Beruflich Dokumente
Kultur Dokumente
Question 1: Is there a relationship among the variables measuring different aspects of customer
satisfaction?
1. Run a Pearson correlation matrix using delivery reliability, product satisfaction, technical supp
ort, sales satisfaction, overall satisfaction in January and overall satisfaction in June.
Correlations
Delivery
Product
Tech
Sales
Overall
Overall
Reliability
Satisfaction
Support
Satisfaction
Satisfaction
Satisfaction
in January
in June
Satisfaction
Pearson
Correlation
Delivery
Reliability
Correlation
Sig. (2tailed)
N
Pearson
Tech Support
Satisfaction
Correlation
Sig. (2tailed)
N
Pearson
Sales
Satisfaction
.112
.206
.628**
.180
.002
.439
.152
.000
50
50
50
50
50
50
.193
.317*
.726**
.195
.484**
.025
.000
.174
.000
tailed)
Pearson
Satsifaction
.436**
Sig. (2-
Product
.193
Correlation
Sig. (2tailed)
N
Pearson
Overall
Correlation
Satisfaction
Sig. (2-
in January
tailed)
N
.180
50
50
50
50
50
50
.436**
.317*
.133
.340*
.555**
.002
.025
.356
.016
.000
50
50
50
50
50
50
.112
.726**
.133
.173
.326*
.439
.000
.356
.229
.021
50
50
50
50
50
50
.206
.195
.340*
.173
.479**
.152
.174
.016
.229
50
50
50
50
.000
50
50
Pearson
Overall
Correlation
Satisfaction
Sig. (2-
in June
tailed)
N
.628**
.484**
.555**
.326*
.479**
.000
.000
.000
.021
.000
50
50
50
50
50
50
2. Create a scatterplot for the following pairs: (1) delivery reliabilityoverall satisfaction in June
; (2) product satisfactionoverall satisfaction in June.
Scatter Plot
Delivery reliability overall satisfaction in June
3. Report the descriptive statistics, assumptions tests, as well as tests of statistical significance id
entify of positive and negative relationships.
Descriptive Statistics
Descriptive Statistics
N Minimum Maximum
Delivery Reliability
50
1.0
7.0
Product Satisfaction
50
3.0
8.0
Overall Satisfaction in June 50
3.0
7.0
Valid N (listwise)
50
The mean delivery reliability score is 4.34 with a standard deviation of 1.6734. This indic
ates that the consumers rated the delivery reliability as better. The minimum and maximum recor
ded delivery reliability scores are 1 and 7 respectively. The mean product satisfaction score is 5.3
4 with a standard deviation of 1.1537. This indicates that the consumers rated the product satisfa
ction as good. The minimum and maximum recorded product satisfaction scores are 3 and 8 resp
ectively. The mean overall satisfaction in June is 4.7 with a standard deviation of 0.9530. This in
dicates that the consumers rated the overall satisfaction in June as better. The minimum and maxi
mum recorded delivery reliability scores are 3 and 7 respectively.
To test whether there is significant relationship between Delivery Reliability and Overall
Satisfaction score, we perform Pearson Correlation test for significance. The null and alternate h
ypotheses are given below:
Null Hypothesis: H0: = 0
That is, there is no relationship between Delivery Reliability and Overall Satisfaction
Alternate Hypothesis: Ha: 0
That is, there is a significant relationship between Delivery Reliability and Overall Satisf
action
Level of Significance:
Let the level of significance be = 0.05
Sample size = n = 50
Correlation Coefficient = = 0.628
Test Statistic:
The test statistic is:
n2
50 2
0.628
8.903
2
1
1 0.6282
Thus, the value of t test statistic is 8.903.
Sample size = n = 50
Correlation Coefficient = = 0.555
Test Statistic:
The test statistic is:
n2
50 2
0.555
8.329
2
1
1 0.5552
Thus, the value of t test statistic is 8.328.
Model
Unstandardized
Coefficients
B
Std. Error
Coefficientsa
Standardized
Coefficients
Beta
(Constant)
3.147
.297
1 Delivery
.358
.064
Reliability
a. Dependent Variable: Overall Satisfaction in June
Sig.
10.595 .000
.628
5.596 .000
95.0% Confidence
Interval for B
Lower
Upper
Bound
Bound
2.550
3.744
.229
.486
2. Report the descriptive statistics, assumptions tests (scatter plots), as well as tests of statistical s
ignificance.
The simple regression equation to predict overall satisfaction in June using delivery reliab
ility as independent variable is:
N
50
50
50
ANOVAa
Model
Sum of Squares df Mean Square
F
Sig.
Regression
17.569 1
17.569 31.314 .000b
1 Residual
26.931 48
.561
Total
44.500 49
Regression
23.674 2
11.837 26.713 .000c
2 Residual
20.826 47
.443
Total
44.500 49
Model
Unstandardized
Coefficients
B
Std. Error
(Constant)
1 Delivery
Reliability
(Constant)
Delivery
2 Reliability
3.147
.297
.358
.064
1.662
.479
.316
.058
Coefficientsa
Standardized
Coefficients
Beta
Product
.312
.084
Satsifaction
a. Dependent Variable: Overall Satisfaction in June
Sig.
10.595 .000
.628
95.0% Confidence
Interval for B
Lower
Upper
Bound
Bound
2.550
3.744
5.596 .000
.229
.486
3.467 .001
.697
2.626
.556
5.464 .000
.200
.433
.377
3.712 .001
.143
.481
2. Report the descriptive statistics, assumptions tests (scatter plots), as well as tests of statistical s
ignificance. Write a brief conclusion statement summarizing your results. What can you tell this
manufacturing company about the relationship among satisfaction variables? Are there any areas
they need to improve? Does adding a second variable to the regression equation increase predicti
on of customer satisfaction?
The simple regression equation to predict overall satisfaction in June using delivery reliab
ility and Product Satisfaction as independent variables is:
Overall Satisfaction in June = 3.147+ 0.316 * Delivery Reliability + 0.312 * Product Satisfaction
Going through the ANOVA table, we see that the value of F test statistic is 26.713 and its
corresponding p value is 0.000. Since this p value is less than 0.05, we conclude that the esti
mated regression equation is good fit in predicting the dependent variable overall satisfaction in J
une
The coefficient of determination is 0.532. This indicates that 53.2% of the variation in the
dependent variable is explained by the regression model, while the remaining 46.5% left unexpl
ained.
On comparing the two regression models, we see that the regression model used two inde
pendent variables delivery reliability and product satisfaction to predict overall satisfaction in Ju
ne is better when compared with the regression model used deliver satisfaction as independent va
riable to predict overall satisfaction in June.
Going through normal probability plot, we see that the plotted points fall close to trend li
ne, indicating that the distribution of errors follows normal. Going through the scatter plot of resi
duals, we see that there exists random pattern in the distribution, indicating that the assumption o
f errors are independently distributed is satisfied
Test Statistics
Type of Purchasing
Chi-Square
.640a
df
2
Asymp. Sig.
.726
a. 0 cells (0.0%) have expected frequencies less than 5. The minimum expected cell frequency
is 16.7.
From the above SPSS output, we see that the value of chi square test statistic is 0.64
and its corresponding p value is 0.726. Since the p value of chi square test statistic is
greater than 0.05, there is no sufficient evidence to reject the null hypothesis at 5% level of
significance. Therefore, we conclude that the distribution of customers (private label, brand
label, or both) do not differ from what one would expect by chance