Computers R Us Executive Summary
Data Analysis Report Case Study – Computers R Us
Executive summary
This report aims to figure out two basic questions, current consumers’ satisfaction and strategy that would be most potent to increase overall satisfaction. At the beginning, a survey of three parts was designed and conducted across different age groups.420 samples were collected and coding and editing of a variety of data ensued. By resorting to hypothesis test, regression model, this report found out that current consumers’ satisfaction level is much lower than management expected and two gender groups show significantly different satisfaction .Also, consumer satisfaction was similar across five age groups. In respect of determinants of consumer satisfaction, satisfaction …show more content…
with response time, satisfaction with the level of advice from staff at call centre, satisfaction with the level of communication and satisfaction with loyalty rewards program from consumers are all key contributors while only satisfaction with the level of communication has positive relationship with overall consumer satisfaction. Thus, in order to make effective strategy to improve overall satisfaction, Computer R Us should increase methods of communication.
Introduction
Computer R Us, whose main business focuses on manufacture and retail of computers, recently set up a division called Completecare. This division offers consumers service and repair for its computers. However, in its everyday operation, R Us has received many complaints from consumers because the division is suffering from such problems as lack of trained work staff in the call centre, distribution problem and availability of computer parts.
This report aims to investigate what current level of satisfaction is and which strategy is the most effective to improve the overall satisfaction. In order to set out most potent strategy to improve overall satisfaction, this report studies that satisfaction with response time, satisfaction with the level of advice from staff at call centre ,satisfaction with the level of communication and satisfaction with loyalty rewards program from consumers which one contributes significantly to overall satisfaction and then recommends respective strategy.
Research Design
In primary research, a consumer satisfaction survey which consists of three parts, personal information, current satisfaction and determinants of consumer satisfaction, has been designed. This survey was randomly distributed among 500 consumers of R Us. This sample size is sufficient to represent the population and the sampling is at random. Out of 500 copies of questionnaires sent, 420 samples were collected .This set of data is the basis to further research.
In the process of collecting data, there are several human ethical considerations .Consent with this survey should be given by participants before data process begins. Make sure consumers everyday life would not be disturbed by this sample survey both physically and emotionally. The most salient ethical values implicated by the use of human participants in research are beneficence, fidelity and trust within the fiduciary investigator/participant relationship, personal dignity, and competent decision making and the privacy of personal information. (Kapp, 2006)In our surveying, consumers were able to respond anonymously, which protects the privacy of personal information. Even in reporting results, the private information provided by those 420 consumers must not be disclosed.
Analysis
Six hypotheses tests were conducted to work out current consumer satisfaction and the determinants of overall satisfaction. From appendix, it can be seen that current consumer satisfaction with the business doesn’t reach management’s expectation (6 out of 10)by using p value one tail though analysis of two tails are illustrated .It’s necessary to take right actions to boost consumers’ satisfaction with profound insight into the determinants of overall satisfaction. Also it’s worth noting that male satisfaction differs from female satisfaction with means of 3.626 and 5.885 respectively, indicating that female consumers are much more satisfied with current situation more than male are. With regards to satisfaction across the specified age groups, the hypothesis test shows no difference at all. By means of Chi-Square test, we conclude that composition of male and female is quite the same in those five age groups. In addition, consumer’s satisfactions in response to the initiatives of ‘decreasing response times in the CompleteCare division’ and the ‘new loyalty rewards program’ are quite different.
With respect to the determinants of overall satisfaction, satisfaction with response time, satisfaction with the level of advice from staff at call centre, satisfaction with the level of communication and satisfaction with loyalty rewards program from consumers all are significant factors to overall satisfaction, evidenced by the t-statistics of table 6. From the regression model, it can be seen that coefficient of determination reaches 98% showing that the obtained model explains all those variables quite well. However, it is odd that only consumer’s satisfaction with the level of communication has the positive relationship with the overall satisfaction since its coefficient is 0.167 while the other three have negative relationship with overall satisfaction. This result is crucial to determine which strategy is the most effective one, in which management is most interested.
