Data visualization using Tableau Public

Data visualization is the presentation of data in a graphical or pictorial format. Because of the way the human brain works, in processing information, it is faster for the human brain to understand many data points when they are displayed in visuals (charts or graphs) rather than going through many spreadsheets or pages and pages of reports. Let’s say for example you have thousands of spreadsheet reports that you need to understand in a short time to make a decision that is informed by the data. To make that data-informed decision, you need visualization.

Tableau is one example of software that can be used to create visualizations. Tableau helps us to quickly analyze, visualize and share information. We are going to go through a step by step process of creating a visualization using the freeTableau Public software.

We shall use this sample data set which is in Excel format (.xlsx). It looks like the diagram below when open in Microsoft Excel.It’s a sample of premature admissions in Kenyan Hospitals in different regions.

Pre-mature Admissions

1)      Download a copy of Tableau Desktop software from if you don’t have it installed already. Install the downloaded software. Download and save the Excel file or use your own data set. Drag and drop your data set to the Tableau Public shortcut on your desktop or onto an open Tableau Desktop application. Tableau will detect the settings of the file you are trying to open.

Check diagram below:

Import Data

Click ok. Tableau will momentarily display a small window written “processing request”. After it has completed, you will get a window that has field names, dimensions and values that is left aligned to the parent window and a working pane that is right aligned to the parent window. Measures will contain numerical data like number of number of admissions while dimensions will contain categorical data like sex.

Open Workbook

Lay out to create the visualization

Tableau has many charts and graphs that you can use to visualize data. Each one is suitable for different visualization depending on what you want to achieve. Some guidelines to choosing the right type of chart are as below;

  • A line chart is suitable when you are trying to show change over time.
  • A map is most suitable to show location i.e. where something is. Maps can also be used to show concentrations through color gradients.
  • A bar chart is most suitable when you are comparing data categories. A double bar chart can be used to show comparison.
  • For more information on selecting which chart is best suited for your needs, please visit the tableau website and download the white paper.

Let us now create a double bar chart that shows the number of pre-term admissions in hospitals from the highest to the lowest. At the same time, it will show a comparison between the number of pre-term admissions and the number of incubators in each hospital. The final visualization will look like the diagram below:

Final Viz

To achieve this layout, drag Hospital from Data to Columns, Pre-term admissions to Rows and Number of Incubators to Rows. See diagram below:

Drag and Drop

Next, we need to sort the number of pre-term admissions from highest to lowest so that it is easier to make inferences like which Hospital has the highest number of pre-term admissions or which one has the lowest. To do this, click on the sort icon at the toolbar, as shown below:

Sort Highest to Lowest

Next, we need to sort the number of pre-term admissions from highest to lowest so that it is easier to make inferences like which Hospital has the highest number of pre-term admissions or which one has the lowest. To do this, click on the sort icon at the toolbar, as shown below;

Double bar Graph

Since we have a legend showing what the two colors represent, we don’t need the labels below the bars. It clutters the graph. So we want to remove the labels. To do this, right click on any of the labels and uncheck the Show Header checkbox.

Hide Labels

We should now have the layout as below:

Labels Removed

You have just created your first interactive double line bar graph using Tableau. Hover over any of the bars and you should get more information.


To save your workbook, go to File Menu in Tableau desktop then select save to web. Tableau requires that you are logged in for it to save your workbook into their server. If prompted, enter your username and password. Tableau will ask for a name for your workbook. Enter a name and your workbook will be saved onto the server. After saving, you will be presented with a window that gives you a preview of your graph, a link to send via email, and html code that can be embedded onto a website/blog.

Workbook saved

You will notice how easy it is to make inferences at a glance just by looking at this graph as compared to the spreadsheet. For example, you can easily tell that Kenyatta and Pumwani Hospitals have the highest discrepancies in ratio of pre-term admissions to incubators. Pumwani Maternity Hospital had the highest number of pre-term admissions while Garissa Provincial Hospital had the lowest number of pre-term admissions.

A Tableau dashboard is a more powerful way of visualizing data. It enables the display of several related visualizations on a single pane.Click here to view a video on how to make a Tableau dashboard.

For more information on how to use Tableau, visit

Daniel Cheseret
About the Author: Daniel Cheseret
Few developers in Kenya would consider applying for a data journalism fellowship but not Daniel Cheseret. Having realized that the growing importance of data journalism has created new career opportunities in the newsroom, he applied for Internews Data and Health Journalism Fellowship. I will learn more about producing infographics, data mining and creating news applications,” says the software, developer and webmaster. Among his achievements is developing a news management software for Citizen TV. While working with Royal Media Services, the company that owns Citizen TV, he developed over 10 websites for the group. After five years he left to start a consultancy firm.
Author Website:

Leave a Reply

Your email address will not be published. Required fields are marked *