You are on page 1of 4

REGRESSION ANALYSIS

PROBLEM:
To understand the role of quality and price in influencing the patronage of drugstores, 14 major
stores in a large metropolitan area were rated in terms of preference to shop, quality of
merchandise and fair pricing. All the ratings were obtained on an 11 point scale, with higher
numbers indicating more positive ratings.

Table: 1

Correlations

QUALITY PRICE
QUALITY Pearson Correlation 1.000 .531
Sig. (2-tailed) . .051
N 14 14
PRICE Pearson Correlation .531 1.000
Sig. (2-tailed) .051 .
N 14 14

Table 2

Model Summary

Std. Error
Adjusted of the
Model R R Square R Square Estimate
1 .975a .951 .942 .7618
a. Predictors: (Constant), PRICE, QUALITY

Table 3

ANOVAb

Sum of Mean
Model Squares df Square F Sig.
1 Regression 122.831 2 61.415 105.826 .000a
Residual 6.384 11 .580
Total 129.214 13
a. Predictors: (Constant), PRICE, QUALITY
b. Dependent Variable: PREF
Table 4

Coefficientsa

Standardi
zed
Unstandardized Coefficien
Coefficients ts
Model B Std. Error Beta t Sig.
1 (Constant) .535 .471 1.136 .280
QUALITY .976 .097 .798 10.096 .000
PRICE .251 .071 .278 3.522 .005
a. Dependent Variable: PREF

ANALYSIS:
Initial regression model:

  
Y  a  b1 x1  b2 x 2  

Y  Store Preference
x1  Quality
x 2  Price

GLOBAL TEST: Testing whether the regression model is valid or not. This test signifies the
ability of the independent variables to explain the behavior of the dependent variable.

HYPOTHESIS
Null Hypothesis H 0 : R 2  0
Alternative Hypothesis H 1 : R 2  0

Considering Table 2, we can see the adjusted r 2, which is the coefficient of determination, is
0.942. The value indicates that the independent variables (Quality and Price) have strong
influence (94.2%) of determining the dependent variable (Store Patronage). That is 94.2%
variation can be explained by Price and Quality. Considering the ANOVA test (Table 3), we can
measure whether the null hypothesis can be accepted or rejected. That indicates whether we
can accept the association of the regression model. In Table 2, we can see that F value
(105.826) significantly rejects the null hypothesis. That means the independent variables are
significantly capable of estimating the store patronage.

MULTICOLLINERITY TEST
In the beginning we need to get a look into the multicollinerity testing of the independent
variables. In this analysis we have taken price and quality are independent variable which might
have influence the preference of shopping from distinct store. The multicollinerity testing It is
required to take out one variable from the regression model if the correlation coefficient of those
variables ranges between -0.70 to 0.70. The Table 1 shows that the correlation coefficient of
independent variables PRICE and QUALITY lies within acceptable level (0.53). That means that
we can put all independent variables for analysis.

EVALUATING INDIVIDUAL REGRESSION COEFFICIENT


This step is to test whether the independent variables individually have any influence on
dependent variable. That means, if a  coefficient is equal to 0, it implies that the particular

independent variable is of no value in explaining any variation in the dependent variable. In that
situation we will eliminate that variable from the regression equation. We need to conduct two
separate tests of hypothesis for two independent variables;

For Quality: H 0 :  1  0
H 1 : 1  0

For Price: H 0 :  2  0
H1 :  2  0

Table 4 of the regression analysis depicts that t value of two independent variables (X 1 and X2)
are in rejection region. The t value of X1 and X2 is 10.096 and 3.522 with significance level less
than 0.01. The result indicates that the beta coefficient is not equal to ZERO. Table 4 also
denotes the unstandardized beta coefficients. The unstandardized beta coefficient of X 1 and X2
is 0.976 and 0.251. The value indicates that one unit attitude change to the quality (X 1) will
change the store patronage (Y) by 0.976 and one unit attitude change to price (X 2) will change
the store patronage (Y) by 0.251. The final regression model is given below:
Y  0.535  0.976 x1  0.251 x 2

You might also like