Recommendations
In short, current overall consumer satisfaction is below the expected level and different between genders. The key determinants to consumers’ overall satisfaction are satisfaction with response time, satisfaction with the level of advice from staff at call centre, satisfaction with the level of communication and satisfaction with loyalty rewards program from consumers.
It’s of great necessity to improve the overall consumers’ satisfaction by increasing methods of communication between Computer R Us and its consumers. Even though strategies such as decrease in response times, training and continuous education for staff and a new loyalty rewards program could improve consumers’ satisfaction with response time, level of advice from call centre and loyalty rewards program, those strategies might dent the overall consumers’ satisfaction due to the negative relationships. Therefore, R Us should focus on methods of communisation. In addition, male consumers should be placed more importance because they showed less satisfaction.
Appendix
1.
H0: The current level of customer satisfaction is not different from management’s goal of 6 out of 10.
HA: The current level of customer satisfaction differs from 6 out of 10.
One sample t-test is used to prove whether the current level of customer satisfaction differs from 6 out of 10 since it is an appropriate method to test the difference between one variable and a pre-determined mean (Zikmund, 2012).The result is displayed below;
Table 1: t-Test: Paired Two Sample for Means
Variable 1
Variable 2
Mean
4.561904762
6
Variance
6.127423571
0
Observations
420
420
Pearson Correlation
#DIV/0!
Hypothesized Mean Difference
0
df
419
t Stat
-11.90620462
P(T<=t) one-tail
1.17648E-28
t Critical one-tail
1.64849841
P(T<=t) two-tail
2.35295E-28
t Critical two-tail
1.965641842
From table 1, we can see that the p-value (2.35295E-28) two-tail is much less than 5%, the null hypothesis should not be accepted. The mean of current level of customer satisfaction (4.562) differs from the minimum level of 6 out of 10.
2. H0: there is no difference between the overall satisfaction of male and female customers. HA: overall satisfaction of male is different from that of female.
As these two samples are independent and come from the same population, two-sample t-test is appropriate to prove if there is a difference. The result is listed below:
Table2: t-Test: Two-Sample Assuming Equal Variances
Variable 1
Variable 2
Mean
3.62601626
5.885057471
Variance
4.65956529
5.23526676
Observations
246
174
Pooled Variance
4.89783408
Hypothesized Mean Difference
0
df
418
t Stat
-10.30480201
P(T<=t) one-tail
1.21806E-22
t Critical one-tail
1.648507149
P(T<=t) two-tail
2.43611E-22
t Critical two-tail
1.965655464
It is evident that t Statistics 10.304 is much larger than t Critical two-tail 1.965(5% significance level).Thus the null hypothesis cannot be accepted and the conclusion can be drawn that there is a difference between the overall satisfaction of male and female customers.
3. H0: There is no differences in the overall customer satisfaction across the following age groups. HA: At least the overall customer satisfaction of one age group is different from other age groups.
Since the comparisons of the overall customer satisfaction across the following age groups, ANOVA is necessary to tell if there is are differences. The result is below:
Table3 Anova: Single Factor
SUMMARY
AGE Groups
Count
Sum
Average
Variance
<=20
47
182
3.87234
6.983349
21-30
109
513
4.706422 …show more content…
7.061162
31-40
105
473
4.504762
6.367766
41-50
107
492
4.598131
5.186034
>=51
52
256
4.923077
4.699849
ANOVA
Source of Variation
SS
df
MS
F
P-value
F crit
Between Groups
31.89138
4
7.972844
1.304962
0.267408
2.393438
Within Groups
2535.499
415
6.109636
Total
2567.39
419
From table 3, we can see that F value (1.305) is smaller than F critical value that is 2.393.Thus the null hypothesis should be accepted. Therefore, there is no difference in the overall customer satisfaction across the following age groups.
4. H0: There is no difference in the gender compositions across five age groups. HA: There is a difference in the gender compositions across five age groups.
