ABSTRACT
Despite anecdotal evidence of a COVID-19 induced decline in the intensity of interstate migration in Australia, population-level evidence is limited. The recent release of the 2020 wave of the Household, Income and Labour Dynamics in Australia (HILDA) Survey provides a unique opportunity to robustly assess the effect of the COVID-19 pandemic on the level, direction, determinants, and reasons for migration in Australia. By applying a series of regression models to individual-level longitudinal microdata, and measuring migration at a range of spatial scales, this paper shows that COVID-19 has somewhat accelerated the long-term decline in the intensity of internal migration-particularly for residential mobility, short-distance migration, and migration due to employment and involuntary reasons. The socio-demographic determinants of migration have remained broadly stable, despite a slight increase in the deterring effect of duration of residence and a reduction in the impact of education. Finally, we show that the increase in net migration gains in regional areas is underpinned by a decrease in outflows. Juxtaposing these results with aggregate-level migration statistics from the Australian Bureau of Statistics from 2021, we conclude that the effect of COVID-19 on internal migration to date has been minimal and is likely to be short-lived. However, it may still be too soon to make a definitive judgement, as shifts in work patterns stemming from the pandemic may further transform the level, direction, and composition of internal migration.
ABSTRACT
BACKGROUND: Mental health disorders are ranked globally as the single largest contributor to non-fatal ill-health. Social support can be a means of reducing and managing depression. However, depression can also impact on a person's level of social support. OBJECTIVE: As men typically have fewer sources of social support than females, this study investigated the bi-directional associations between depressive symptoms and perceived levels of social support among Australian males, aged 18-63. METHODS: Three waves of panel data from Ten to Men: The Australian Longitudinal Study on Male Health collected over 7 years (2013-2020) were used. A random intercept cross-lagged panel analysis with 5112 participants was undertaken. Mediating effects and indirect and total effects for lagged and cross-lagged pathways were also examined. RESULTS: Over time, greater social support was found to be associated with lower depression levels, and simultaneously greater levels of depression was found to be associated with lower levels of social support. Standardised cross-lagged effects between waves were mostly similar (ß = 0.10). However, mediation analyses identified that only the total effect size of the association for depression at wave 1 predicting social support at wave 3 (ß = -0.29) was significant. Mediated effects of social support at wave 1 predicting depression at wave 3 were not significant. LIMITATIONS: These include the number of years between each wave, and data were collected during the COVID pandemic. CONCLUSION: The study provides robust longitudinal evidence supporting the notion that social support and depression are both a cause and consequence of the other. However, the long-term effects of depression reducing social support were longer lasting than the effects of social support reducing depression.
ABSTRACT
We address the use of selfie ocular images captured with smartphones to estimate age and gender. Partial face occlusion has become an issue due to the mandatory use of face masks. Also, the use of mobile devices has exploded, with the pandemic further accelerating the migration to digital services. However, state‐of‐the‐art solutions in related tasks such as identity or expression recognition employ large Convolutional Neural Networks, whose use in mobile devices is infeasible due to hardware limitations and size restrictions of downloadable applications. To counteract this, we adapt two existing lightweight CNNs proposed in the context of the ImageNet Challenge, and two additional architectures proposed for mobile face recognition. Since datasets for soft‐biometrics prediction using selfie images are limited, we counteract over‐fitting by using networks pre‐trained on ImageNet. Furthermore, some networks are further pre‐trained for face recognition, for which very large training databases are available. Since both tasks employ similar input data, we hypothesise that such strategy can be beneficial for soft‐biometrics estimation. A comprehensive study of the effects of different pre‐training over the employed architectures is carried out, showing that, in most cases, a better accuracy is obtained after the networks have been fine‐tuned for face recognition. [ABSTRACT FROM AUTHOR] Copyright of IET Biometrics (Wiley-Blackwell) is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)