Improvement Plans

[Siddhant-96]

1. In figure 5.1, rather than comparing distribution of tuition fees of residents and non-residents in a single distribution graph, making 3 separate side by side to compare tuition fees of public, private and for profit schools individually.

2. In figure 5.2 using net_cost incurred instead of scholarship amount to display the burden a student has taken across various income groups.

New Question : Does paying high tuition guarantee a higher salary? Which college has more potential in career growth on the basis of initial salary offered?

[aarathybabu97]

1. In figure 5.3, Using the improvement percentage to show a better comparison of student development across different universities.
2. In figure 5.5, Using a different plot in figure 5.5 like a histogram instead of the present figure as it is difficult to draw inferences out of it!

New Question : Are all States and universities accommodative of students of different ethnic and economic backgrounds?

[dicook]

@aarathybabu97 and @Siddhant-96,

You've inherited a really good report. The pressure is on for you to make this a great report.

How will the changes to figures you have listed above contribute to improving? You have mentioned comparison - yes, spot on - but how will the comparison be made, and what do you mean by better comparison.

These two comments on the issue don't give a sense that you are working together to make a cohesive completion to the initial report. It looks like you have both carved out a little segment that you will work on separately. How do the two plans merge into one improved document?

[Siddhant-96]

For better comparison between residents and non-residents in figure5.1 we are planning to use a density plot, histogram(probability density instead of frequency) and plot it together.

[aarathybabu97]

Also , As we explored the dataset, we also found out that there were 2 different types of courses offered i.e., a 2 year and a 4 year course so we used 2 distinct plots accordingly as part of improving fig 5.1

[aarathybabu97]

In the original report, one of the secondary questions intends to find out how much burden a student has taken. But figure 5.2 uses and speaks about the scholarship earned. But upon discussing the same we thought that "net_cost" is a better indicator for the same and we have instead used that. Also, we plan to show the scholarships students of various income groups earn in a separate graph altogether.

[Siddhant-96]

Also, we noticed that the courses offered were on-campus and off-campus. Although, there is not a lot of difference between the two, we thought it should be shown separately as the total cost for both the courses are not same.

[aarathybabu97]

In the original report for figure 5.3 sum of the improvement was used which we felt was not appropriate to make comparisons so for better comparison we plan to group by the state and find out the average and reorder the states according to the improvements.

[Siddhant-96]

For table 5.3 in the original report, should show universities with improvement above 75% but upon inspecting it closer, we found out that it instead shows improvement above 0.75% and we corrected it accordingly.

[aarathybabu97]

For figure 5.4, the variable "making world a better place" has 33 missing values which were not taken into account in the original report. Also instead of summation, we plan to group by states and take the average of the variable "make the world a better place" to draw better inferences.

[Siddhant-96]

Initially, we planned to do away with the circular bar chart. But upon discussing, we thought with a little extra effort we could present the earlier circular bar chart in a better manner. To understand the circular bar chart better, empty bars will be added to distinguish between different states also we will fill the chart with colors according to the mean and arrange it appropriately.

[aarathybabu97]

Since there were a lot of outliers present in the boxplot representing diversity in fig 5.5 , we looked up for plots that represent outliers in a better manner. So we found the letter value plot and will try to incorporate it instead of the boxplot.

[Siddhant-96]

When we were paired up together, we discussed the entire original report of team magpie together and thought of certain changes which could be inculcated to improve the report. Then we divided the work equally so it doesn't burden the other person. We assure you after improving every tiny change we make in the original report, we communicate and check if we both agree upon it and then and only then we incorporate in the new report. Sincere apologies if it came off as if we are not working in tandem.