07 - Visualisations, Functions, Maps and Density Plots Flashcards
How can I create an hierarchy between these two dimensions with Grouping being the highest level?
On the Data pane, drag Grouping and drop it over Sub-grouping. Then, name this hierarchy accordingly.
Finally, ensure that field corresponding to the highest level of the hierarchy is at the top of the stack of fields. Beneath is the next level and so on and so forth.
How can I overlay the two charts so that there is just a single x-axis?
Click on the right-most measure pill and select Dual axis.
When using dual axis in a visualisation, what are some changes that we can make to reduce the cognitive load for readers to interpret the data?
- Format the axis so they have the colour of the corresponding measure.
- Make the primary measure of interest stand out by adding transparency to the other measury.
- Use colors, make the primary measure of interest have a “strong” color (e.g. red) while the secondary measure should have a “softer” color (e.g. grey).
When is reasonable to synchronize axis and how can we do it?
When the two axis have similar scales. Just right-click on the axis on the right and select Synchronize Axis.
Can I drag and drop a measure directly on the area shown on the screenshot?
Yes, this is an alternative way of displaying the data as Text rather than dragging the pill over the Text button on the Marks card.
What is one scenario where adding multiple measures to Measure value is not recommended?
If one measure is on scale that is orders of magnitude higher than the other, then the latter will likely look like a flat line, adding no value to the visualisation.
Moreover, it can even be misleading, making people think that there are no changes to the measure with the smaller scale, when, in fact, these are just too small for the scale used.
When using dual axis, we can only have two measures at a time on the same axis. How can we overcome this and add as many measures as we want?
We need to use Measure Names and Measure Values. A simple way is to drag a second measure and drop it on the y-axis once the icon changes to two parallel green bars. Then we can simple drag and drop additional measures to the Measure Value pill on the Measures Values card.
Explain what are Measure Values and Measure Names.
Measure Values field contains all the measures in your data, collected into a single field with continuous values.
Measure Names field contains the names of all measures in your data, collected into a single field with discrete values.
Explain the difference between the two calculated fields below.
Calc_Field = [Profit] + 2 Calc_Field_2 = SUM( [Profit] ) + 2
Calc_Field is a calculated field that operates at the row level, i.e. for each row, 2 wil be added to the value of Profit.
Calc_Field_2 is a calculated field that operates at the aggregate level, i.e. the measure will be aggregated using sum and then 2 will be added to the result.
Looking at the example below when Sub-Grouping is equal to Action:
- SUM( [Profit] ) = 18,132.
- Calc_Field_2 just adds 2 to this total, resulting in 18,134.
- Calc_Field_1 will add 2 for each row where Sub-Grouping = 'Action'. As there are 466 rows that meet this criteria, this field will be equal to the SUM( [Profit] ) + 466 *2, resulting in 19,064.
How can we exclude the points inside the red box from the visualisation? Imagine that these are considered outliers and are not of interest to the stakeholders.
Just select the points that need to filtered out, then hover over them and select Exclude .
What is one limitation from adding labels by using the Labels button on the top bar compared with dragging a pill over to the Text button on the Marks card?
When we add labels using the Labels button, we cannot format the resulting labels on the chart.
The labels on the screenshot below were added by clicking on the Labels button. See how they have different units compared with the x-axis and there is no way to format this.
When adding labels by dragging pills to Labels on the Marks card, what is something that we need to be careful of?
Ensure that the pill on the Labels has the same AGGREGATION FUNCTION as the pill on the shelves.
On the example below, we can see that the labels are using SUM while the measure is using AVG, misleading readers as the actual value for each bar.
One way of avoiding this is by holding CTRL and dragging the pill from the SHELVES to the Label card, instead of from the Data pane.
How can I format the labels so they appear as a percentage with 1 decimal place?
Click on the pill under the Marks card and then select Format. Move to the Pane tab and set the Numbers field accordingly.
If I format the numbers on one sheet and then create a new one, then I need to replicate the formatting rules on it. How can I avoid having to repeat this over and over again?
Edit the Default Format Property. Click on dimension/measure on the Data pane, select Default Properties and then Number Format. Set it accordingly.
I double-clicked on a Geographical field to create a map, but nothing gets displayed as shown below. What is the problem and how to fix it?
Tableau does not recognize to which country the Geographical roles on the dataset belong to.
To fix this, I need to set up the Region for maps accordingly. To do that, double-click on any Geographical dimesion, then go Map (top bar) and choose Edit Locations.
How can I see the underlying data only for the select data point?
Select the desired data point, hover over it and click on View Data Point. A new window will open, showing you the data at the aggregate or row level.
How can we format the background on maps?
Click on Maps on the tool bar, then select Background Layers. A pane will appear on the left-hand side where we can edit the formatting options.
Using the example below, explain how do density plots work? Moreover, list a few ways on how we can tweak the visualisation.
Each data point is turned into circle with a blurred area. Whenever two or more circles overlap, the degree of “blurriness” will change.
If we put this change of blurriness on a spectrum of color, the more areas overlap on a given section of the map, the more the color will tend to one end of this spectrum. Contrarily, the less areas overlap, the more the color will approach the other side of the spectrum.
This visualisation can be tweaked by changing:
* Change size of the density plot
* Change color/intensity of the density plot
* Change background layers
* Add a dimension to the columns shelf to have multiple plots, one for each value from that dimension.