Because the this set of data is categorical data,chi-square test is employed .the result is shown as follows,
Observed value
<=20
21-30
31-40
41-50
>=51
SUM
MALE
27
47
43
41
23
181
FEMALE
20
62
62
66
29
239
SUM
47
109
105
107
52
420
Expected value
MALE
20.25
46.97
45.25
46.11
22.41
FEMALE
26.75
62.03
59.75
60.89
29.59
CHI-TEST P-value
0.270557
It can be seen from Table 4 that p-value of chi-test is 0.270557, even larger than the significance level 10%.Thus we fail to reject the null hypothesis and therefore, no difference exists in the gender compositions across five age groups.
5. H0: There is no difference in consumer’s satisfaction in response to the initiatives of ‘decreasing response times in the CompleteCare division’ and the ‘new loyalty rewards program’. HA: There is a difference in consumer’s satisfaction in response to the initiatives of ‘decreasing response times in the CompleteCare division’ and the ‘new loyalty rewards program’.
As these two samples are independent and come from the same population, two-sample t-test is appropriate to prove if there is a difference. The result is listed below,
Table5 t-Test: Two-Sample Assuming Equal Variances
Variable 1
Variable 2
Mean
4.371428571
5.645238095
Variance
6.348585066
7.842817366
Observations
420
420
Pooled Variance
7.095701216
Hypothesized Mean Difference
0
df
838
t Stat
-6.929732637
P(T<=t) one-tail
4.20464E-12
t Critical one-tail
1.646673991
P(T<=t)
two-tail
8.40928E-12
t Critical two-tail
1.962798881
It is obvious that P value two-tail (8.40928E-12) is way less than 1% significance level and thus the null hypothesis can be rejected. The conclusion can be drawn that there is a difference in consumer’s satisfaction in response to the initiatives of ‘decreasing response times in the CompleteCare division’ and the ‘new loyalty rewards program’.
6. H0: There is no relationship between any initiatives and overall consumer satisfaction. HA: There is a relationship between any initiatives and overall consumer satisfaction.
To determine any of the initiatives proposed by management related to the overall satisfaction of Computers R Us customers, regression model is demanded. By setting the overall satisfaction as the dependent variable, satisfaction with response time, satisfaction with the level of advice from staff at call centre and the level of communication and satisfaction with loyalty program as independent variables, we can get the result below,
SUMMARY OUTPUT
Regression Statistics
Multiple R
0.988
R Square
0.977
Adjusted R Square
0.977
Standard Error
0.379
Observations
420.000
ANOVA
df
SS
MS
F
Significance F
Regression
4.000
2507.8
627.0
4369.7
0
Residual
415.000
59.5
0.1
Total
419.000
2567.4
Coefficients
Standard Error t Stat
P-value
Lower 95%
Upper 95%
Intercept
9.121
0.0505
180.5457
0.0000
9.0214
9.2200
response time
-0.085
0.0114
-7.4327
0.0000
-0.1074
-0.0625 level of advice
-0.234
0.0280
-8.3374
0.0000
-0.2889
-0.1787 level of commun.
0.167
0.0132
12.5860
0.0000
0.1405
0.1925 loyalty rewards
-0.676
0.0239
-28.3216
0.0000
-0.7234
-0.6295 It can be seen from the ANOVA that P value of F statistics is 0, indicating that there is a linear relationship between independent variables and dependent variable. To see if the each independent variable has significant impact on the dependent variable (overall satisfaction), t-stat is taken. With respect to independent variables, all p-values are 0, indicating that satisfaction with response time, satisfaction with level of advice from staff, satisfaction with level of communication and satisfaction with loyalty rewards program all determinants to the overall consumer satisfaction. Thus, the four initiatives proposed by management are related to the overall satisfaction of Computers R Us customers. It is worth noticing that only satisfaction with level of communication positively contributes to the overall consumer satisfaction while the other three variables have negative relationship with overall satisfaction.
References
Azzalini, A. (2012). Data Analysis and Data Mining: An Introduction. Oxford University Press Inc.
Kapp, M. B. (2006). Ethical and legal issues in research involving human subjects: do you want a piece of me? Journal of Clinical Pathology, 335-339.
Zikmund, W. B. (2012). Business research methods (9th ed.). Cengage Learning.
