Multiple Linear Regression Analysis Analysis of The Coefficient of Determination R Partial Test T Test

58 X4, merchandise assortment X5, and pricing X6 on the dependent variable customer satisfaction. If the significance level yield number 0.05, it can be said regression model does not contain any heteroscedasticity. Ghozali, 2006: 129.

3. Multiple Linear Regression Analysis

According to Malhotra 2004:502, the regression analysis is a statistical procedure to analyze the relationship between the dependent variable and the independent variables. If there are two or more independent variables then using multiple linear regression analysis. Thus it can be seen how the influence of independent variables on the dependent variable. According to Malhotra 2004:512 formulas that can be used as a multiple linear regression analysis calculation is as follows: Y = a + b1X1 + b2X2 + b3X3 + b4X4 + b5X5 + b6X6 Description: Y = customer satisfaction dependent a = constanta X1 = cutomer service independent X2 = store design display independent X3 = communication mix independent X4 = location independent X5 = merchandise assortment independent 59 X6 = pricing independent b1 = regression coefficient of X1 b2 = regression coefficient of X2 b3 = regression coefficient X3 b4 = regression coefficient X4 b5 = regression coefficient X5 b6 = regression coefficient X6

4. Analysis of The Coefficient of Determination R

2 According to Ghozali 2006:87, the coefficient of determination R 2 essentially measures the how far the ability of model explain the variation of independent variables. The coefficient of determination is between zero and one. Small value of R 2 means the ability of the independent variables in explaining the variation in the dependent variable is very limited. Value that close to one means the independent variables provides almost all the information needed to predict the variation in the dependent variable.

5. Partial Test T Test

According to Setiawan and Kunto 2013:7,t-test is a statistical method used in the testing to test the influence of all independent variables on the dependent variable partially. The usefulness of this t-test is to test whether the variable customer service X1, store design display X2, communication mix X3, location X4, merchandise assortment X5, 60 and pricing X6 partially affect the customer satisfaction Y in SB Mart Bukit Sawangan Indah Housing. Steps to perform t test are as follows: a. Formulate statistical hypothesis b. Determine the critical value t table Selected level of significant α 2 = 5 2 0.025 Divider degrees of freedom df = n - k – 1 c. Calculating the value of t statistic t count can be found using the formula: t = bi SE bi Where: bi = regression coefficient SE bi = standard error of regression coefficient d. According to Ghozali stated in Jama, 2013: 46 the calculation criteria are: 1 If -t table t test t table, then Ho is accepted and Hais rejected,it means there is no influence betweenindepedent variable X toward dependent variable Y. 2 If t test t table, or -t test -t table then Ho is rejectedand Ha is accepted, it means there is influence between independent variable X toward dependent variable Y. While criteria in making a decision with significancy α = 0,05 are: 61 1 If probability α 0.05, so Ho acceptedand Ha rejected. 2 If probability α 0.05, so Ho rejected and Ha accepted.

6. Simultaneous Testing F Test