Analysis Of Customer Attitudes At Retail Surge

Problem Definition

Technological advancement in the field of business coupled by global inflation has made business world to be a ground of competition. Many companies in various industries which have not been able to keep up with the competition in the market have been forced to collapse. For this reason, majority of businesses have strove to embrace various technological advancements which have made business operations easier. For example in tours and travel industry, customers are now able to book their tickets online at the comfort of their couches instead of appearing physically in the offices. Such kind is what technology can do. It also enables the company to understand the market better by analysing customer centric data. Retail Surge, an online company, deals in clothes and shoes for both men and women.  It has also been experiencing a down surge in revenues in due to global dynamics in the market. For this reason, the business collected customer data to be analysed so as to be able to better understand the market and their customers.  

• The product category is made the most profit

A graph was used to present the products and their profits. This descriptive was employed since it is visual thus easy to interpret.

• Product category that cost the most

A graph was used to present the products at Retail Surge and their costs. This descriptive was employed since it is visual thus easy to interpret.  

• Is there a difference in payment methods?

Independent sample t-test was used in this test because there were only two independent variables that were being compared (PayPal and credit card).

• Are there any differences in the user groups on all of the customer attitudes? (6 outcomes)

Analysis of variance (ANOVA) was used in this test because there were more than two variables that were being compared. These were light, medium and heavy users customers at Retail Surge.

• Are there any differences in gender on all of the customer attitudes?

Independent sample t-test was used in this test because there were only two independent variables that were being compared (males and females).

It can be observed from the results in the bar graph above that customised products are doing well in terms of profit. The mean total profit foe customized products was 25 dollars. The men’s shoes followed with a mean total profit of 16 dollars while the third product was girls’ shoes that made a mean total profit of 7 dollars.

It can be observed from the results in the bar graph above that customised products are doing well in terms of mean cost of goods. The mean cost of goods for customized products was about 10 dollars. The girls’ shoes followed with a mean cost of goods of 8 dollars while the third product was women’s shoes that made a mean cost of goods of 5 dollars.

Descriptives
	N	Mean	Std. Deviation	Std. Error	95% Confidence Interval for Mean	Minimum	Maximum
Lower Bound	Upper Bound
Knowledge of the company	Light Users	104	2.85	1.711	.168	2.51	3.18	1	6
Medium Users	204	4.90	1.664	.116	4.67	5.13	2	7
Heavy Users	284	6.45	.601	.036	6.38	6.52	5	7
Total	592	5.28	1.840	.076	5.14	5.43	1	7
Satisfaction with the company	Light Users	104	2.54	1.157	.113	2.31	2.76	1	6
Medium Users	204	5.39	1.033	.072	5.25	5.53	2	7
Heavy Users	284	6.07	.776	.046	5.98	6.16	2	7
Total	592	5.22	1.585	.065	5.09	5.34	1	7
Preference for Nike	Light Users	104	2.46	1.507	.148	2.17	2.75	1	6
Medium Users	204	2.82	1.572	.110	2.61	3.04	1	5
Heavy Users	284	4.45	1.654	.098	4.26	4.64	1	7
Total	592	3.54	1.826	.075	3.39	3.69	1	7
Purchase Intention for Nike	Light Users	100	4.16	1.835	.184	3.80	4.52	1	7
Medium Users	204	4.04	1.893	.133	3.78	4.30	2	7
Heavy Users	284	5.01	1.195	.071	4.87	5.15	3	7
Total	588	4.53	1.648	.068	4.40	4.66	1	7
Would recommend company to a friend	Light Users	104	3.46	1.131	.111	3.24	3.68	2	6
Medium Users	204	4.92	.494	.035	4.85	4.98	4	6
Heavy Users	284	6.35	.477	.028	6.29	6.40	6	7
Total	592	5.35	1.260	.052	5.25	5.45	2	7
Loyalty for Nike	Light Users	104	3.92	1.499	.147	3.63	4.21	2	6
Medium Users	204	4.14	1.563	.109	3.92	4.35	2	6
Heavy Users	284	3.92	1.575	.093	3.73	4.10	2	7
Total	592	3.99	1.559	.064	3.87	4.12	2	7