Executive summary
This report aims to figure out two basic questions, current consumers’ satisfaction and strategy that would be most potent to increase overall satisfaction. At the beginning, a survey of three parts was designed and conducted across different age groups.420 samples were collected and coding and editing of a variety of data ensued. By resorting to hypothesis test, regression model, this report found out that current consumers’ satisfaction level is much lower than management expected and two gender groups show significantly different satisfaction .Also, consumer satisfaction was similar across five age groups. In respect of determinants of consumer satisfaction, satisfaction …show more content…
with response time, satisfaction with the level of advice from staff at call centre, satisfaction with the level of communication and satisfaction with loyalty rewards program from consumers are all key contributors while only satisfaction with the level of communication has positive relationship with overall consumer satisfaction. Thus, in order to make effective strategy to improve overall satisfaction, Computer R Us should increase methods of communication.
Introduction
Computer R Us, whose main business focuses on manufacture and retail of computers, recently set up a division called Completecare. This division offers consumers service and repair for its computers. However, in its everyday operation, R Us has received many complaints from consumers because the division is suffering from such problems as lack of trained work staff in the call centre, distribution problem and availability of computer parts.
This report aims to investigate what current level of satisfaction is and which strategy is the most effective to improve the overall satisfaction. In order to set out most potent strategy to improve overall satisfaction, this report studies that satisfaction with response time, satisfaction with the level of advice from staff at call centre ,satisfaction with the level of communication and satisfaction with loyalty rewards program from consumers which one contributes significantly to overall satisfaction and then recommends respective strategy.
Research Design
In primary research, a consumer satisfaction survey which consists of three parts, personal information, current satisfaction and determinants of consumer satisfaction, has been designed. This survey was randomly distributed among 500 consumers of R Us. This sample size is sufficient to represent the population and the sampling is at random. Out of 500 copies of questionnaires sent, 420 samples were collected .This set of data is the basis to further research.
In the process of collecting data, there are several human ethical considerations .Consent with this survey should be given by participants before data process begins. Make sure consumers everyday life would not be disturbed by this sample survey both physically and emotionally. The most salient ethical values implicated by the use of human participants in research are beneficence, fidelity and trust within the fiduciary investigator/participant relationship, personal dignity, and competent decision making and the privacy of personal information. (Kapp, 2006)In our surveying, consumers were able to respond anonymously, which protects the privacy of personal information. Even in reporting results, the private information provided by those 420 consumers must not be disclosed.
Analysis
Six hypotheses tests were conducted to work out current consumer satisfaction and the determinants of overall satisfaction. From appendix, it can be seen that current consumer satisfaction with the business doesn’t reach management’s expectation (6 out of 10)by using p value one tail though analysis of two tails are illustrated .It’s necessary to take right actions to boost consumers’ satisfaction with profound insight into the determinants of overall satisfaction. Also it’s worth noting that male satisfaction differs from female satisfaction with means of 3.626 and 5.885 respectively, indicating that female consumers are much more satisfied with current situation more than male are. With regards to satisfaction across the specified age groups, the hypothesis test shows no difference at all. By means of Chi-Square test, we conclude that composition of male and female is quite the same in those five age groups. In addition, consumer’s satisfactions in response to the initiatives of ‘decreasing response times in the CompleteCare division’ and the ‘new loyalty rewards program’ are quite different.
With respect to the determinants of overall satisfaction, satisfaction with response time, satisfaction with the level of advice from staff at call centre, satisfaction with the level of communication and satisfaction with loyalty rewards program from consumers all are significant factors to overall satisfaction, evidenced by the t-statistics of table 6. From the regression model, it can be seen that coefficient of determination reaches 98% showing that the obtained model explains all those variables quite well. However, it is odd that only consumer’s satisfaction with the level of communication has the positive relationship with the overall satisfaction since its coefficient is 0.167 while the other three have negative relationship with overall satisfaction. This result is crucial to determine which strategy is the most effective one, in which management is most interested.
