Low-cost air temperature sensors are an emerging theme in environmental monitoring. These sensors offer the advantage of making microclimate monitoring feasible due to their affordability. However, they are limited by the quality of the data they provide; in many cases, they have been reported to have presented errors in the sensor readings. These errors have been shown to improve after calibration was applied. The lack of a comprehensive understanding of the available calibration techniques, models, and sensor types has led to studies presenting heterogeneity in models and techniques alongside different performance metrics. To address this gap, this study conducted a systematic review following the PRISMA guidelines, reviewing studies from 2015 to 2024 across the databases Web of Science and Scopus, alongside the search engine Google Scholar. The aim was to identify the calibration techniques and models, the commercially available low-cost air temperature sensors used, the performance metrics utilised, and the calibration settings. The findings presented three main categories of calibration models utilised in the collected studies: linear, polynomial, and machine learning. Twenty-two commercially available low-cost sensors were identified, with the DHT22 sensor being the most utilised. Indoor settings were identified as the most preferred for conducting calibrations. Key challenges included limitations in reported results for calibration by the studies, the use of different performance metrics across studies, insufficient studies conducting calibration, and the diversity in sensor types utilised.