Test of Homogeneity of Variances
	Levene Statistic	df1	df2	Sig.
Knowledge of the company	Based on Mean	137.679	2	589	.000
Based on Median	43.167	2	589	.000
Based on Median and with adjusted df	43.167	2	366.541	.000
Based on trimmed mean	130.226	2	589	.000
Satisfacition with the company	Based on Mean	34.012	2	589	.000
Based on Median	19.318	2	589	.000
Based on Median and with adjusted df	19.318	2	543.743	.000
Based on trimmed mean	28.470	2	589	.000
Preference for Nike	Based on Mean	3.007	2	589	.050
Based on Median	2.032	2	589	.132
Based on Median and with adjusted df	2.032	2	534.316	.132
Based on trimmed mean	3.259	2	589	.039
Purchase Intention for Nike	Based on Mean	51.499	2	585	.000
Based on Median	48.655	2	585	.000
Based on Median and with adjusted df	48.655	2	555.807	.000
Based on trimmed mean	49.574	2	585	.000
Would recommend company to a friend	Based on Mean	155.697	2	589	.000
Based on Median	75.229	2	589	.000
Based on Median and with adjusted df	75.229	2	489.408	.000
Based on trimmed mean	152.051	2	589	.000
Loyalty for Nike	Based on Mean	.438	2	589	.645
Based on Median	1.134	2	589	.322
Based on Median and with adjusted df	1.134	2	466.890	.323
Based on trimmed mean	.333	2	589	.717

Table 1

In the test above, homogeneity of variance has been used to assess Lavene’s test. Lavene’s test uses the F-test to if the variance between variables is equal (null hypothesis) (Derrick, Toher, & White, 2017). Knowledge company has a critical value of 0.00 < 0.05. The null hypothesis is rejected thus it is concluded that the variance is significantly different across the variables. Another variable is ‘satisfaction with company”. It has a critical value of 0.00 < 0.05. The null hypothesis is rejected thus it is concluded that the variance is significantly different across the variables. However, ‘loyalty for nike’ and other items have critical values of that are greater than 0.05. This forces the research to fail to reject the null hypothesis  (Leigh, 2008) and conclude that there is equality of variance across these two groups in terms of customer attitudes. ANOVA
	Sum of Squares	df	Mean Square	F	Sig.
Knowledge of the company	Between Groups	1034.437	2	517.218	315.401	.000
Within Groups	965.888	589	1.640
Total	2000.324	591
Satisfaction with the company	Between Groups	959.259	2	479.630	538.032	.000
Within Groups	525.065	589	.891
Total	1484.324	591
Preference for Nike	Between Groups	461.224	2	230.612	89.966	.000
Within Groups	1509.803	589	2.563
Total	1971.027	591
Purchase Intention for Nike	Between Groups	129.379	2	64.690	25.830	.000
Within Groups	1465.070	585	2.504
Total	1594.449	587
Would recommend company to a friend	Between Groups	692.399	2	346.199	829.181	.000
Within Groups	245.919	589	.418
Total	938.318	591
Loyalty for Nike	Between Groups	6.460	2	3.230	1.331	.265
Within Groups	1429.513	589	2.427
Total	1435.973	591

Table 2  

Attitude 1: Knowledge of the company 

Alpha value = 0.05

The p-value calculated (0.00) is less compared to the alpha value (0.05). This means that null hypothesis is rejected thus at least one mean is different.

Attitude 2: Satisfaction with the company

Versus 

Alpha value = 0.05

The p-value calculated (0.00) is less compared to the alpha value (0.05). This means that null hypothesis is rejected thus at least one mean is different (Howell, 2007).

