06 - Filtering Dates, Measures & Strings Flashcards
Explain how can I filter out all the values of Sub-Grouping that are not selected on the screenshot below.
Select the data to be kept/excluded, hover over the selection and choose either Keep Only or Exclude. This will create a filter on the Filters card.
How can I quickly filter data to show only the sum of Quantity for the months of May throughout the years. By quickly, it is meant that we do not want to drag pills to the Filter card and configure the filter (several steps).
Click on the May label in any of the years and select Keep. If I click one of the bars for May, it would filter by that specific month of May, i.e. the combination May-YYYY.
Using this approach, a filter is created with just two clicks instead of many more.
How can I create a filter using Ship Date as a continuous field?
Right-click and drag it to the Filters card. Then select, Range of Dates. Set the filter accordingly.
Then, we will set a green pill for Ship Date inside the Filters card, indicating that is being used as a continuous field.
What is the difference between Discrete and Continuous dates in Tableau Desktop?
Discrete dates allow us to view parts of the date such as year, quarter and month. Whereas Continous dates look at these parts in combination over the full time period.
How can I create a filter by Ship Date that shows only the data for the previous 3 months relative to 31/08/2021?
- Right-click on Ship Date and drag it to the Filters card.
- Choose *Relative Date
- Select Month and set it to the Last 3 Months.
- Anchor it relative to the date provided (31/08/2021)
- Click OK.
When filtering a measure, what is the main difference between selecting the red or orange options?
Filtering by All Values (red box) will filter at row level, meaning every row that does not meet the criteria defined, will be excluded from the visualisation.
Filtering by Aggregate functions (orange box) will filter at the level of the aggregation, meaning it will not touch the individual rows on the table.
Setting the same threshold for the two options will result in completly different results.
Explain how can we create a filter that prevents ingestion of data that is not needed. Also, why would this kind of “pre-filter” be benefitial?
Note that we do not want to create a filter at Sheet/Dashboard level but rater at the Data Source level.
Either go to the Data Source page and click on Add on the “Filter corner”, or right-click on the Data Source Connection on any Sheet and select Edit Data Source Filters
If we select to keep only the data where Region = “Central”, we will see that the number of rows in our dataset will be reduced accordingly, i.e. there will only be rows where this condition is met.
Filtering data at this stage is particularly useful when dealing with big datasets, where loading and rendering results might take several seconds/minutes. Also, we can think it through an efficiency perspective, i.e. why bring in data that is not going to be used for analysis?
The visualisation below displays the count of sales per customer. Explain how can I filter the data by the Sales measure to only include sales with a total between $10 and $50.
We want to filter data at a row level, meaning sales outside the $10-50$ range must be excluded.
1. Right-clicking the Sales measure and drag it to the Filters card.
2. Select All Values to filter out at the row level instead of the aggregate level.
3. Set the filter accordingly using the option Range of values. Then, click OK.
**Note how the value for Aaron Bergman went from 6 to **3, meaning he had 3 purchases where value under the Sales column lied outside the defined range.
How can we show the filters applied so users can quickly change them once the dashboard has been published?
Right-click the pill on the Filters card and select Show Filter.
A window will come up on the right-hand side with filters applied and will allow user to edit them.
Explain data is being filtering when there are multiple filters in place.
Only the data that meet the three filters will be showed. Think about multiple criteria combined together using AND.
Associate each visualisation to either explanatory or exploratory, and explain the underlying rationale.
Left-hand side: Explanatory
Lack of controls and addresses a very specific question (which sub-groupings have been underperforming). The conclusion is already evident, not need to play with data.
Right-hand side: Exploratory
Controls (filters) allow users to deep dive into the data and try to draw their own conclusions.
What is the benefit of having the Apply / Cancels buttons on the Filter? How can I show these?
Without the Apply button, every time the user changes the filter, the visualisation will render again to reflect the change. If we are dealing with a** big dataset**, that means stakeholder might have to wait several seconds after each change they’ve made, which can be annoying. By adding the Apply button, the rendering will take place only when they hit the button, saving them from unnecessary loading time.
To show these buttons, on the right-hand side, on the Legend section, hover over the filter and click the expand button. Then, select Customize and click on Show Apply Button.
What is another way of creating a filter without having to drag the corresponding pill to the Filters cards?
Click on the pill, and then select Show Filter.
This is more efficient than dragging the pill to the Filters card.
The visualisation has been filtered by the state of Alabama, but City and Postal Code filters still show all options regardless of whether these belong to the this state or note.
How can I fix that so when users select a state from the dropdown list, only the relevant cities and postal codes are shown on the filter?
On the right-hand side, on the legend, click on the expand button for the filters and select Only Relevant Values
