数据分析代写|CMT218 Data Analysis and Visualisation Creation Assignment

这是一个CS数据分析代写的相关案例

Assignment

 

You are still working for the excellent and very real website ‘martins-cool-datavis-website’ as a key contributor. Your last contribution to the website was received very well, and the editorial team would like you to take on a more involved task. They would like to produce a feature on the website where a data visualisation practitioner creates a piece of data visualisation and then explains *why* they have created the visualisation that way, to show the process that goes in to developing a successful data visualisation.

 

To do this, you are first asked to carry out an analysis of a dataset and to identify:

 

  1. What key message can you communicate about this data through a data visualisation?
  2. Who are the audience that would be interested in a data visualisation on this topic?

 

You should then create a maximum of two (2) visualisations, (or a single (1) dashboard comprising a set of linked sub-visualisations) communicating the key message about the dataset you have selected to the audience you have identified.

 

Alongside this data visualisation you should write a very short (2 page, 800 word) article evaluating why the visualisation you have created is a good visualisation to communicate the key finding from your data analysis to your chosen audience. This evaluation should not focus on *what* the data says, or *what* you are trying to communicate, but rather *why* your visualisations do a good job of communicating the information that you have found out about the data. It must *not* just be a description of what the visualisations show and what they tell us about the data.

 

The data you choose to analyse should be one or more freely available dataset(s) on any topic, (with a small number of restrictions, see below) from a reliable source.

 

You should analyse this data to determine what the data tells you about its particular topic and should visualise this data in a way that allows your chosen audience to understand the data and what the data shows. You should create a maximum of two (2) visualisations of this data that efficiently and effectively convey the key message from your chosen data. It should be clear from these visualisations what the message from your data is.

 

You can use any language or tool you like to carry out both the analysis and the visualisation, with a few conditions/restrictions, as detailed below. All code and data used must be included alongside the article, so that readers can reproduce your visualisations. You should include a small appendix to the article that must include enough instructions/information to be able to run the code and reproduce the analysis and/or visualisations.

 

 

 

 

 

Tool usage

 

Although you are free to use any tool, language, library that you like, there are some exceptions/conditions to this for you to be aware of. The editorial team for the website want to ensure the piece you are creating is accessible to the widest audience possible.

 

In order to view this piece, the audience need to be able to see it! You absolutely *must* submit everything that is needed to create and view your visualisation. This should include all code used to clean and filter data, any data required, and so on.

 

You must include enough instructions/information to be able to run the code and reproduce the analysis/visualisations.

 

Tableau

If you use Tableau to create your visualisations, you *must* ensure you are either creating a single dashboard that combines multiple sub-visualisations together with some form of linked functionality between the sub-visualisations, or alternatively an effective story presentation. Simply creating individual non-linked visualisations will not suffice.

If you use Tableau to create your visualisations you must submit a packaged tableau workbook that includes all needed resources within the packaged .twbx file[1].

 

PowerBI

If you use PowerBI to create your visualisation, please make sure it is possible for someone to view your submitted visualisation. You must submit the .pbix file for your dashboard, and should also submit a link to the online version of the dashboard. As with Tableau, you *must* ensure you are creating a single dashboard that combines multiple sub-visualisations together with some form of linked functionality between the sub-visualisations

 

Python/JavaScript/R…

Please submit a list of all libraries required to run your code/visualisations. This might be a pipfile or a requirements.txt for Python, or a package.json for javascript, and so on.

 

Java

No. No Java. It’s the only programming language that’s banned. We just can’t deal with the classpath issues.

 

Dataset Selection

 

You are free to choose data on any topic you like, with the following exceptions. You cannot use data connected to the following topics:

 

  1. COVID-19. We’ve seen too many dashboards of COVID-19 data that just replicate the work of either John Hopkins or the FT, and we’re tired of seeing bar chart races of COVID deaths, which are incredibly distasteful. Let’s not make entertainment out of a pandemic.
  2. World Happiness Index. Unless you are absolutely sure that you’ve found something REALLY INTERESTING that correlates with the world happiness index, we don’t want to see another scatterplot comparing GDP with happiness. It’s been done too many times.
  3. Stock Market data. It’s too dull. Treemaps of the FTSE100/Nasdaq/whatever index you like are going to be generally next to useless, candle charts are only useful if you’re a stock trader, and we don’t enjoy seeing the billions of dollars hoarded by corporations.
  4. Anything NFT/Crypto related. It’s a garbage pyramid scheme that is destroying the planet and will likely end up hurting a bunch of people who didn’t know any better.