Tableau is one of the most popular visualization tool in current era. Needs less to say that Tableau has lot of functionalities and is continuously evolving. It can not only be used for Data Visualization, but has powerful analytical capabilities too. And the key to master it is constant practice
Tableau provides multiple certificates in order to give official credit to users who want to have a career related to Data Visualization.
In this particular blog post, I want to specify few things that we need to keep in mind before taking up the Tableau Desktop Specialist Certification. We can register for the certification for any of the available from their site, and post registration we can schedule a date for taking up the Exam.
Please not that this is not a tutorial for Tableau, and simply a guide which we can use to brush up the concepts
For this particular Desktop Specialist Certification, the details are available in the link , and the guide to prepare for the exam is available here (PDF). It is recommended that we take up the class, but it is paid one and it is as per wish, not mandatory.
The skill that will be measured are mentioned in the PDF above and so we can keep that as our syllabus to have a high level picture to our preparation. If you are completely new to Tableau, I highly recommend to go through the videos, which will not only give you the knowledge on using Tableau, but will also cover the syllabus for this exam with hands-on examples, which you can also practice. Additional resources which will give you a hand for preparation would be link1 and link2. So in the next section we will jump to the actual purpose of this blog post.
In the upcoming sections we will go through the basics, which we might need to remember while taking the actual exam
There are few major parts that have been listed in our syllabus PDF above. Let’s cover those in the same order for convenience.
Connecting to & Preparing Data
Data Connections
New Data Source option will add an entirely data source, while Add in Connections, will add a new data to the existing data enabling a cross data base join with existing data
Connecting live leaves the data will leave in the data base or Source file.
Best when we need high performance data base capabilities, updates on real time, but will result in low experience
Extract
Will take query load of critical systems, will be faster inside Tableau high performance fast data engine database. We can select to import only few options choosing Extract and click on Edit
Metadata
In tableau server , we can’t edit the metadata like changing or removing the hierarchy, but we can create Calculated fields
If we extract the data after using a Live connection from the Data pane, we will be prompted for saving the extract as a .tbe, if the work book has already been saved as .twbx packaged workbook, the extract will also be saved there
Data Blend
One way of combining the data, instead of joining the data at row level, data blend will aggregate data in Tableau
One common field between the two data sources,
We will have primary and secondary data source based on which data source we use,
We can create calculated fields in blend data source
* important aspect in Blending
* indicates multiple values from one source to another one
Manage data properties
The activities such as renaming the fields and assigning the alias could be done both while connecting to the Data source and while in the Dimensions and measures shelves. To change the default properties for a data field, right click on the field and select the filed that you wan to modify, so that the next the when you use it, it will be defaulted to that selected option on the start
Create basic charts
We have few charting types that are listed in our PDF, which we will come to know as we watch the entire video series. So I am not digging much into the charts and its types
Organize data and apply filters
Group: Lets us create new fields based on a combination of members and we can create based on the data that we are handling in multiple ways
Easiest way is to select the fields from the visualization and clicking on the paper clip icon to group those fields. We have another option of grouping as shown here to group fields from the Data pane. We can select the items in the window that pops up and we can rename the group. We also have the option to Add, Edit , or Ungroup the items. We also have the option to ‘Include Other’, and can be used when necessary.
Alternatively we can select the items from the view and can group the data points. This kind of grouping is called Visual Group
The below image shows us another way of using Calculated field along with Parameter so that the calculation can be made dynamic based on a field
Bins: Creating bins is very useful in scenarios such as segregating people based on Age Groups, Job sector, and so on We can create bins in Tableau as shown here
Sets
Depending upon the way in which sets are created they can take any one of the icons shown here. One way to create a Set is to select the data points and Click on the set icon on the on the pop-up that appears, which will create the set.
