ANALISIS PENGARUH PRODUK, LOKASI, PROMOSI DAN KUALITAS PELAYANAN TERHADAP KEPUTUSAN PEMBELIAN PRODUK MERCHANDISE (Studi Pada Mandiri Art, Tembalang Semarang)

Abstract

This study aims to analyze how big the influence of product, location, promotion and service quality of the purchase decision merchandise product on the Mandiri Art, Tembalang Semarang. And this research also aims to analyze the most dominant factors that influence on purchase decisions merchandise product on the Mandiri Art, Tembalang Semarang. The population used in this study is consumers who ever buy in the Mandiri Art, Tembalang Semarang. The sample in this study are 100 respondents and the techniques used are non-probability sampling technique with the approach of accidental sampling (sampling based on chance). Analytical methods used were multiple regression analysis. This analysis includes: validity and reliability, the classic assumption test, multiple linier regression analysis, and expedience model test. From the analysis of the regression equation was obtained as follows, Y = 0,320 X1 + 0,178 X2 + 0,230 X3 + 0,183 X4 The product variables (X1), Location (X2), Promotion (X3), and Service Quality (X4) have a positive impact to purchase decisions (Y). The product variables have the most impact on purchase decisions for 0,320, followed by promotion variable amounting to 0,230 and then by service quality variable amounting to 0,183. While the location variables have an influence lowest compared to other variables for 0,178. Testing using t tests showing that the four independent variables are product (X1), location (X2), promotion (X3) and servive quality (X4) which investigated proved positively and significantly affect the dependent variable is the purchase decision (Y). Then through the F test can be seen that the variables product, location, promotion and servive quality eligible to test the dependent purchasing decisions. Adjusted R Square explained figures that 34,1% of purchase decisions variation can be explained by the four independent variables in multiple regression equations. While the rest of 65,9% was explained by other variables outside of the four variables used in this study.