A Comparative Study of Using the Methods of Multiple Linear Regression and Artificial Neural Networks in Organizational Correlations for the Fields of Management and Organization

Abstract

The purpose of this study is to compare the performances of the Multiple Linear Regression (MLR) and Artificial Neural Networks (ANN) used in correlations for the fields of management and organization, and to demonstrate that the ANN method can be used in organizational studies. Therefore, first, comprehensive information is provided about the ANN method and its use in literature and comparative studies. Work-family conflict, affective commitment, and turnover intention have been used as the variable group to compare the performance of the ANN with that of the MLR. Validity and reliability tests have been carried out, and the correlation coefficient R (correlation coefficient) and RMSE (Root Mean Square Error) have been considered as the criteria for performance. Within the scope of the findings, the ANN method is found to demonstrate higher R and lower RMSE values, when compared with the MLR method, although the effect coefficients obtained through both methods regarding correlations are noted to be similar.

TOPSIS-based Regression Algorithms Evaluation

“…Furthermore, different measurements attempt regression performance from different perspectives. Subsequently, benchmarking the regression algorithms in various fields, or even in a single field, is not trivial given the previously mentioned variations in concerns and measurements (Zorlu, 2012).…”