|---|---|
| A) Discrete time (year): 2007 (ref) | 1.00 | 1.00 | 1.00 | 1.00 |
| 2009 | 0.99 | 0.99 | 0.93 | 0.97 |
| (0.84, 1.18) | (0.83, 1.18) | (0.72, 1.08) | (0.81, 1.17) | |
| 2011 | 1.13 | 1.13 | 1.16 | 1.14 |
| (0.94, 1.35) | (0.94,1.35) | (0.99, 1.37) | (0.94, 1.38) | |
| 2013 | 1.18 | 1.19 | 1.16 | 1.15 |
| (1.01, 1.39) | (1.01, 1.40) | (1.00, 1.36) | (0.97, 1.36) | |
| 2015 | 1.43 | 1.42 | 1.41 | 1.40 |
| (1.19, 1.71) | (1.18, 1.72) | (1.18, 1.69) | (1.16, 1.70) | |
| 2017 | 1.57 | 1.57 | 1.47 | 1.57 |
| (1.34, 1.83) | (1.33, 1.84) | (1.25, 1.75) | (1.33, 1.86) | |
| B) Continuous (linear) time | 1.05 | 1.05 | 1.05 | 1.05 |
| (1.03, 1.07) | (1.04, 1.07) | (1.03, 1.07) | (1.03, 1.07) |
Notes: 95% confidence intervals in parentheses.
Source: Department of Social Services/Melbourne Institute of Applied Economic and Social Research (2018).