טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
M.Sc Thesis
M.Sc StudentGoldberger Sagi
SubjectA Field Study of a Data Analytics Training
Program to Upgrade Firm Performance: The Role
of On-The-Job Active Learning Methods
and the Technogical Context
DepartmentDepartment of Industrial Engineering and Management
Supervisor PROF. Eitan Naveh
Full Thesis textFull thesis text - English Version


Abstract

The accelerating growth in information technology is opening new directions for organizations to draw decisions based on data, and thus upgrade performance. The cyber-physical connectivity enables firms to digitize manufacturing plants, install sensors, which send out data heretofore unavailable, allowing firms to make decisions based on data, rather than on intuition or "gut feelings". Making decisions based on data is possible when a strong data analytics infrastructure exists within the organization-skillful personnel capable of processing, analysing, transforming, and visualizing data. One way to enhance organizations' data analytics skills is through employee training programs. Hence, organizations face the challenge of allocating resources towards effective data analytics training programs, which would ultimately boost their ability to make data-driven decisions. However, the transfer of trained knowledge and skills to be used on the job is influenced by various factors, such as learner's characteristics, training delivery method, and the organizational context. Thus, evaluating the existence and strength of factors influencing organization's data analytics skills -data-driven decision making relation holds both theoretical (i.e., understanding mechanisms of change) and practical (i.e., accommodating training programs) implications. The current study investigated how on-the-job active learning methods influence organization's data analytics skills - data-driven decision making relation. On-the-job active learning methods have yet to be thoroughly evaluated in training data analytics professionals.  Thus, it was unclear whether on-the-job active learning methods could provide the setting needed for data analytics training to effectively upgrade data-driven decision making.  Data was collected from Thirty-nine small and medium size' manufacturing organizations which participated in a data analytics training program. The data analytics training program provided an excellent platform for our examination, since it was comprised of two-stages in-class and on-the-job active learning. The measures of organization's data analytics skills, and data-driven decision making, were collected from four different training cycles. The cycles differed in the progress they had made in the training program as of the point in time of data collection. This allowed us to examine the organization's data analytics skills -data-driven decision making relation given progress in the training program, in a cross-sectional field study design. Results show that data analytics skills and data-driven decision making were significantly and positively related only when organizations completed both in-class and on-the-job active learning stages.  Organizations which underwent only the in-class training stage, did not exhibit a statistically significant relation between their data analytics skills and data-driven decision making. This illustrates the importance of on-the-job active learning in utilizing the organizational data analytics skills as a platform for decision making.  Contributions, limitations, as well as directions for future research are discussed.