Tuesday, 4 September 2012

Business Analytics :- Session 3 & 4 (Part 1) | Group F


Business Analytics :- Session 3 | Group F
 
We were provided with the data of Retail Stores in our third session of Business Analytics.
The data that was to be analysed gave the first impression of a pool of information relating to customer satisfaction. We were required to analyse the data provided and figure out possible variables (factors) that would have an influence on the customer satisfaction.
The satisfaction scores available to us were Price Satisfaction, Organization Satisfaction, Variety Satisfaction, Item Quality Satisfaction and Overall Satisfaction.
Some of the variables that we used in our analysis were Gender, Shopping Frequency, Method of Payment, etc. We had to find a relation or a correlation if it existed between the variable and the customer satisfaction. This would enable us understand and identify the key factors influencing customer satisfaction.

Crosstab enables us to find the relationship between the two selected variables. We can select any two variables and measure the extent of influence one has over the other. For example, Crosstab allows us to find if the gender has any link to the customer satisfaction or if the method of payment has any influence on the customer satisfaction score. Store managers would benefit if the key variables to customer satisfaction are identified and the extent to which they influence the same is measured. They would be able to use this analysis to their advantage and improve upon the customer satisfaction levels. A successful analysis of the same would facilitate them to channel their energies and help them focus on the specific issue that if resolved or enhanced would lead to improved customer satisfaction.

Example :- Our group did an analysis to find out if the Shopping Frequency was related to Service Satisfaction.


Chi-Square Test is also selected from the Statistics Section in the Crosstab Function.
The Chi-Square is a test that determines if the variables (one in row and the other in column ) are related to each other or not. The null hypothesis is created that stated there is no relation between the variables compared and is accepted if the test result is more than 0.05. This indicates that the variables used are not in relation to each other. A test result of less than 0.05 means that there is a relation between the variables and hence the null hypothesis is rejected.
 

The use of Data Selection was taught that helped in filtering out the irrelevant data which was found out after the first analysis. This resulted in an in-depth analysis and helped us streamline our findings.
Our group used the above tests with many different combinations of variables and their relationships were tried like price satisfaction and gender etc. The use of these with different combinations helped us get familiarised with the software, but more importantly it helped us improve our analytical skills.

By
Piyush Upmanyu
11020841093
Group F

No comments:

Post a Comment