Gender Age And Apparel Product Return Case Study

4.1 Gender, age and apparel product return
In the first phase, the real transaction data of the garment company were conducted. There were total 42,703 consumers included in this data. Since the sample size of women (31,897) and men (10,806) were different, Welch robust t-test should be used (Ramsey, 1980). The result in Table 2 shows that women (mean = 1.43) were more significant (p-value =0.000 < 0.05) related to apparel product returns than men (mean = 1.36) of online shopping. Thus, Hypothesis 1: “Women are more related to impulse buying than men” was supported in the first phase.
Table 2
Independent Variable Sample size Mean Stad. Deviation Sig.
Female 31897 1.4331 .95817 .000
Male 10806 1.3555 .00814
Total 42703 1.4135 .93170
Gender and apparel product return (data from the garment company)

In K-S Test (Table 3), p-value = 0.000 < 0.005, which means the null hypothesis, that the data is following a normal distribution, was rejected. Neither the data of age or return frequency were following the normal distribution, thus a non-parametric test should be used. Hypothesis 4: “H4: Age is negatively related to impulse buying” was rejected since there was no significant correlation between the age and apparel product return (Table 4, p-value = 0.621 > 0.05) in the first analysis