Data Visualisation Assignment

The topic chosen by me to visualise was the number of charities relating to Cork city and county. The information was gleamed from the Revenue website some time ago and placed in an Excel spread sheet by me.  It was very time consuming.  It is basic information.  The number of charities were manually drawn down by me for Cork city and county.  However, there were a number of issues.  The data was not very user-friendly on the Revenue website.  It was some time ago, however, the data was either in a PDF or website format and not an excel sheet which made it very difficult to deal with the data and carry out any kind of analysis on the data.  With perseverance the data was placed into an Excel spread sheet and a number of attempts were made to carry out an analysis of the data which was quite limited given the little data available to me.  Datahero could not successfully load the file.  Tableau software managed to make some sense of the data after playing around with it.

Firstly, to be clear in my mind, a simple bar graph analysis was performed to check which had more registered charities, Cork city or Cork county with the following graph:

one

 

This analysis was carried out using a simple Google Fusion table. The city has slightly more registered charities over the county.  However, this data is open to interpretation, some of the data is questionable over whether it lies in the city or county as it may have stated that it lay in the county when in fact it was in the city or vice versa and this may be further added to by the redrawing of constituency lines etc. in various parts including Cork city and county.  Initially a question mark was placed to indicate this, however, for the purposes of this exercise it was eliminated.  The Charity number and addresses served no purpose this time around, however, it would be interesting to label each area and see which area contained according to clusters the most registered number of charities or place it on a Google or Open street Map and carry out an analysis.  However, as clearly noted the addresses may need to be re-formatted or standardised.  This may be resolved by postcodes or Eir Codes.  The CHY number really amounts to nothing in this case, however, t may have significance to the Revenue.

 

Next a tree map was visualised of the named data:

two

 

As you will note it is not very easy on the eye or visually understood in an easy way.

However, finally came the breakthrough with a live visualisation simulation.

three

The blue circles represent the city and the orange the county and the green represents other for example a charity which mentioned Cork but was based elsewhere. The data on the Excel sheet was gleamed using the find function, that is, any mention of Cork was noted by me.  This visualisation was really appreciated after the painstaking effort that was made to represent the data in some way without adding further to it and creating any biases etc.  This is the data visualised without any written indication:

four

 

 

And if you hovered over the circles each one indicated which CHY registered charity was represented. This was fascinating stuff to me.

One from the city:

five

 

One from the county:

six

 

Now obviously the only thing really to be taken from this is that there appears to be more registered charities in the city rather than the county. However, if there were more time allocated and further data provided you could perhaps look at what types of charities under different headings in a standardised format, look at each one by geo-location according to a map and there are potentially more besides these.  The main point here is that with a minimal amount of data a very visually appealing data visualisation was created using Tableau software and one that is easily understood with a legend/headings.  The live version creates another layer to give data meaning and significance.  It is very powerful, however, it is not without dangers of its own which was emphasised in the lectures.  Data can be skewed, can contain errors- some of which have been outlined.  Neverthless, it is a very useful and worthwhile exercise.  Toiling over a dataset can make you really appreciate data visualisation.  After all there is more than one way of looking at anything.

Leave a Reply

Your email address will not be published. Required fields are marked *