Sign in or start a free trial to avail of this feature.
4. Understanding the Tableau Prep User Interface
The Tableau Prep user interface may be quite different from what most Tableau users are used to. In this lesson, we go through the basic UI features, such as the starting screen and the flow pane.
To explore more Kubicle data literacy subjects, please refer to our full library.
This goal of this lesson is to explore Tableau Prep’s interface.
Tableau Prep’s actions all take place on the Flow pane. It contains a series of icons which represent a data cleaning process. The steps connect with lines which show their relationship to each other.
This pane shows us a summary of our data for reference. Tableau Prep represents each field as a card. Cards provide a quick overview of the data.
If the data is text-based, we’ll see a bar chart of unique values.
If the data is numeric or date based, the information will be viewable as a histogram. This is very useful because it allows us to spot outliers in the data.
You can hide the profile pane to get a clear view of the data in a regular table format.
In the previous lesson, we imported our data set to Tableau Prep. In this lesson, we'll take a break from our sales data set and explore Tableau Prep's user interface using one of the preloaded flows.
We'll start by simply opening up a Tableau Prep. Here, we can see a few options. As demonstrated in the previous lesson, to the left we have the Connections pane for connecting our data to Tableau Prep. In the center of the screen, we can see a variety of recently used flows.
A flow is a type of project file. It stores all of our data cleaning actions for future use.
To the right, we have the Discover pane with a selection of tips and resources, such as official Tableau blog posts.
At the bottom are two sample workflows accessible to all users.
Let's have a quick look at a complete flow by clicking on Superstore.
This might look a little intimidating at first but it'll be much clearer when we break it down.
Each icon in the flow represents a step in the data cleaning process.
The lines connecting the steps indicate their relationship to each other.
Let's have a look at the flow's path by starting from scratch.
Tableau Prep automatically places any new step in the top left-hand corner. The very first step appears here. This first step simply added data to the flow.
This data was then cleaned with a cleaning step. Because the cleaning action and the input step interact with each other, they're linked with a line.
After this, three new data input steps were added.
None of the four data input steps interact with each other so they all appear on their own line.
Each of these inputs are cleaned so they're each linked to cleaning steps on the right. All this data is then merged together with one single step.
Below another set of input and cleaning steps were added, this was also merged with the rest of our data.
Our merged data set was then cleaned.
It was then split into two branches.
In one branch the data set was outputted to a local file.
We won't review the steps in the second branch as these consist of actions similar to the ones we've seen so far.
This is a lot of new information, so no need to worry if you find it a bit overwhelming. You'll be familiar with each of these steps soon.
Let's now click on a simple cleaning step to get a view of the profile pane.
The profile pane shows us a summary of data for our reference. Each of our fields are presented as cards. At the top of the card, we have the field name, the data type, the number of unique values, a quick sort tool and a search tool. These cards display our data differently depending on the data type. Date/time, and numerical values are presented as histograms in ascending order.
A histogram separates the data into bins; for example, in sales the first bin contains sales values between zero and 1,000. The second bin contains values between 1,000 and 2,000 and so on. This is useful because it allows us to quickly identify outliers in our data.
We can see that all but one sales entry have values less than 9,000 but then there's a single value of over 22,000. This may well be a legitimate value but it could also be an entry error, perhaps this should have read 2,200, either way the possibility of an error in our data makes it worth investigation.
We see something similar for text data.
Each unique value is laid out in alphabetical order.
We can also see an overlay of a bar chart in the card view.
This gives us a brief view of the frequency of each value; for example, we can see in the State card that the bar for Florida is just over twice as long as a Georgia bar.
When we hover over Florida, the summary tells us that there are 383 rows which read Florida. Hovering over Georgia shows only 184 rows, just under half.
Beneath our profile pane, we can see all of our data for reference.
If we click the Hide profile pane button, we can see a more complete view of our data.
Let's stop here. This completes our lesson on the Tableau Prep user interface.
In the next lesson, we'll start the data manipulation process with a cleaning step.