Recommendations
In short, current overall consumer satisfaction is below the expected level and different between genders. The key determinants to consumers’ overall satisfaction are satisfaction with response time, satisfaction with the level of advice from staff at call centre, satisfaction with the level of communication and satisfaction with loyalty rewards program from consumers.
It’s of great necessity to improve the overall consumers’ satisfaction by increasing methods of communication between Computer R Us and its consumers. Even though strategies such as decrease in response times, training and continuous education for staff and a new loyalty rewards program could improve consumers’ satisfaction with response time, level of advice from call centre and loyalty rewards program, those strategies might dent the overall consumers’ satisfaction due to the negative relationships. Therefore, R Us should focus on methods of communisation. In addition, male consumers should be placed more importance because they showed less satisfaction.
Appendix
1.
H0: The current level of customer satisfaction is not different from management’s goal of 6 out of 10.
HA: The current level of customer satisfaction differs from 6 out of 10.
One sample t-test is used to prove whether the current level of customer satisfaction differs from 6 out of 10 since it is an appropriate method to test the difference between one variable and a pre-determined mean (Zikmund, 2012).The result is displayed below;
Table 1: t-Test: Paired Two Sample for Means
Variable 1
Variable 2
Mean
4.561904762
6
Variance
6.127423571
0
Observations
420
420
Pearson Correlation
#DIV/0!
Hypothesized Mean Difference
0
df
419
t Stat
-11.90620462
P(T<=t) one-tail
1.17648E-28
t Critical one-tail
1.64849841
P(T<=t) two-tail
2.35295E-28
t Critical two-tail
1.965641842
From table 1, we can see that the p-value (2.35295E-28) two-tail is much less than 5%, the null hypothesis should not be accepted. The mean of current level of customer satisfaction (4.562) differs from the minimum level of 6 out of 10.
2. H0: there is no difference between the overall satisfaction of male and female customers. HA: overall satisfaction of male is different from that of female.
As these two samples are independent and come from the same population, two-sample t-test is appropriate to prove if there is a difference. The result is listed below:
Table2: t-Test: Two-Sample Assuming Equal Variances
Variable 1
Variable 2
Mean
3.62601626
5.885057471
Variance
4.65956529
5.23526676
Observations
246
174
Pooled Variance
4.89783408
Hypothesized Mean Difference
0
df
418
t Stat
-10.30480201
P(T<=t) one-tail
1.21806E-22
t Critical one-tail
1.648507149
P(T<=t) two-tail
2.43611E-22
t Critical two-tail
1.965655464
It is evident that t Statistics 10.304 is much larger than t Critical two-tail 1.965(5% significance level).Thus the null hypothesis cannot be accepted and the conclusion can be drawn that there is a difference between the overall satisfaction of male and female customers.
3. H0: There is no differences in the overall customer satisfaction across the following age groups. HA: At least the overall customer satisfaction of one age group is different from other age groups.
Since the comparisons of the overall customer satisfaction across the following age groups, ANOVA is necessary to tell if there is are differences. The result is below:
Table3 Anova: Single Factor
SUMMARY
AGE Groups
Count
Sum
Average
Variance
<=20
47
182
3.87234
6.983349
21-30
109
513
4.706422 …show more content…
7.061162
31-40
105
473
4.504762
6.367766
41-50
107
492
4.598131
5.186034
>=51
52
256
4.923077
4.699849
ANOVA
Source of Variation
SS
df
MS
F
P-value
F crit
Between Groups
31.89138
4
7.972844
1.304962
0.267408
2.393438
Within Groups
2535.499
415
6.109636
Total
2567.39
419
From table 3, we can see that F value (1.305) is smaller than F critical value that is 2.393.Thus the null hypothesis should be accepted. Therefore, there is no difference in the overall customer satisfaction across the following age groups.
4. H0: There is no difference in the gender compositions across five age groups. HA: There is a difference in the gender compositions across five age groups.
