Executive Summary
It is important to keep the customer requirements and preferences in mind in order to provide the best possible service and Sa Pisitta Supermarket is taking the opportunity in order to conduct a detailed analysis in order to come up with the solutions which can help them to target and attract their customers in a better manner. The variables defined are mainly the age, gender, distance and average monthly spent of the customers. The analysis targets an understanding of the association between different variables and how they impact each other. The overall data variables are converted into numeric responses in order to achieve a normal distribution and this report also comments about the scatter of the customers across variables e.g. on the basis of their age, location and occupation status.
Sa Pisitta Supermarket is a large retail centre located in Pompu, on the Sardinian west coastline. The supermarket’s marketing analysts must use appropriate statistical checks in order to understand the actual situation of the company in the market. The table which is mentioned below, provides a clear understanding of the variables associated with each aspect of the survey is conducted for the customers. There are variables which are used as a part of this analysis, and therefore it is important to define their category before moving to proceed with the actual analysis.
[1]Variable description table
[2]Describing Categorical Data
The marketing analyst wants to understand the target-customer of the supermarket. An appropriate chart to display the variation in customers’ status and occupation is shown below. The association between status and occupation is clearly visible from the comparison of the line chart, which is created, as it provides the understanding that people with statuses single and married are the ones who are employed and are earning. With this understanding the analyst can also conclude that targeting the population as mentioned above would be beneficial in order to increase the sales. A close study of the data clearly reveals that, single and employed are modal attributes. However, at the same time widow and unemployed are the least common attributes.
If we aad the car ownership factor to the previous analysis, it becomes very easy to understand that most of the car owners are also in the same criteria of people, who have a status as single or married. The availability of this information is very helpful in providing an idea how the class of customers who maybe spending more in the supermarket, and by identifying them. It will be your target them to generate further sales. In the next section, the analysis would also involve the actual spend, which can then be compared to other variables to obtain the required results.
The analysis clearly indicates unknown uniform distribution of the overall customer database, and therefore if we compare the status and an ownership of the car, it is simply supporting the fact that single or married customers are the ones who have maximum number of cars and are also employed. This pattern indicates a specific class of customers which is either single or married, has a car ownership and are employed, which gives a clear indication of their strong financial condition and also the capability to sustain as a long-term customer is targeted efficiently.
[3]Describing Quantitative Data
The marketing analyst is interested in how customers’ spending changes with regard to different characteristics
The analysis which is done below is on the basis of a single chart containing box-plots of average monthly expenditures according to gender, status and occupation. This is a very interesting and important analysis for the market analysts as it will provide the actual date of how the average spends is related to variables like gender, status and occupation. If we look at gender at first, it is clearly visible that females have better average monthly spending, which can be clearly understood by the fact that they are the ones who run the household and therefore they are more likely to visit the supermarket in order to make a purchase.
If we discuss about the status, the analysis explains the fact that the customers fall into the criteria of being married or divorced, have a better average monthly spend which can be understood by the fact that they have more number of family members and may require larger amount of household goods. Moving on to occupation and the data clearly presents the fact that the population which is employed has the best average monthly spend, thus providing us with a common input that the customers were employed are more likely to spend more at the supermarket.
The figure above shows the single chart containing box-plots of average monthly expenditures according to level of appreciation.
Std. Error
Avg Monthly Spending
Very low
95% Confidence Interval for Mean
Lower Bound
Upper Bound
5% Trimmed Mean
Std. Deviation
Interquartile Range
95% Confidence Interval for Mean
Lower Bound
Upper Bound
5% Trimmed Mean
Std. Deviation
Interquartile Range
95% Confidence Interval for Mean
Lower Bound
Upper Bound
5% Trimmed Mean
Std. Deviation
Interquartile Range
95% Confidence Interval for Mean
Lower Bound
Upper Bound
5% Trimmed Mean
Std. Deviation
Interquartile Range
Very high
95% Confidence Interval for Mean
Lower Bound
Upper Bound
5% Trimmed Mean
Std. Deviation
Interquartile Range
The above data provides a clear indication of a relationship between the average mean for each of the results related to, the appreciation on the basis of the average spend. Although, on the basis of the box plots, it is clear that there is no as such direct relationship of the average spend with the level of appreciation provided by the customers. The average spend about 150 seems to fetch the maximum appreciation. However, the same about 175 has a low feedback which again was higher remains in different than the amount is reduced. Hence, it is clear that apart from the indifference factor, most of the means of the average spend around appreciation factors remain close to the 145 mark however there is no specific co-relation between the average spend and the level of appreciation given by the customers.
[4]Distributions
The marketing analyst wishes to know if the distributions related to customers’ average monthly expenditure and travelled distance are normally distributed. Hence, it is important to create an analysis of the data to find out whether the above-mentioned two variables are normal in terms of the distribution and therefore test is conducted on, where the results are shared as below. The result is received, clearly signifies the normality of the data and provides us with the fact that the average monthly expenditure and the travel distance are normally distributed.
The normal distribution of the data is being inferred however, there is a close association factor between the distance and the average monthly spend. If we look at the box plots above, the higher becomes the difference. The lower the amount of average spend however, this trend is not homogeneous. Considering the fact that in the above cases, there are events where the average spend has fluctuated down despite the rise in the distance.
