The human brain is wired to process visual data very rapidly. A large amount of its energy is devoted to visual processing. This is probably an evolutionary trait that has helped humans to survive in early age hostile surroundings. During those times, quickly noticing even a small movement, recognizing animal footprints, and correctly interpreting danger signs often played a crucial role in saving one’s life.
This explains the reason why we are better at perceiving data when presented in visual forms (charts, graphs, etc) rather than in raw format (tabular). We have a tendency to better identify patterns or trends in a picture or a visualization. There is substantial truth in the age-old adage “A picture is worth thousand words”
However, the correct interpretation of data visualization is an art. Charts and graphs can often mislead people, sometimes due to the wrong interpretation or many-a-times due to the wrong intentions of the creator. Many political and marketing campaigns are geared towards seducing people and misinforming them by intentionally presenting dubious information.
This article is an attempt to explain the ideas behind commonly used data visualizations and how to interpret them effectively.
Scales and Proportions
Changes that occur over a period of **_time _**can be thought of as a “trend”. e.g Changes in Stock price over a week/month/year, GDP growth of a country over years, rise and fall in temperatures over month, etc. The best chart to visualize trends is a Line Graph. Consider below dummy data for sales of a product over 6 months:
This data can be visualized in two ways as below:
The graph on the left shows that sales have increased over the months, while the graph on the right shows that sales have more or less remain the same. What is going on?
Tip 1: Always check the scales on both the axises.
On the left graph, the vertical axis starts at 3 while on the right graph, it starts at 0. Technically both graphs show same data but the interpretation can differ significantly by the way it’s scaled on the axis.
#storytelling #productivity #data-science #data-analysis #data-visualization #data analysisa