Big Data at Work: The Data Science Revolution and Organizational Psychology (SIOP Organizational Frontiers Series) by

Language: eng
Format: azw3
ISBN: 9781848725812
Publisher: Taylor and Francis
Published: 2015-11-05T16:00:00+00:00

In this case, let’s say the analytics team lead is an I-O psychologist whose first instinct is to look internally to see what research the organization has conducted in the past with regard to turnover. As it turns out, the organization doesn’t have much of a history of conducting turnover-related work, so there is not much to go on there. Furthermore, let’s say the organization resides in an industry where little is known about drivers of turnover, beyond the general theoretical perspectives and models provided in the I-O literature.

Being empty handed up to this point, the lead reviews the I-O literature for theories and models of turnover. Upon review of the prevailing theories and models and comparing their substantive content to the survey and HRIS data available in-house, the team finds that they have the ability to develop and test several theoretically-informed structural models of turnover. However, the team realizes that measures of some variables within the models will be limited because the organization’s annual employee surveys were designed for broader purposes than evaluating pre-existing models of turnover.

Upon specifying several turnover models based on prevailing I-O turnover theory and research, the I-O team lead shares the models with a labor economist and econometrician affiliated with the analytics team. They provide additional insights into variations in economic and market conditions present throughout the past four years that research from their respective fields’ literatures suggest would also predict stay-level decisions, and they highlight data from group-level databases that could be leveraged. Based on this work, the team now has multiple sets of structural models, some based purely on I-O theories and research, and others reflecting a blend of research from I-O and labor economics.

After completing the steps above, the team conducts a content analysis of available open-ended responses from the past several years of employee exit survey data. Upon analyzing these data, the team discovers several themes that suggest there are occupation-level, supervisor-level, and region-level differences in the reasons why individuals are leaving. These reasons are inductive surprises in that they fall outside the scope of both the working models the team has developed based on multidisciplinary theories and research on turnover. With these qualitative themes identified, the team searches its data sources for data fields that might account for such differences, gives thought to how those variables would fit into the existing structural models, and specifies yet another set of structural models that incorporate these factors as adequately as possible. It is important to note that at this point in the process, no models have actually been fit to the data—at this point the team is simply focused on formulating multiple alternative structural models for subsequent evaluation.

To summarize, the team has already identified predictors of turnover that have been informed by theories and research in I-O and labor economics, and has effectively mined its qualitative exit survey data and identified potential occupation-level, supervisor-level, and region-level differences to be further examined. Next, the team decides to take stock of what it can learn from the substantial quantitative data to which it has access.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.