Another way to create a Set is to select the field and choosing the Set option from the menu, and specify the condition with which we need to create the Set. We can have custom rule with which we can create the Set, with the pop-up that comes while creating the Set
The Sets video is a great resource and I recommend to listen it to twice for better understanding
Filters
Filtering in Tableau is a widely used option to minimize the amount of data shown, or to give importance to specific details in the data. There are a lot of ways in which you can filter the data, which are on the filter shelve, within interactive filter, or in the View
- Drag a Categorical variable to the filters shelve
- Right Click on the field and select Show Filter
- By selecting the data on the view itself and clicking on ‘Keep Only’ or ‘Exclude’ will also filter, and place that in the Filter Shelve
Context Filters : Context Filters ca be thought of as an independent filter. All filters that are being set up will pass through the Context Filters. Context Filters are created for two purpose 1) Improve the performance 2) Create a dependent numerical or top N filter. Additional details regarding context filters can be found here
Date Filters : Filtering using dates is available here. Default pill type for a date is Discrete. Continuous date when applied to view and when we use drill down it will change itself to the next hierarchy instead of creating a new pill, where ass if the Date pill is used in discrete format, it will create a new pill along side if we expand the hierarchy
If we want to have a particular filer apply to different sheets, we can right click on the filter and choose the different options available in the menu there The order in which we use the filters matter a lot, know when and where we place it
Order of filters used in the Tableau are 1) Context Filters 2)FIXED LOD Expressions 3) Filters on Dimensions 4) INCLUDE or EXCLUDE LOD filters 5) Filter on Measures 6) Filters on Table Calculations
Apply Analytics to Worksheet
Sorting: The sorting functions similar to what we have in MS Excel, which arranges the data either in Ascending or Descending order, based on the data that is present in the column. But the column and the view in which you perform the sorting highly influences the way in which the Visualization comes up. Keep in mind that Continuous Green pills can’t be sorted from the pills but continuous fields can be sorted using the sort in the column header or the Sort button in the tool bar while the Discrete Blue pills can be sorted. Dimensions define the number of marks
When we want a Discrete pill to be sorted , we click on the Discrete pill to be sorted and click Sort, and fine tune that sort by using the option in the dialog
Reference Lines: Adds a reference line to our view, as it sounds and we can set the aggregation and whether is can with Table, Cell, Pane at our convenience
Trend Lines: Trend lines provide useful insights on our data. You can see the Trend Line option available in the image aside, but it is disabled, as Trend Line is not relevant in the current view and it has similar options like Reference lines. The P-value, R² values can be found for a Trend line after adding it to the visualization and hovering over the trend line.
The residual for a Trend line can be got by following the procedure shown in the image and the only format available is Access File
More info on Help: Reference Lines, Bands, and Boxes , Help: Reference Distributions , Help: Drop Lines , Tableau Blog: Why Bullet Graphs Pack More Punch Than Gauge Charts
Table Calculations : The Table Calculations are performed on the view that is returned and the way we access the table calculation is as shown
The point to note is that we compute the Table Calculation using fields that are already in the view which is rendered and depending on the field the option shown for the table calculation changes
Calculated Field: This is one of the most important aspect of Tableau, where you can create your own calculations in the way you like.
The side image shown here is one way to create a calculated field and the image below that shows you where you type in your actual calculation
We have a drop down where we can choose the Category of calculation and browse through the list of options available or else use the search bar to find on our own, the list of possible calculation types. As you create your calculations we can see if the calculation is valid or not in the same dialog box. Not only calculation, but we can also create a parameter using the above selection, by clicking on the ‘Create Parameter’ option. Parameters allow you to come up with scenarios or options that are not available in your data and create these values to put into your visualization. In short it will be helpful in creating what if analysis within our views
Sharing Insights
Formatting: We can open the formatting pane from the Format menu, alternatively from any field or region we can right click and check to see if we have format option and on selecting that we can see the Format pane . Format pane differs from what we have selected and not a constant one
We can copy the formatting from one sheet to another as shown by the option in the image
Highlighting
Dashboard and Stories:
We can use the sheets that we made in the dashboard and, sheets and dashboards in stories Updates to sheets reflect in both sheets and dashboards
Updates made to dashboard sheets will reflect in individual sheets and dashboards, but updates made to sheets inside stories do not reflect back to sheets
Action inside a Dashboard : there are three types of Actions that can be used inside a Dashboard, which are 1) Highlight 2) Filters 3) URL
We can save the Tableau Work space with the following formats mentioned here , here (image and pdf)
Points to note
Measure are metrics and inside Tableau they are aggregations
Dimensions are Categorical fields
Most of time Measures are continuous and and dimensions are discrete, but not always true, we can convert them also Discrete pills are Blue and Continuous pills are green
Continuous pills when brought to view forms axes, while Discrete forms labels
Continuous pills gradient when brought to color, while discrete pills create color palette
Discrete and continuous dates are shown in the below image
Cross database joins integrates two data sources at the row level to create a single data source, blending takes separate query and aggregates in views
Conclusion
I have listed few major points that we need to keep in mind before we take up the exam, also it is best if we have constant practice and key to master is the same. We can use the Tableau Public for practice and having used that, there is not much difference between Tableau Public and Tableau Desktop
But if you want Tableau Desktop we can get the trial version for limited time from their Website, or take up the course here , which provides Tableau Desktop for free for extended limited time
If there seems to be any correction that is required in the above blog, please do mention them in the comments section, so that myself and people who refer can get to know about them
