Know your data: how to choose a compelling plot or chart

In order to have a successful data visualization, it is important to know the hidden relationships in your dataset. There are several common relationships in the data, that are described as:

  1. Normal comparison: this relationship includes comparing any numerical variables to each other to show a trend based on values of a quantitative variable or parameter.

  2. Deviation: this relationship measures the deviation of data points with respect to a certain value such as mean or median of data points to data points are distributed or related to each other.

  3. Time-series: this relationship is important for dynamics dataset that evolves in time.

  4. Distribution: shows how data points are distributed such as showing the histograms.

  5. Correlation: this relationship shows how data points are correlated to each other for a certain variable in two or more different categories of groups.

  6. Part-to-whole relationships: this relationship tries to compare a subset of data points to the whole.

  7. Ranking: this relationship ranks the variables or parameters based on a numerical variable.

There are several chart and plot types that would be covered in this workshop. The most common plots or charts are categorized as:

  1. Bar chart

  2. Pie chart

  3. Line chart

  4. Area chart

  5. Scatter plot

  6. Bubble chart

  7. Heat map

Now, the mentioned relationships and the plot and chart types are connected together based on this roadmap for choosing the most informative visualization platform based on knowing your data type:

In the next section, the technical details of working with different plots and charts as well as choosing the best visual components to create a compelling data visualization will be discussed in detail.