104 CHAPTER IV RESULT S AND DISCUSSION Employing a mixed-methods approach, this study seamlessly integrates the depth of qualitative analysis with the empirical rigor of quantitative data. Initiated by qualitative semi-structured interviews, this phase captures the intricate realities and nuanced understandings of employee green behavior (EGB) and green human resource management (GHRM) practices from the perspectives of individuals within the company. These interviews play a pivotal role in extracting the subtleties of behaviors, motivations, and organizational culture that pure numerical data might overlook. Subsequently, the study transitions into its quantitative phase, with surveys distributed across a broader employee population to test hypotheses and validate the emerging research model on a larger scale. This combination not only enhances the understanding of the subject matter but also anchors the findings in a robust empirical foundation, ensuring both internal richness and external validity of the research. Through this mixed-methods lens, the study is well-positioned to offer actionable insights with a balance of detail and generalizability, thus constructing a comprehensive picture of the green practices within the targeted companies. This exploration is anticipated to provide a comprehensive understanding of how the antecedents of EGB manifest within the coal mining sector, offering valuable insights into the intricate dynamics of sustainable practices in this specific organizational context. IV.1 Results IV.1.1 Qualitative Results The analysis process is essential for deriving high-quality insights from qualitative data. Employing NVivo, a sophisticated qualitative analysis tool, this research systematically organized, coded, and comprehended complex datasets. Koleksi digital milik UPT Perpustakaan ITB untuk keperluan pendidikan dan penelitian 105 In the initial phase, all data, including interview transcripts, observational notes, and multimedia files, was imported into NVivo. The software’s flexibility facilitated seamless integration of diverse data formats, which were then meticulously organized into folders or cases, enhancing efficient analysis by participant group, data source, or other relevant categories. The coding process is pivotal, involving the tagging of text segments with codes representing specific themes or ideas. NVivo accommodates both inductive and deductive coding, enabling the exploration of emergent themes while testing pre- defined concepts. Nodes, representing themes or concepts, serve as repositories for related coded references, simplifying the retrieval and examination of specific themes. NVivo’s powerful querying capabilities allow detailed searches, exploration of word frequencies, and examination of how themes occur across the dataset. Visualization tools within the software generate models, mind maps, and charts, aiding in the identification of patterns and relationships. Memoing, a feature in NVivo, supports a rigorous analytical process by allowing researchers to record insights and interpretative decisions, enhancing the transparency and auditability of the research process. To ensure the validity of the qualitative analysis, we employed several validation methods, such as saturation and the pattern-matching technique. Firstly, the saturation was achieved when additional data collection and analysis no longer contributed new insights or themes related to the research objectives. This was evident in this study as the coding of interviews from ten participants yielded recurring codes and themes, indicating that further data collection would likely not introduce new concepts.