BI for Beginners Session 4

Session materials

Ideally, you’ll use the data that you wrangled during the previous session. But if that’s not suitable, use this data:

• .xlsx session data

There’s also a starting-point pbix file which you might prefer to use instead:

• .pbix starting file

Previous attendees have said…

• 60 previous attendees have left feedback
• 100% would recommend this session to a colleague
• 97% said that this session was pitched correctly

Three random comments from previous attendees

• Sessions are great - some info went over my head a bit, but Brendan is excellent at describing everything and also having alternative methods of working so everyone can follow.
• I’m a beginner with some previous experience of being an end user getting data from powerBi so I have an understanding of the possibilities. I found the method of clicking along very helpful tied with Brendan’s positive presentation skills in openly checking issues and answering queries most beneficial. The level of content was right for me until the last session - I have no experience of DAX so this was all newer. I have been able to apply the skills I have learned with confidence. I feel I’m still lacking a better understanding of the semantics model and if I have a recurring set of data ie datix incidents, how I can update the data in a dashboard without having to rebuild the dashboard each time.
• Really interesting about use of measures.

Session outline

• In this session, we’re going to build out the GP practice size dashboard from last time
• we’re going to use some new features of Power BI to help us with that
  • we’ll spend a little bit of time looking at the model
  • but most of the session is about DAX and measures

Session files

• BfB_s03.pbix as starting point
• BfB_s04.pbix as completed example

Recap

• we did lots of data wrangling last week using Power Query
• that’s designed to do data loading and tidying - so we tried removing some data, reshaping data, and joining data,…
• but there’s another whole language in Power BI to let us do things with that tidy data: DAX (annoyingly, Data Analysis eXpressions)
• similar to Excel formulas

Major differences between DAX and Excel formulas

• Excel takes a cell or range of cells (like A4)
  • DAX functions take a column or table as reference, and cannot use e.g. cell ranges
• Excel formulas show their results directly in the cell
  • DAX formulas need to be added to a card (or similar)
• also a grab-bag of minor differences (no mixed columns data type, date and time functions, different lookup mechanics)

Set up

Task

1. open your dashboard from the last session (or you can use BfB_s03.pbix)
2. add a new report page, and name it measures
3. remind yourself about the data we’re working with

The model

• let’s start by looking at the model view
• we should see our data tables as blocks in the model
• thre are also connections between some of those blocks

Task

1. Mouse-over the relationships in the model view
2. What’s the difference between and ?
3. Now expand the Properties pane

DAX first steps

Task

1. in either the report or the data views, click New measure
2. in the formula bar, add largest_practice = MAX(practices[PracticeListSize])

DAX queries

The general way of building a DAX formula is as follows:

measure = FUNCTION(table[column])

• note the [square brackets] for column names
• we wrote a simple DAX formula in the first session, when we calculated a column using the formula AvailableBeds = [BedComp] - [ClosedBed]
• in this session, we’re going to concentrate on using DAX to write measures

WTH are measures?

• measures are Power BI lingo for calculated data summaries
• calculated columns, as we’ve seen before, let you make new columns from existing data
• measures allow you to make non-column summaries of your data - like an average

DAX queries

For our example, our measure is called largest_practice, and we want it to report the largest practice in each health board. Our DAX-based measure works as follows:

• our data table is called practices
• our column is PracticeListSize
• we’re using the MAX() function to find the largest value (exactly like Excel)

largest_practice = MAX(practices[PracticeListSize])

Adding our measure to our report

• we should be able to add our largest_practice measure to our report
• go back to the report view, and refresh
• you should see the largest_practice measure appear in the data pane
  

Adding our measure to our report

Task

1. in our new measures page, drag the largest_practice measure into a new card
2. we also want to be able to look at different boards, so add a slicer using the HBnames from the practices table
3. we’ll then play with the interactions to make sure that this measure gives us a sensible result
   

Making a measures table

• it’s helpful to keep your measures in a stand-alone measures table - but requires a bit of creativity

Task

• manually create an empty table using Enter data
  
• select your largest_practice measure, go to the data view, and use the Home table menu to move it to your new measures table
• then delete the blank column from your measures table

Concatenating

• so far, our largest_practice measure is ugly-but-functional
• we can use the concatenate operator (&) to beautify it

