Adoption trend of residential solar photovoltaics (PV) and influencing factors: an Australian study

Abstract

Australia leads the world in per capita residential solar adoption, yet a definitive understanding of the factors driving consumer decisions remains elusive. Existing studies highlight various determinants to solar PV adoption, but their relative importance is often overlooked. This research addresses this gap by evaluating the significance of various factors influencing solar photovoltaic (PV) adoption, with a specific focus on the Australian context. Identifying the most critical factors driving the adoption will facilitate more targeted deployment strategies and hasten the transition to clean energy grids.

Employing a rigorous methodology, a comprehensive analysis was conducted integrating archival data examination and a broad-based survey covering diverse consumer variables, such as socio-demographics, economic considerations, technical aspects, and government policy influences. Using data from an online questionnaire survey (n=574) of respondents from Queensland, a robust statistical analysis was completed to identify the key factors affecting solar adoption. Feature importance was determined using logistic regression, decision trees, and random forest models, offering insights into the most critical adoption factors. Additionally, structural equation modelling (SEM) was employed to assess the foundational constructs shaping the residential solar PV adoption.

The findings illuminate key determinants shaping consumer decisions in Queensland. Trend analysis revealed that a reduction in feed-in tariffs and lower subsidies contributed to a slower adoption rate, counterbalanced by the declining levelized cost of electricity (LCOE). Results from the survey indicated that homeownership, government policy, property rights, installation costs, and electricity bills emerged as the most influential features to adoption of solar PV. In addition, SEM analysis indicated a substantial mediation of policy by financial factors, reinforced the significance of financial factors on adoption and did not establish any meaningful moderation effect of location, homeownership, and household size on the key constructs of the SEM model.

Date of Award	6 Jun 2024
Original language	English
Supervisor	Jasper Mbachu (Supervisor), Matthew Moorhead (Supervisor) & Alan Patching (Supervisor)