Ridderprint

Chapter 7 146 topic is how to manage people analytics in practice, with scholars discussing the setup of the department, the required individual and organizational capabilities, and the delivery of people analytics projects (e.g., Andersen, 2017; Baesens, de Winne, & Sels, 2017; Levenson & Fink, 2017; Minbaeva, 2017b; Van der Togt & Rasmussen, 2017; Schiemann et al., 2017; Simón & Ferreiro, 2017). Fewer studies in this stream discuss actual applications of people analytics and those that do are mostly limited to high-level case study information (e.g., Kryscynski et al., 2017; Rasmussen & Ulrich, 2015; Van der Laken, Bakk, Giagkoulas, Van Leeuwen, & Bongenaar, 2017). In general, the stream is nearly void of empirical investigations, let alone applications of advanced analytics (e.g., text mining, predictive modelling, deep learning). Paradoxically, the mainstream people analytics literature consists mostly of meta-research , discussing but not demonstrating what people analytics is or what it brings to the table. 7.3.1.3 Data Mining and Decision-Support Systems Third, there is a large scholarly community exploring applications of people analytics, but residing mostly outside of the HRM domain and refraining from use of the popular terminology. For instance, scholars have explored data mining algorithms and decision-support systems within HRM, mostly for selection purposes, but also in light of compensation, development, performance management, and retention (see Strohmeier & Piazza, 2013). These publications are highly technical and focused specifically on the data mining process, the predictive modeling process, and/or the decision-support quality. The theoretical implications, the strategic value, and the practical and ethical feasibility of the people analytics applications are discussed to a lesser extent (e.g., Al-Radaideh, & Nagi, 2012; Chien & Chen, 2008; Dursun & Karsak, 2010; Jantan, Hamdan, & Othman, 2010; Saradhi & Palshikar, 2011; Valle, Varas, & Ruz, 2012). Additionally, these studies are published almost exclusively outside of the (HR) management domain and do not use the popular management terminology (e.g., people analytics). However, this line of research very closely embodies what I would consider people analytics. This “ prospering new field of data mining research […] provides ample insights in how to generate advanced information and decision support within the HR domain ” (Strohmeier & Piazza, 2013, p. 40), thus exploring how information technology and predictive modelling allow for data- driven decisions on HRM topics (e.g., Davenport & Harris, 2007; Marler & Boudreau, 2017). Unfortunately, such studies are infrequently conducted by HRM researchers or published in management and psychology journals. In light of Wright and Boswell’s (2002) typology of HRM research, I have only encountered several studies focusing at the individual level of analysis and on a single practice, again in the selection space. For instance, a few scholars have adopted the data mining mindset and are applying machine learning to automatically evaluate jobs and match jobs and applicants (e.g., Kobayashi, Mol, Berkers, Kismihok, & Den Hartog, 2017). Similarly, scholars within recruitment and selection are considering the use of more advanced social network data and analysis, though still far off from creating automated decision-support systems (Caers & Castelyns, 2011; Kluemper & Rosen, 2009; Kluemper, Rosen, & Mossholder, 2012; Roth et al., 2016; Van Iddekinge, Lanivich, Roth, & Junco, 2016). When it comes to terminology, these HRM