Customer Attitude Analysis

Alpha value = 0.05

The p-value calculated (0.00) is less compared to the alpha value (0.05). This means that null hypothesis is rejected thus at least one mean is different.

Attitude 4: Purchase intention for Nike 

Versus 

Alpha value = 0.05

The p-value calculated (0.00) is less compared to the alpha value (0.05). This means that null hypothesis is rejected thus at least one mean is different (Hinkelmann & Kempthorne, 2010).

Attitude 5: Would recommend company to a friend? 

Versus 

Alpha value = 0.05

The p-value calculated (0.00) is less compared to the alpha value (0.05). This means that null hypothesis is rejected thus at least one mean is different.

Attitude 6: Loyalty for Nike 

Versus 

Alpha value = 0.05

The p-value calculated (0.27) is less compared to the alpha value (0.05). This means that null hypothesis is accepted thus the means are equal across the three groups (Gelman, 2005).  . 

Multiple Comparisons
Bonferroni
Dependent Variable	(I) Webiste User Group	(J) Webiste User Group	Mean Difference (I-J)	Std. Error	Sig.	95% Confidence Interval
Lower Bound	Upper Bound
Knowledge of the company	Light Users	Medium Users	-2.056*	.154	.000	-2.43	-1.69
Heavy Users	-3.605*	.147	.000	-3.96	-3.25
Medium Users	Light Users	2.056*	.154	.000	1.69	2.43
Heavy Users	-1.549*	.118	.000	-1.83	-1.27
Heavy Users	Light Users	3.605*	.147	.000	3.25	3.96
Medium Users	1.549*	.118	.000	1.27	1.83
Satisfaction with the company	Light Users	Medium Users	-2.854*	.114	.000	-3.13	-2.58
Heavy Users	-3.532*	.108	.000	-3.79	-3.27
Medium Users	Light Users	2.854*	.114	.000	2.58	3.13
Heavy Users	-.678*	.087	.000	-.89	-.47
Heavy Users	Light Users	3.532*	.108	.000	3.27	3.79
Medium Users	.678*	.087	.000	.47	.89
Preference for Nike	Light Users	Medium Users	-.362	.193	.183	-.83	.10
Heavy Users	-1.989*	.184	.000	-2.43	-1.55
Medium Users	Light Users	.362	.193	.183	-.10	.83
Heavy Users	-1.627*	.147	.000	-1.98	-1.27
Heavy Users	Light Users	1.989*	.184	.000	1.55	2.43
Medium Users	1.627*	.147	.000	1.27	1.98
Purchase Intention for Nike	Light Users	Medium Users	.121	.193	1.000	-.34	.58
Heavy Users	-.854*	.184	.000	-1.30	-.41
Medium Users	Light Users	-.121	.193	1.000	-.58	.34
Heavy Users	-.975*	.145	.000	-1.32	-.63
Heavy Users	Light Users	.854*	.184	.000	.41	1.30
Medium Users	.975*	.145	.000	.63	1.32
Would recommend company to a friend	Light Users	Medium Users	-1.455*	.078	.000	-1.64	-1.27
Heavy Users	-2.887*	.074	.000	-3.06	-2.71
Medium Users	Light Users	1.455*	.078	.000	1.27	1.64
Heavy Users	-1.432*	.059	.000	-1.57	-1.29
Heavy Users	Light Users	2.887*	.074	.000	2.71	3.06
Medium Users	1.432*	.059	.000	1.29	1.57
Loyalty for Nike	Light Users	Medium Users	-.214	.188	.763	-.66	.24
Heavy Users	.008	.179	1.000	-.42	.44
Medium Users	Light Users	.214	.188	.763	-.24	.66
Heavy Users	.222	.143	.364	-.12	.57
Heavy Users	Light Users	-.008	.179	1.000	-.44	.42
Medium Users	-.222	.143	.364	-.57	.12
*. The mean difference is significant at the 0.05 level.

Table 3

The research also sought to determine to use multiple comparisons to establish whether there was significant difference in mean attitude levels. 

