Design principles for data-driven employee profiling : With a case study on recruitment and selection at Sunsational Swim School
Dolle, Alexandra (2022-06-10)
Design principles for data-driven employee profiling : With a case study on recruitment and selection at Sunsational Swim School
(10.06.2022)
Julkaisu on tekijänoikeussäännösten alainen. Teosta voi lukea ja tulostaa henkilökohtaista käyttöä varten. Käyttö kaupallisiin tarkoituksiin on kielletty.
suljettu
Julkaisun pysyvä osoite on:
https://urn.fi/URN:NBN:fi-fe2022062750090
https://urn.fi/URN:NBN:fi-fe2022062750090
Tiivistelmä
This design science research aims at exploring a data-driven model in order to create employee profiles and consequently help in the external recruitment and more specifically the selection of new employees in companies. A case study is performed with the use of the employee records from a North American company named Sunsational Swim School.
Firstly, literature research is used in order to define the important concepts of Human Resource Management (HRM), Human Resource Analytics (HRA), Strategic Workforce Planning (SWP), Human Resource Information System (HRIS) and recruitment and selection process. Importantly, a definition of employee profiling is provided for recruitment and selection.
Secondly, we deal with the exploration of models for employee profiling for the case study with the use of the CRISP-DM method for process guidance. A classification model is found as the most appropriate for Sunsational Swim School in order to meet the requirements. The company needs a model that classifies the candidates in order to discover patterns behind good, medium and candidates that are not a good fit for the company. In addition, the company wishes to have a performant classification model that predicts if new applicants are likely to fall into a good, medium or bad fit for the company.
The output of this research highlights nine general principles that other companies can learn from, in order to apply data-driven employee profiling for recruitment and selection. The most important principles are that companies needs an HRIS with required data to conduct profiles, it is necessary to have an evaluation process after the first interview within the recruitment process in order to obtain a certain score, it is highly recommended to follow the CRISP-DM framework to solve such data mining challenges, it is required to have a significant amount of applicant records in order to have a more reliable model, and it is required to identify “what is the best model” for the company to make a good choice in the assessment of it.
Firstly, literature research is used in order to define the important concepts of Human Resource Management (HRM), Human Resource Analytics (HRA), Strategic Workforce Planning (SWP), Human Resource Information System (HRIS) and recruitment and selection process. Importantly, a definition of employee profiling is provided for recruitment and selection.
Secondly, we deal with the exploration of models for employee profiling for the case study with the use of the CRISP-DM method for process guidance. A classification model is found as the most appropriate for Sunsational Swim School in order to meet the requirements. The company needs a model that classifies the candidates in order to discover patterns behind good, medium and candidates that are not a good fit for the company. In addition, the company wishes to have a performant classification model that predicts if new applicants are likely to fall into a good, medium or bad fit for the company.
The output of this research highlights nine general principles that other companies can learn from, in order to apply data-driven employee profiling for recruitment and selection. The most important principles are that companies needs an HRIS with required data to conduct profiles, it is necessary to have an evaluation process after the first interview within the recruitment process in order to obtain a certain score, it is highly recommended to follow the CRISP-DM framework to solve such data mining challenges, it is required to have a significant amount of applicant records in order to have a more reliable model, and it is required to identify “what is the best model” for the company to make a good choice in the assessment of it.