Task

1. edit your largest_practice measure
2. you can insert text by “quoting” it, and using & to join it to the result of your MAX()
   • for instance, largest_practice = "Largest practice: " & max(practices[PracticeListSize]) & " patients"
3. feel free to experiment - make sure you go back to the report view and refresh to see changes
4. you can also use the format visual section to change the font size, and drop/alter the category label

Core DAX functions

• several useful DAX functions work in basically the same way as MAX()
• let’s try them out now

Task

1. in your measures table, make a new measure called average_practice
2. try average_practice = AVERAGE(practices[PracticeListSize])
3. add this to a new card to test it
4. now tweak with concatenated text
5. COUNT(), and MIN() should also work in the same way

Combining functions

• most measures will need more than one DAX function
• we combine functions just like Excel, by bracketing them together
• function1(function2(data)) does function 2 on the data, then function 1 on that result
• when you’re reading/building formulas with many functions, two top-tips:
  1. make sure you match your brackets
  2. read from the inside out

Combining functions

• our average_practice measure is pretty ugly
• a whole number value would be more suitable

Task

• click on the average_practice measure, which should read average_practice = AVERAGE(practices[PracticeListSize])
• add the INT() function, which converts a number to an integer (whole number)
  • we want INT() to wrap around the AVERAGE() function
  • average_practice = INT(AVERAGE(practices[PracticeListSize]))

Combining functions

• we can keep combining functions together in the same way
• e.g. we could round that result to the nearest hundred using round(..., -2)
  • the -2 specifies the number of decimal places, so 3 here would give us 3 decimal places (say 444.444).
  • average_practice = ROUND(INT(AVERAGE(practices[PracticeListSize])), -2)
• care with the brackets!
• you might prefer to do simple rounding in the format visual area
  ){height=“150px”}

Filter

• FILTER() is the last DAX function we’re going to look at
• it lets us filter our data based on conditions
• say we want to count how many big practices (more than 10000 patients) we have…
• big_practices = countrows(filter(practices, practices[PracticeListSize] > 10000))

Filter

• big_practices = countrows(filter(practices, practices[PracticeListSize] > 10000))
• let’s read this from the inside-out
• filter(practices, practices[PracticeListSize] > 10000) applies the filter function over the whole practices table
• it finds all the rows where PracticeListSize is greater than (>) 10000
• those rows are then counted by countrows()
• finally, we save that number into our measure named big_practices

Filter

• that’ll be another ugly-but-functional measure
• we’ll use an alternative way of beautifying this one, to avoid making the formula too complicated
• measures can contain measures
• instead of the table[column] format we’ve been using, we refer to measures by wrapping them in single square-brackets: [measure]

Filter

• let’s create this now

Task

1. add a new measure to your measures table
2. use the formula big_practices = countrows(filter(practices, practices[PracticeListSize] > 10000)) to obtain the filtered count of big practices
3. create a second new measure
4. nice_big_practices = [big_practices] & " practices with more than 10000 patients"
5. add nice_big_practices to a new card

Beautifying average_practice

Task

1. Please make a new meaure named nice_average_practice
2. Use concatenation to make the average_practice measure nicer to read
3. Update your average measure card in the report

More interesting filters

• FILTER() is one of the big building-blocks of Power BI
• we can use it as a way of looking up matching data
• for example, to find the name of the largest practice, we can:

largest_practice_name = CALCULATE (
    VALUES (practices[Practice Name]),
    FILTER (
      ALL(practices[PracticeListSize]), 
      practices[PracticeListSize] = MAX (practices[PracticeListSize]
            )
            )
)

• FILTER() finds the largest practice
• VALUES() returns the corresponding value from the Practice Name column
• CALCULATE() joins the FILTER() and VALUES() parts of the expression

DAX resources (beginners)

• DAX sandbox - dax.do
• DAX introduction on the Microsoft site
• cheat sheet from Pragmatic Works - good as an introduction
• introduction to DAX from Data Flair is particularly useful if you have a bit of prior experience
• Saurabh dasgupta’s DAX cheat sheet

DAX resources (advanced)

• DAX reference
• DAX formatting utility
• Helpful tutorial on DAX variables
• Marco Russo and Alberto Ferrari’s book The Definitive Guide to DAX: Business intelligence for Microsoft Power BI, SQL Server Analysis Services, and Excel is the standard big reference book
• I’ve also heard good things about DAX Studio as a helper for writing more complex DAX functions - broadly analogous to an IDE for DAX