Visualising trends over time

 Visualising trends over time on Kaggle

During this data visualisation course, I have been using an application called Kaggle, from there I learnt various different ways to extract data and write it on a graph. The image above is a graph that I created by writing three lines of code and extracting data from a spreadsheet in order to create this graph. This graph shows the monthly visitors to a museum named Avila Adobe from the years of 2014 to 2019. The data here shows a trend of the march-august seasons having peak visitors while September-February every year. This shows that the solution to the problem when Avila Adobe should employ seasonal workers, it would be during spring-summer and not autumn-winter.

Pivot Table - Data Manipulation

Pivot Table

In todays' lesson, we were tasked to go through different data manipulation tasks in order to understand how to move data and compile data. For Pivot tables, this was a task to gather data on different criteria instead of having a single property type dispersed throughout the sheet. Pivot tables allowed us to group up the data on the property types then also provide the sum of the sale prices of those property types. This was an interesting way to group data together.

Misleading Graphs

Graphs are very useful in our daily lives. It is a way to present statistics as a form as evidence in order to back up a claim. However, there are also many ways a graph can be used to mislead an audience in order to try and fool someone. For example, in this graph the data represented here is on the number of COVID-19 tests per million of people. When the graph is formed this way, many people will assume Norway has had the most Covid tests out of all of these countries, however this graph fails to mention the population of each country and how small the country of Norway is compared to the country of US or Brazil. Below are some examples as to how a graph can fool you

1. Different Colours (colours are generally correlated for positive or negative effects)

2. Unique name of Axis (May cause the graphs to alter in shape and suggest other things)

3. Causation vs Correlation (Two different variables put in a graph may not mean that they are affected by each other)

4. Scaling of the axis and whether it starts at 0