Hypothesis

H₀: Mean 1 = Mean 2 = Mean 3

Versus

H₁: At least one mean is different

The p-value calculated (0.00) is less compared to the alpha value (0.05). This means that null hypothesis is rejected thus at least one mean is different.

However for Nike products, the hypothesis and conclusion was as below; 

Versus 

Alpha value = 0.05

The p-value calculated (0.27) is less compared to the alpha value (0.05). This means that null hypothesis is accepted thus the means are equal across the three groups

Group Statistics
	Gender	N	Mean	Std. Deviation	Std. Error Mean
Knowledge of the company	Female	388	5.02	1.961	.100
Male	204	5.78	1.463	.102
Satisfaction with the company	Female	388	5.18	1.595	.081
Male	204	5.29	1.567	.110
Preference for Nike	Female	388	3.19	1.876	.095
Male	204	4.22	1.516	.106
Purchase Intention for Nike	Female	388	4.67	1.619	.082
Male	200	4.26	1.675	.118
Loyalty for Nike	Female	388	3.44	1.588	.081
Male	204	5.04	.768	.054
Would recommend company to a friend	Female	388	5.40	1.255	.064
Male	204	5.25	1.267	.089

Table 4

Above in table 4 is a summary statistics for levels of attitude towards the six items. The table compares the levels between the male and females. The males had higher (5.78) mean attitude levels compared to their counterparts who had 5.02 when it came to knowledge of the company. When it came to satisfaction with the company, the males had higher (5.29) mean attitude levels compared to their counterparts who had 5.18. On Nike preference, the males had higher (4.22) mean attitude levels compared to their counterparts who had 3.19. The females had higher (4.67) mean attitude levels compared to their counterparts who had 4.26 when it came to purchase intention for Nike. When it came to loyalty for Nike, the males had higher (5.04) mean attitude levels compared to their counterparts who had 3.44. 

Independent Samples Test
	Levene’s Test for Equality of Variances	t-test for Equality of Means
F	Sig.	t	df	Sig. (2-tailed)	Mean Difference	Std. Error Difference	95% Confidence Interval of the Difference
Lower	Upper
Knowledge of the company	Equal variances assumed	56.606	.000	-4.892	590	.000	-.764	.156	-1.070	-.457
Equal variances not assumed			-5.347	522.857	.000	-.764	.143	-1.044	-.483
Satisfaction with the company	Equal variances assumed	.024	.877	-.867	590	.386	-.119	.137	-.388	.150
Equal variances not assumed			-.872	419.112	.384	-.119	.136	-.387	.149
Preference for Nike	Equal variances assumed	14.901	.000	-6.765	590	.000	-1.030	.152	-1.329	-.731
Equal variances not assumed			-7.223	493.730	.000	-1.030	.143	-1.310	-.750
Purchase Intention for Nike	Equal variances assumed	2.724	.099	2.876	586	.004	.410	.143	.130	.690
Equal variances not assumed			2.845	390.123	.005	.410	.144	.127	.694
Loyalty for Nike	Equal variances assumed	246.135	.000	-13.543	590	.000	-1.596	.118	-1.827	-1.364
Equal variances not assumed			-16.471	586.589	.000	-1.596	.097	-1.786	-1.406
Would recommend company to a friend	Equal variances assumed	.157	.692	1.442	590	.150	.157	.109	-.057	.371
Equal variances not assumed			1.437	409.319	.151	.157	.109	-.058	.372

Table 5

Test for equality of variance in attitude (knowledge of the company) between male and female 

Hypothesis

H₀: Var₁ = Var₂

Versus

H₁: Var₁ ≠ Var₂

The p-value calculated (0.00) is less compared to the alpha value (0.05). This means that null hypothesis is rejected thus inequality of variance.

