Ridderprint

Discussion 141 matching analysis helps to gain insights from new high volume and high velocity HRM data formats, such the data gathered via continuous listening. 16 Researchers outside of scholarly HRM communities are already leveraging the value of these new methods in the HRM domain. Data mining scholars are using decision trees, cluster analysis, association analysis, support vector machines, and neural networks to solve HRM issues related to employee selection, development, and turnover, performance management and reward allocation, and career and cost planning (Strohmeier & Piazza, 2013, p. 2410). Similarly, Chapter 2 demonstrated that scholars are optimizing work conditions for well-being and productivity using sales and scheduling data (Tan & Netessine, 2014), improving team behaviors via machine learning (Johnson & Gonzalez, 2014), and evaluating safety climates by text mining employee interviews (Colley & Neal, 2012). Moreover, HRM data scientists and people analytics analysts in practice are applying advanced analytics to solve a variety of HRM issues. To illustrate, my direct colleagues in Shell have quantified team diversity in a multi-dimensional fashion via multiple correspondence analysis (Bongenaar & Van Leeuwen, 2016), predicted cyber security risks via tree-based models (Giagkoulas & Hawkes, 2016), digitized recruitment and assessment processes (Lam & Hawkes, 2017), and many more (Van der Togt & Rasmussen, 2017). However, such applications seem less widespread inmainstreamHRM practitioner and scientific communities. 7.1.3 RQ1: Intermediate Conclusion & Future Directions In conclusion, the spread of analytics seems slow within the HRM domain. Chapter 2 demonstrated that business value through data analytics and machine learning applications is achieved more frequently in other functional domains. HRM scholars and practitioners are often not familiar with the different methodological approaches that are needed for people analytics and the analysis of complex data formats (Chapter 3). While there are visible developments related to people analytics in scientific and practitioner communities (Chapter 2), the overall adoption is slow. These are early conclusions on limited research and further exploration via alternative designs, angles, and perspectives would be valuable. Chapter 2 relied on open- access data on scientific publication networks. Here, organization-level surveys regarding the use of, or familiarity with, (big) data, analytics, and algorithmic intelligence in the different functional management domains would have been a great addition in light of my first research question. Similarly, instead of exploring potential methodological applications, such as in Chapter 3, it would have been valuable to study the perceptions and knowledge of scholars and practitioners regarding different research designs, methods, and algorithms. In both cases, it would have been valuable to discriminate more explicitly between the rise of (people) analytics in HRM research and practitioner communities. 16 Process whereby organizations gather structured or unstructured data, such as employee experiences or feedback, at relatively frequent intervals (e.g., hourly, daily, weekly).