Visualising and Analysing Data in Tableau

Visualizations are created based on the type of data you have. But you must know which combination of features (dimensions and measures) you can plot together to create the type of visualizations you want. Tableau automatically selects the best visualization technique based on the features you have selected, but it can always be changed manually. You can leverage elements of Tableau, such as color, size, and shape to create interactive visualizations. There are several charts that can be created in Tableau. Some are discussed below:

BAR CHART IN TABLEAU

Bar Chart: A bar chart is a very simple chart and is most useful when you want to compare data across dimensions. For example, visualizing sales across regions. It is mostly preferred when you want to create a visualization for one categorical and one numeric feature. You can add other dimension variables to the colour option in the Marks card to make a stacked bar chart and to make the visualization more interactive. A grouped or a stacked bar chart are plotted when you want to visualize a multi-categorical dimension with a common quantitative variable.

SCATTERPLOT IN TABLEAU

A scatter plot is used when you want to visualize the relationship between two quantitative variables. It gives you a very good idea of how one variable changes or reacts to different values of the other variable. Scatter plots can be made very interactive by adding other dimension variables to the colour and shape options in the Marks card.

TREEMAPS IN TABLEAU

In scatter plots, when you use a dimension to display colour, Tableau uses different colours for each category in the dimension. However, you can use a measure for colour too and Tableau will present a scale of colour ranging from light to dark depending on the measure’s values. The same concept is applied in treemaps which are used to visualise two quantitative and one categorical variable. The quantitative variables exploit the size and colour parameters of the treemap.

DUAL AXES IN TABLEAU

Dual axes charts is a very powerful way to visualise quantitative variables of different scale together. In these charts, there are 2 y-axes representing one quantitative variable and 1 x-axis representing the dimension variable. These are created when the ranges of the quantitative variables differ a lot.

HISTOGRAMS IN TABLEAU

Histograms plot quantitative data with ranges of data grouped into bins which are decided automatically by Tableau. Histograms show the distribution of a single variable across the bins. The bin width can also be changed manually.The histogram above shows the Count of Quantity on the y-axis and the Quantity bins on the x-axis with a bin width of 1. Such a histogram can be used to manage inventory of a store.

BOXPLOTS IN TABLEAU

Box plots enable us to study the distributional characteristics of a variable and show the overall patterns of a variable. They tell you about the existence of outliers and the overall spread of the data. They are also very effective when you want to compare the distribution of two quantitative variables. The line inside the box denotes the median value: 50% of the data lies both above and below that value. The boundaries of the box denote the 25th and the 75th percentile, with the whiskers denoting the lowest and the highest values in the data set. Any observations lying below or above the whiskers are outliers. However, box plots can only be created for quantitative variables.

DASHBOARD & STORIES IN TABLEAU

Dashboard and Stories: Dashboards and stories are an effective way to present your finding of the data you have visualized. To create a dashboard, click on the new dashboard icon at the bottom. The size of the dashboard can be selected from the bottom left. Sheets can be added by simply dragging and dropping.To create a filter, simply select any chart on the dashboard and select the filter icon on its upper right that says “Use as filter”. This enables some options on the menu. Go to Analysis, select filters, and the variables of that chart are available for filtering. To create a custom filter, create a new sheet and add the variable name to rows and on the labels card on the marks shelf. On the tab in the rows shelf, right click and uncheck “Show header”. To modify the font, right click on the tab, select format and choose any color. Go to the dashboard and this sheet can be dragged. Select it and use as a filter. The show title at the bottom left toggles the visibility of titles for each chart and the dashboard, depending upon the selection.

A new story can be created by clicking on the new story icon at the bottom. New slides or points can be created by clicking on “New Blank Point”. A duplicate will create a copy of the existing sheet. Captions can be used as a part of the story. To mention other interesting facts about the data, use annotate to highlight areas and enter the text. A story can also contain dashboards. If the dashboard looks too cluttered, go to the dashboard and in the size option, select “Fit to the story”. The titles can also be disabled to make more space in the story.

Baldev Bhatt