Diego's journal (DS12) - Week Three

by Diego Parker

Week Three (17/12/18 – 21/12/18)


During week three we went back to ETL, starting with a deep dive in Tableau Prep followed by a very intensive couple of days learning Alteryx Macros. We finished our week with an introduction to statistics in data visualisation and the different tools that Tableau and Alteryx provide to do descriptive and inferential analytics. One of the different things about week three was that we didn’t have any class with our usual coaches (Carl and Andy) but previous Data Schoolers who now work with the Core Team.


What did we do this week?


Monday was about Tableau Prep.

The day was called “Tableau Prep Deep Dive”, which meant an intense eight hours hands-on session in Tableau’s ETL tool. The person in charge of teaching us that day was Sasha, an “original” (DS1) who now works with the Core Team.

The day started with a review of the basics of Tableau Prep, such as Unions and Joins, pivoting and aggregations. Like in every class at The Data School, we quickly moved into ‘learn by doing’ and started doing a series of exercises that involved working with data roles, multiple data sources and scaffolding. In the afternoon, Ian (DS4 who works at TIL Ireland) took over and as the Alteryx expert he is, asked us to try different Alteryx challenges using Tableau Prep.


Tuesday and Wednesday were about Alteryx Macros and Apps.

The two following days we were introduced to the concept and workings of macros in Alteryx. In small words, macros are workflows that are saved as a single tool that later can be inserted into other workflows. By doing this, the user to saves time by building just once a workflow that may have to use in repeated occasions. For example, one of the macros we had to build in Alteryx had to simulate the “% from previous” calculation in Tableau. Now, instead of having to build that workflow again and again every time I have to do that calculation, I can just throw the macro in between.

Whilst the previous example correspond to a standard macro, Ian also introduced us to Iterative and Batch macros. An iterative macro runs a workflow repeatedly until a condition set by the user is met. In the other hand, a Batch macro runs a workflow repeatedly until for each selected record of data.


Thursday was about statistics.

After our two intense macro days we moved into an even more intense topic, statistics. For someone without a background in statistics, having an eight-hour introduction to the topic can be a bit overwhelming. This time we had Beth (DS4) teaching us, who did her very best in making it easier for us to understand the different intricacies of the topic.

The first part of the day we focused on descriptive analytics. We mainly used Tableau to understand and build histograms, boxplots, standard deviation lines and control charts. Besides the theoretical part, we also learnt some Tableau skills, such as how to create bins, parameter controls and window calculations. For the second part of the day we used both Tableau and Alteryx to do basic inferential analysis. We learnt (or tried to) about regression analysis, z-scores and p-values.


Friday and the macro project.

Ian wasn’t in the office anymore, but he left us a task for our macro project. We had to create a suite of table calculation tools in Alteryx that resemble the calculations available in Tableau. The main task was to create a percent of total and a percentage change from previous tool. If we had time, we also had to create a rank and a moving average calculation.

Whilst in the last couple of weeks we have been having help from DS11, this week we had to rely in each other and a limited amount of core team and DS’ers that were in the office that day. I didn’t have many problems figuring out the required calculations but configuring the macro to be dynamic and user friendly was a different story. Luckily, we got some help, which also involved learning new Alteryx tools. However, the time still wasn’t enough for me to complete all the required tools with the user-friendly configuration and I had to go home with some homework for the break.

As the previous weeks, the day finished with the presentations from DS11. This time they presented two different dashboards from their dashboard week, one of their own choice and the last one. This was a good opportunity for us to grasp the feeling of dashboard week, which will come in a couple of months and see the different challenges that it means (such as using Power BI instead of Tableau).


If you have any doubts or comments feel free to use the box below or contact me in Twitter @DiegoTParker

© 2022 The Information Lab Ltd. All rights reserved.