Representative data collection is fundamental to social science research and evidence-based policy development, yet standard survey methodologies struggle to adequately capture hard-to-reach (HTR) populations essential to understanding social inequality and economic vulnerability. This literature review synthesizes current knowledge on sampling strategies designed for HTR populations, including mobile or migrant workers, and geographically isolated communities. The review examines prominent sampling methodologies developed for HTR contexts, including non-probability approaches like snowball sampling and respondent-driven sampling (RDS), location-based probability methods such as time-location sampling (TLS), design-based innovations like adaptive cluster sampling (ACS), and indirect estimation techniques including capture-recapture and the network scale-up method (NSUM). The review emphasizes that ethical considerations are paramount, including informed consent challenges, confidentiality protection, and harm prevention. The review concludes that no single method is universally optimal; local context must shape methodological choice, and triangulation using multiple methods is strongly recommended to address inherent limitations and improve equity in research representation.