Best tool for exploring and analyzing data
- Although Orange seems daily lightweight and competent at managing data, I think that the best tool for organizing and analyzing data will have to be Tableau as it allows you to export results as a PDF (Assuming that the PDF is vector based and can be edited using Adobe Illustrator)
Best tool for visualizing data
- I have a strong preference for using Adobe Illustrator to visualize information, I am familiar with the software, enjoy using the application, and it offers limitless freedom in terms of presentation.
|Orange||LightweightHighly modular||Limited or difficult to learn visual controlsComplexUnintuitive UI|
|Tableau||Easy to useAbility to export to PDF||Costly|
|Illustrator||Complete freedom of dataComplete visual controlI am familiar with the software||Time consuming|
Find a correlation in your data set
(Made using Tableau)
Interpret the findings
There is a clear relationship here. For commercial / institutional consumers and residential consumers gas consumption steadily declined. Industrial consumption of gas dropped in April, but remained relatively consistent after that point.
Explore normal distributions in your data set
(Generated using Orange)
I struggled to get normal distributions working in orange or tableau, I was turned off by the scripting and advanced math that popped up when I looked up how to do it. I suspect that the result is inaccurate, as I had expected a positive skewed mesokurtic curve. If it had just been the green and blue distributions it would have been a Leptokurtic curve. If it had just been industrial consumption it would have been closer to a platykurtic curve.
I don’t entirely understand the math behind normal curves, so my expectations may be wrong, although I see no reason why industrial consumption would occur so late. I wonder if the values are being accumulated between time periods and rather than occurring 5 times (the amount of time points) they occur less often.
Variables that are not normally distributed:
- I would speculate that none of the variables are being displayed normally due to user error or ignorance.
Why they are not normally distributed
- One possible explanation is that the variables have been established wrong or contain values that are not properly recognized
- Another possible reason is a failure of understanding mathematics and or a failure to establish proper commands.
- A third possible reason is that there was there is insufficient data points to properly establish a curve
This graph communicates the chronology of life on earth.
- It is easy to parse visually and get a general understanding. The data elements are visually distinct, communicate using symbols, and are unified within the piece to assume grouped elements.
Issues of communication
- It is not visually immediately apparent that the numbers are in millions of years or that it is measured relative to current time. This could be solved by adding a ‘-’ (minus) in front of the reference number and a ‘M’ after to signify ‘million’.
- There is also some confusion regarding the radial divisions, particularly the inclusion of ‘long ago’ that seems out of place and the black bards which could simply be replaced by the colour of the category.
Words related my topic:
Words relating to personality of topic:
- Cold vs compassionate
Made using Canva
(Death by causes in 2019 of Canadian residents)
This data set may have too many categories to be visually successful. Although it appears most successful in its third form.
Here is a visual representation of how the data could be displayed
8.1 Relational information
Here is what might appear or be included while looking at youth health relational data
Improving the class-level relational visual
A possible visualization of instance-level relations
An example of displaying the data would look like this
9.1 Trend Analysis
Total population of canada represents maximum value, population of P&T is a value represented as a colour (0 = 0, 255 = maximum value);
Population of P&T displayed as size, (P/T size = population / total canada population)
For this I improved the sense of scale in my sketch, using the full visual range to display the information.