Abstract
After a literature review and hypothesis preparation, data were collected through a questionnaire method of 40 people working in the EDT division obtained by using saturated sampling, which means that all populations are sampled. Then analyzed the data obtained by using multiple regression analysis. This analysis includes: validity and reliability test, classical assumption test, multiple regression analysis, multiple correlation analysis, correlation coefficient test, coefficient of determination test (R2), from the analysis obtained regression equation: Y = 8.771a + 0,426 X1 + 0,403 X2, and hypothesis testing through t test and f test.
Where variable Employee Productivity (Y), Internal Conflicts (X1), and Communications (X2). Hypothesis testing using t test shows that both independent variables studied proved to significantly affect the dependent variable employee productivity. Then through the test f can be seen that the internal conflict and communication variables worth to test the dependent variable employee productivity. The R square number of 0.790 indicates that 79 percent of the employee productivity variation can be explained by the two independent variables in the regression equation. While the remaining 21 percent is explained by other variables outside the three variables used in this study.