In: Statistics and Probability
A research team collected some data to address several research questions. The coding scheme is included below (Table 2).
Table 2. Coding Scheme for Question 3.
Variable Name |
Label |
Values |
iuser |
Online store user or not |
1 = user; 2 = nonuser |
age |
Age group of participants |
1 = below 20; 2=20-39; 3=40 and above |
income |
Monthly personal income |
1= below 10,000; 2=10,000 – 19,999; 3=20,000- 29,999; 4=30,000 and above |
SAT |
Satisfaction towards the store |
1-7(1 = not satisfied; 7=satisfied) |
iattitude |
Attitude towards online store |
1-7(1 = dislike very much; 7=like very much) |
oattitude |
Attitude towards offline store |
1-7(1 = dislike very much; 7=like very much) |
a) “Does the average level of satisfaction towards the store exceed 5.0?”: In order to analyze this problem one would have to test whether the average level of satisfaction is greater than 5 or not. In such a case one would perform one sample t test for mean using the hypothesis that:
H0: The average level of satisfaction towards the store is 5.
vs H1: The average level of satisfaction towards the store is greater than 5.
b) “Do customers differ in their attitude towards online stores and attitude towards offline stores?”: In order to analyze this problem one would have to test whether the average level of attitude towards online store is same as the average attitude of customers towards offline store. In such a case one uses the t test for the difference of means for the paired data if both the questions were asked from the same customer. The hypothesis is as follows:
H0: Customers do not differ in their attitude towards online stores and attitude towards offline stores i.e.
H1: Customers differ in their attitude towards online stores and attitude towards offline stores i.e.
c) “Are online store users younger?”: In order to analyze this problem one would have to first define the younger category i.e. are indicators 1 and 2 belong to the young category or just the indicator 1. Depending on the definition the entire data will need to be classified in a 2x2 table wherein the four categories are user, non-user, younger and older. For each combination cells will be filled with appropriate numbers. Then to analyze the interdependence between usage and age one would perform a chi square test for the goodness of fit. The hypothesis that will be used are:
H0: There is no association between being an online user and the age of the customer.
H1: There is a significant association between being an online user and the age of the customer.
d) “Do the levels of satisfaction towards the store differ across different age groups?”: In order to analyze this problem one would have to test whether the means of the levels of satisfaction towards the store are different across the age groups. This will analyze the mean scores held by each age group. Since we have 3 categories of age we will use one-way ANOVA to test the hypothesis that:
H0: The levels of satisfaction towards the store are same across different age groups.
H1: The levels of satisfaction towards the store differ significantly across different age groups.
For all the above tests one would use an appropriate level of significance.