Predictors of Psychosocial Adjustment among Retirees in Zimbabwe: Testing a Resource-Based Model

Background: Retirement is a critical life transition that can significantly affect psychological well-being, yet the predictors of successful adjustment in low-resource African settings remain understudied (Wang & Shi, 2014; Hsu et al., 2025). Grounded in the resource-based dynamic model of retirement adjustment (Wang et al., 2011; Froidevaux et al., 2024), this study examined the psychosocial adjustment of retirees aged 60 and above who were beneficiaries of the National Social Security Authority (NSSA) pension scheme in Masvingo Province, Zimbabwe. Materials and Methods: A cross-sectional survey was conducted with a sample of 80 retirees (52.5% female; mean age 64.2 years) who had retired within the preceding 24 months. Participants completed an adapted version of the Health and Retirement Study Psychosocial and Lifestyle Questionnaire (Smith et al., 2013), which demonstrated acceptable internal consistency (α > .70 for all composite scales). Hierarchical logistic regression analysis was used to identify predictors of life satisfaction, the primary indicator of adjustment (Cheung & Lucas, 2014). Results: Results indicated that 72.5% of retirees reported low life satisfaction. Significant predictors included ongoing financial strain (OR = 0.29, p < .001), perceived negative social support (OR = 0.35, p = .004), and everyday discrimination (OR = 0.46, p = .022), while engagement in volunteer work was protective (OR = 2.63, p = .031). Pre-retirement preparation from institutions was near-absent. The findings support a resource-based perspective (Wang et al., 2011; Zhan et al., 2023), suggesting that material, social, and personal resources are differentially associated with adjustment. Conclusion: The study highlights the need for holistic pre-retirement programmes that extend beyond financial provision to address psychosocial preparedness (Pang et al., 2016; Nyamubarwa, 2025). Policy and practice implications are discussed, alongside limitations related to sample size (Peduzzi et al., 1996) and cross-sectional design.