[5] Hypothesis test
The marketing analyst is now interested to know if there is any significant average age difference between males and females and if there is which is the older group.
Std. Error
95% Confidence Interval for Mean
Lower Bound
Upper Bound
5% Trimmed Mean
Std. Deviation
Interquartile Range
95% Confidence Interval for Mean
Lower Bound
Upper Bound
5% Trimmed Mean
Std. Deviation
Interquartile Range
Null Hypothesis - There is no significant difference between the average age of the males and female customers of the supermarket.
Alternate Hypothesis - There is significant difference between the average age of the males and female customers of the supermarket.
If we look at the above data, it is clear that the average age of men and women entering the supermarket is different by almost 7 years, and therefore the alternate hypothesis holds true and the null hypothesis is nullified.
The above-mentioned box plots show a correlation between the gender of the customers and their average age, and it is clearly visible that the average age of the male customers is higher than that of the females. Now the analysts would be able to use this data in order to target the customers according to their age, and now that the average age of both males and females were mostly shop around is available, the target customers through this information can be easily chosen.
[6]Data manipulation and analysis
Std. Meviation
Std. Error
95% Confidence Interval for Mean
Lower Bound
Upper Bound
Very low
Very high
If we look at the above analysis to find the gender based segregation of the appreciation on the basis of the values which are changed and associated with the new set of values assigned to the variable. It is important to notice that although there is no significant change in values which is done by reducing the number of variables in the dataset, there is a steep impact f the same to the overall mean satisfaction level which drops down, which accounts to the fact that the mediocre values of 2 & 3 which were initially low values suddenly have further gone down and are not included in the satisfaction levels t all.
The change in scale of the evaluation of teh appreciation is the primary reason behind the impact of the analysis results that have been obtained as a part of the comparison in values after the changes have been made.
[7]Correlation and regression
The marketing analyst has become so enthralled with the benefits of using statistical analysis as a performance indicator toolkit, then decides that further analyses on customers’ behaviour. If we look at the age factor from the distance associated with it, it is clearly visible that the data is quite scattered one and there is no significant continuity between the results that are received as a part of the analysis.
If we look at the distance as a factor, again, it does not have an active co-relation with the age and the data is pretty much scattered and is oddly randomized. The distance and the monthly visits have a close association and although with perfect homogeneity, as the distance increases the monthly visits in number also decreases.
Hence, if we look at this from an analysts point of view, it is clear that the association of the customers on the basis of their age to the distance provides the fact that as the age increases the number of monthly visits also decrease accordingly. The next result that is inferred out of the data is that, higher is the distance and lower are the monthly visits, hence this will help to increase measures to encourage the customers living at far distances to visit as regularly as the customers who are loving close to the super market.
a. Perform linear correlation and simple regression between travelled distance and monthly visits. Do results provide any significant information?
The results clearly signify an inversely proportional relationship and which means that the higher is the distance been the supermarket and the customer’s residence, lower will be the number of total monthly visits.
Works Cited
Ahamed, S.S.R. (2010) 'Project Planning: An analysis', International Journal of Engineering Science and Technology Vol.2(1), pp. pp.18-29.
Antony, J., Antony, F.J. and Ghosh, S. (2004) 'Evaluating service quality in UK hotel chain: a case analysis of the impact of service changes on study', International Journal of Contemporary Hospitality Management, vol. 16, no. 6, pp. 380-384.
Appelbaum, S.H. (2005) 'The critical success factors in the client-consulting relationship', Journal of Management Development Vol.24 No.1, pp. 68-93.
Banwet, D.K. and Datta, K. (2002) 'Effect of service quality on post visit intentions over time.', Journal of Total Quality Management, vol. 13, no. 4, pp. 537-546.
Bartlett, J.E., Kotrlik, J.W. and Higgins, C.C. (2001) 'Organizational Research: Determining Appropriate Sample Size in Survey Research', Information Technology, Learning, and Performance Journal, Vol. 19, No. 1, pp. 43-50.
Bouma, G. and Ling, R. (2004) The research process, New York: Oxford University Press.
Christian, S., Jens-Peter, M., Ranata, M., Olin, R., Clemens, H., Alexis, H. and Stephen, L. (2008) Design for Six Sigma and LeanToolset: Implementing Innovations Successfully., San Antonio: Prentice-Hall.
Creswell, J. (2009) Research design: Qualitative, quantitative, and mixed methods approaches, 3 edn, Los Angeles: Sage Publications.
Davis, K.D. (2004) Manage Your Writing, Indianapolis: McGraw-Hill.
Dean, A.M. (2004) 'Links between organizational and customer variables in service delivery', International Journal of Service Industry Management vol. 15, no. 4, pp. 332-350.
Gano, D.L. (2011) Seven Steps To Effective Problem-Solving And Strategies, Washington: Apollonian Publications, LLC.
NHS (n.d) Process mapping, analysis and redesign, Annesley: NHS Institute for Innovation and Improvement.
The Banting Institute (2007) Conducting Survey Research Version 4.0, Toronto: University of Toronto.