a. The data set consist of 44 female and 53 males.
b. There are 21 people under the age range of 13-35, which account for 21.6% of the total 97 in the sample.
c. The mean level of the hour spent in the office during week 2 is 47.47 hours, with the standard deviation of 17.06. The distribution is slightly negative skew (-0.237), although in an acceptable range suggest that slightly more samples spend more hours in office than the average. The distribution’s value for kurtosis (-1.005) exceeds the acceptance level (value between 1.0 and -1.0.), implying it has a ‘Platykurtic’ characteristic. This means that value is less cluster around the mean, with heaver tails.
In order to test for significant dependency between two categorical data, we conduct Chi-Square Test. …show more content…
The null-hypothesis is that there is no significant difference in sales performance between males and females. Before we get to the T-value, Levene’s test for equality of variance must conducted. In SPSS, if Levene test produces a significant figure greater than 0.05, then homogeneity of variance is hold, and the top-row corresponding T value is selected. In both weeks homogeneity of variance is hold. In week 1, the T-value (df=95) is 1.848. This generates a p-value of 0.068, which exceeds Fisher’s threshold and therefore null-hypothesis is hold. There is no significant difference in sales performance when compare males (mean=4232.34) to females (mean=5227.86). In week 2, the T-value (df=95) is 2.828 and the p-value derived is 0.006. Here, the possibility that the difference is due to mere random chance is less than 1% therefore the difference in sales performance between males (mean=4175.92) and females (mean=5675.97) in week 2 is strongly significant. A conclusive argument can be drawn that in week 2, females have a better sales performance than male