Analyzing Factors Influencing the Research Progression of Faculty in STEM based Universities

Abstract

progression that prompts University-Industry interaction and also the knowledge transfer between them. As researchers analyze the factors affecting the advancement in research progression of faculty, they consistently show that a lot of factors affecting performance such as time spent on research, academic responsibilities, administrative responsibilities, and other factors such as family responsibilities, life style, health factors, age, inter personnel communication etc., In this study, we figure out the relationship between these factors, and optimizing the schedules by balancing those factors which have greater influence  in research progression and also we can predict the research progression of individuals. The data was collected from 121 faculty members of university working in various departments and pre-processed.  A model is designed and developed with Multiple Regression. Experiments have been conducted and performance of the model is evaluated on collected data.  R² is the coefficient of determination, whose value is determined as 85.04%. The above value indicates that the independent variables describe nearly 85% of the variance in dependent variable. The results were compared with the results of other two models Multilayer Perceptron and SMO in predicting the research progression. From the comparison it is clear that Multiple regression performs well. Not only research progression, factors effecting faculty stress are also analyzed. By analyzing these factors, female faculty married and having children, faculty who are having less experience and the faculty who joined recently in this organization are feeling stress in meeting the research targets. Some hidden factors, such as unshared personal, confidential information by volunteers limited the performance of the model.  Counter measures to improve performance further will be taken in future work.

Keywords: Research Progression, Regression, Prediction, Multiple Regression, Stress Analysis.