Calculations Flashcards
What are table calculations?
“Table Calculations allow you to compare any number on the screen to any other number on the screen.” - Marc Rueter
“They are not computed fields, they summarize the data you’ve already pulled back from the data source.” - Christian Chabot
“Table Calculations are used when you want to COMPARE different summarized values of the same measure (EX Sales from 2011 minus Sales from 2010, EX2 Sales from the East divided by total Sales in all Regions). In contrast, Calculated Fields are used when you want to perform calculations with multiple measures (EX Sales minus Cost).” - Jesse Gebhardt
“Table Calculations allow you to perform an additional level of aggregation on top of the results returned in the table.” - Danton Lee
“They are Tableau-side (client-side) calculations that do additional aggregations, like Running Totals, for example.” - Nigel Stoodley
Cohort Calculations
Cohort Analysis is used to study the behavior or outcomes associated with a group of people over time. The technique of cohort analysis is most often used to study customer retention. A cohort is a group of customers or subjects that have a common characteristic. Cohorts can be determined by the date at which they joined a website, became a customer, their age, demographic, or any other attribute that could be used to group a set of individuals. It is then assumed that something about the cohort drives certain behavior over time.
Calculations that are built into the query and sent to the database?
All calculated fields that don’t use Table Calculation Functions.
Calculations that are computed locally by Tableau after we get the query results back?
All Table Calculations.
What’s the difference between a calculated field that uses aggregation and one that doesn’t?
If you do not put aggregation into the calculated field, it will be evaluated row-by-row through the whole dataset. Then aggregation is applied afterword.
What are some ways you might use logic statements (IF and CASE statements)?
To set KPI’s. To account for outliers or special business rules. To manipulate data (split 1 measure into multiple slices). Cohort Calcs.