• Test for equality of variance in attitude (satisfaction with the company) between male and female

Hypothesis

H₀: Var₁ = Var₂

Versus

H₁: Var₁ ≠ Var₂

The p-value calculated (0.39) is less compared to the alpha value (0.05). This means that null hypothesis is rejected thus equality of variance in attitude levels between the males and females. 

• Test for equality of variance in attitude (preference for Nike) between male and female 

Hypothesis

H₀: Var₁ = Var₂

Versus

H₁: Var₁ ≠ Var₂

The p-value calculated (0.00) is less compared to the alpha value (0.05). This means that null hypothesis is rejected thus inequality of variance.

• Test for equality of variance in attitude (purchase intention for Nike) between male and female  

Hypothesis

H₀: Var₁ = Var₂

Versus

H₁: Var₁ ≠ Var₂

The p-value calculated (0.00) is less compared to the alpha value (0.05). This means that null hypothesis is rejected thus inequality of variance in attitude levels between the males and females.

• Test for equality of variance in attitude (loyalty for Nike) between male and female

Hypothesis

H₀: Var₁ = Var₂

Versus

H₁: Var₁ ≠ Var₂

The p-value calculated (0.00) is less compared to the alpha value (0.05). This means that null hypothesis is rejected thus inequality of variance in attitude levels between the males and females.

• Test for equality of variance in customer attitude (would recommend company to a friend) between male and female 

Hypothesis

H₀: Var₁ = Var₂

Versus

H₁: Var₁ ≠ Var₂

The p-value calculated (0.15) is great compared to the alpha value (0.05). This means that null hypothesis is accepted thus equality of variance in attitude levels between the males and females. 

One sample t-test for the mean satisfaction level (3.5)

Hypothesis

H₀: Mean = 3.5

Versus

H₁: Mean ≠ 3.5

Test results table is as shown below  

One-Sample Statistics
	N	Mean	Std. Deviation	Std. Error Mean
Satisfaction with the company	592	5.22	1.585	.065

Table 6

One-Sample Test
	Test Value = 3.5
t	df	Sig. (2-tailed)	Mean Difference	95% Confidence Interval of the Difference
Lower	Upper
Satisfaction with the company	26.349	591	.000	1.716	1.59	1.84

Table 7

From the results table above, it can be observed that the observed mean is 26.34 t-deviations from the hypothesized mean (3.5). The observed mean is 1.72 greater than 3.5. The 95% confidence interval for the difference in mean is 1.59 and 1.84. Since the p-value calculated (0.00) is less compared to the alpha value (0.05). This means that null hypothesis is not accepted thus the mean satisfaction level is not 3.5.

It was found that customised products are doing well in terms of profit at Retail Surge Company. The mean total profit foe customized products was 25 dollars. The men’s shoes followed with a mean total profit of 16 dollars while the third product was girls’ shoes that made a mean total profit of 7 dollars. This prompts the research to recommend to the management of this conduct an awareness program or marketing and advertising for the less selling items so as to boost their sales and increase revenue for Retail Surge. It was also observed from the results in the bar graph previously that customised products are doing well in terms of mean cost of goods. The mean cost of goods for customized products was about 10 dollars. The girls’ shoes followed with a mean cost of goods of 8 dollars while the third product was women’s shoes that made a mean cost of goods of 5 dollars. The company is therefore advised to balance the less costly goods and the highly costly goods so as to widen the profit margin of the Retail Surge company.  

Reference

Derrick, B., Toher, D., & White, P. (2017). How to compare the mean of two samples that include paired observations and independent observations. Quantitative methods for Psychology, 13(2), 120 – 126.

Gelman, A. (2005). Analysis of variance? Why it is more important than ever. The anals of Statistics, 33, 1 – 53.

Hinkelmann, K., & Kempthorne, O. (2010). Design and analysis of experiments (5 ed., Vol. 8).

Howell, D. C. (2007). Statistical methods for Psychology (3 ed., Vol. 5).

Leigh, E. S. (2008). Consumer rites. Selling of American Holidays, 6(3), 106 – 191.