Multiple Linear Regression Analysis for Identifying the Most Affecting Factor on The Quality of Raw Tea Materials in PT Pagilaran

Abstract

Most of the black tea production in Pagilaran Batang Production Unit is to meet the demands of consumers from abroad. Because of the abundant requests from abroad, the quality of black tea product must be always controlled, therefore it is necessary to conduct a Multiple Linear Regression Analysis of the tea shoots to find out the most affecting factor of the tea shoots quality so that the quality of contributing factors of highest First Grade black tea product can be identified. The methods used in this study was Multiple Linear Regression Analysis and Multivariate Analysis where the Multivariate Analysis was the Factor Analysis. In addition to determine the regression equation, classical assumption test calculations (normality test and heteroscedasticity test), correlation coefficient, determination coefficient, F hypothesis test and t hypothesis test were administered to find the most affecting  factor of the tea shoots quality. From the results of the analysis and discussion on the calculation of multiple linier regression in each garden (y₁, y₂, y₃) it was obtained that the most affecting factor was fertilizer with a value in Afdeling Pagilaran (y₁) of – 1415,934 , Afdeling Andongsili (y₂) of – 265,858 and Afdeling Kayulandak (y₃) of 225,589. Then on the calculation of multiple linier regression in each garden (y₁, y₂, y₃) after summarizing the variables it was obtained that the most affecting factors in Afdeling Pagilaran (y₁) was a Trip Pest with a value of - 53,741 , Afdeling Andongsili (y₂) was fertilizer with a value of - 46,301 and Afdeling Kayulandak (y₃) was fertilizer with a value of 213,073