MS980 : Business Analytics Classification and Clustering - Management Assignment Help

Assignment Task:

Task:

(a) Use the Guardian score to group the universities into two groups – those with a score of 70 and above and those with a score of below 70. How many and what percentage of institutions are in each group?
What proportion of Russell and non-Russell universities have scores of 70+. Display these proportions in a suitable plot.

[3 marks]
(b) Investigate the distributions of, and provide appropriate numerical summaries of the 2 variables, Expenditure.per.student.FTE and Average.Entry.Tariff. Give reasons for your choice of numerical summary statistics. Describe the association between these two variables?

[4 marks]

(c) Using suitable plots, examine the distributions of Student.staff.ratio, and Satisfied.with.course, over the two groups from (a) and also over the two Russell group categories. Describe your findings.

[5 marks]
(d) Take a subset of the universities corresponding to those in the Russell Group. How many universities are in this subset? Within this group how many have a guardian Score of 70+? Also, examine the distribution of the Average.Entry.Tarrif over the two Guardian Score Groups and comment on the patterns you see

[4 mark]
(e) Using the subset of Russell group universities, carry out single link, complete link and average link hierarchical cluster analysis using all 9 performance variables (i.e. exclude Name, Guardian Score and Country from the data set you read in) and the Euclidean distance between the universities. Explain why you scaled or did not scale the variables. For each clustering method write down the agglomerative coefficient. Which of the three linkage methods do you consider to be the best; give reasons for your choice. Plot the dendrogram and the banner from the best method and provide an interpretation of these plots. Hence, decide if there are a small number of groups which capture the structure in these data. Give reasons for your choice of the number of groups to use. How many universities are in each group?

[9 marks]
(f) For each of the groups from (e) calculate the average scores on each of the 9 performance measurements. Hence provide an interpretation of the groups you obtained from the hierarchical clustering analysis.

[6 marks]
(g) Carry out a kmeans analysis of the 9 performance measures using all the data. You need to decide on the most appropriate number of groups to use. Provide an interpretation of the clusters by (i) forming a table of cluster by Guardian Score <70>

[9 marks]
(h) Plot the Average.Entry.Tariff against Satisfied.with.Assessment and label with points to identify which kmeans group each university belongs to. What is your interpretation of this plot?

[3 marks]
(i) Calculate the average Guardian Score in each of the k means groups that you have. Also use an appropriate plot to investigate the distribution of the Guardian Score, the Average.Entry.Tariff and Satisfied.with.Assessment over the k means clusters.

[3 marks]
(j) Briefly, explain how you could use principal components analysis on these data. Which variables would you use and what might this analysis achieve. Also discuss if you could use nearest neighbours classification with these data. Which variables would you use as predictors and what would your response variable be for the groups? You need only write about 1 paragraph for this question and you are not expected to carry out principal components analysis nor nearest neighbours classification.

This MS980: Management Assignment Help has been solved by our Management Experts at My Uni Paper. Our Assignment Writing Experts are efficient to provide a fresh solution to this question. We are serving more than 10000+ Students in Australia, UK & US by helping them to score HD in their academics. Our Experts are well trained to follow all marking rubrics & referencing style.