Because the this set of data is categorical data,chi-square test is employed .the result is shown as follows,
Observed value
<=20
21-30
31-40
41-50
>=51
SUM
MALE
27
47
43
41
23
181
FEMALE
20
62
62
66
29
239
SUM
47
109
105
107
52
420
Expected value
MALE
20.25
46.97
45.25
46.11
22.41
FEMALE
26.75
62.03
59.75
60.89
29.59
CHI-TEST P-value
0.270557
It can be seen from Table 4 that p-value of chi-test is 0.270557, even larger than the significance level 10%.Thus we fail to reject the null hypothesis and therefore, no difference exists in the gender compositions across five age groups.
5. H0: There is no difference in consumer’s satisfaction in response to the initiatives of ‘decreasing response times in the CompleteCare division’ and the ‘new loyalty rewards program’. HA: There is a difference in consumer’s satisfaction in response to the initiatives of ‘decreasing response times in the CompleteCare division’ and the ‘new loyalty rewards program’.
As these two samples are independent and come from the same population, two-sample t-test is appropriate to prove if there is a difference. The result is listed below,
Table5 t-Test: Two-Sample Assuming Equal Variances
Variable 1
Variable 2
Mean
4.371428571
5.645238095
Variance
6.348585066
7.842817366
Observations
420
420
Pooled Variance
7.095701216
Hypothesized Mean Difference
0
df
838
t Stat
-6.929732637
P(T<=t) one-tail
4.20464E-12
t Critical one-tail
1.646673991
P(T<=t)
two-tail
8.40928E-12
t Critical two-tail
1.962798881
It is obvious that P value two-tail (8.40928E-12) is way less than 1% significance level and thus the null hypothesis can be rejected. The conclusion can be drawn that there is a difference in consumer’s satisfaction in response to the initiatives of ‘decreasing response times in the CompleteCare division’ and the ‘new loyalty rewards program’.
6. H0: There is no relationship between any initiatives and overall consumer satisfaction. HA: There is a relationship between any initiatives and overall consumer satisfaction.
To determine any of the initiatives proposed by management related to the overall satisfaction of Computers R Us customers, regression model is demanded. By setting the overall satisfaction as the dependent variable, satisfaction with response time, satisfaction with the level of advice from staff at call centre and the level of communication and satisfaction with loyalty program as independent variables, we can get the result below,
SUMMARY OUTPUT
Regression Statistics
Multiple R
0.988
R Square
0.977
Adjusted R Square
0.977
Standard Error
0.379
Observations
420.000
ANOVA
df
SS
MS
F
Significance F
Regression
4.000
2507.8
627.0
4369.7
0
Residual
415.000
59.5
0.1
Total
419.000
2567.4
Coefficients
Standard Error t Stat
P-value
Lower 95%
Upper 95%
Intercept
9.121
0.0505
180.5457
0.0000
9.0214
9.2200
response time
-0.085
0.0114
-7.4327
0.0000
-0.1074
-0.0625 level of advice
-0.234
0.0280
-8.3374
0.0000
-0.2889
-0.1787 level of commun.
0.167
0.0132
12.5860
0.0000
0.1405
0.1925 loyalty rewards
-0.676
0.0239
-28.3216
0.0000
-0.7234
-0.6295 It can be seen from the ANOVA that P value of F statistics is 0, indicating that there is a linear relationship between independent variables and dependent variable. To see if the each independent variable has significant impact on the dependent variable (overall satisfaction), t-stat is taken. With respect to independent variables, all p-values are 0, indicating that satisfaction with response time, satisfaction with level of advice from staff, satisfaction with level of communication and satisfaction with loyalty rewards program all determinants to the overall consumer satisfaction. Thus, the four initiatives proposed by management are related to the overall satisfaction of Computers R Us customers. It is worth noticing that only satisfaction with level of communication positively contributes to the overall consumer satisfaction while the other three variables have negative relationship with overall satisfaction.
References
Azzalini, A. (2012). Data Analysis and Data Mining: An Introduction. Oxford University Press Inc.
Kapp, M. B. (2006). Ethical and legal issues in research involving human subjects: do you want a piece of me? Journal of Clinical Pathology, 335-339.
Zikmund, W. B. (2012). Business research methods (9th ed.). Cengage Learning.