Hello again everybody, welcome to the final episode of Dashboard Week. In this episode we were given a dataset that essentially compares various countries for their livability based on various other factors. Unlike the other dashboard week days, this day was an extra tight time crunch because we only had until 3pm to build, and had to present both the previous day and the current days dashboards at the same time.

The Data

The dataset itself is very interesting, at least to me. A lot of the aspects that they chose to include were things I had never really considered before, like number of long work days, or how safe you feel walking home at night. Other things I think really should have been included, like how good the food is in that country, or how long typical commute times are. Or even some sort of free health care system presence in that country.

Data Supplementation

By itself, the data wasn’t enough for the story that I wanted to tell. So as always, its time to supplement. I wanted to tackle this challenge from the angle of what is important to me. So I had to find the extra data that supported what was important to me.

If I had to pick my dream life in four-words, it would be WORK, EAT, CREATE, SLEEP. A lot of my enjoyment in life comes from either doing something really productive, or solving some problem. But there’s more to life than work, so it would be good to come home and be able to enjoy my other passions.

I would almost say that the most important thing in life is FOOD. So I managed to find some data on food affordability and food accessibility. I also found data on daily commutes of every country. All of this was starting to build a good selection criteria for a new country.

Data Transformations

The main issue with this data is the lack of standardisation. Some columns were percentages, some were absolute values. But we need a way to compare all values to each other in a standardised way. After finding some info on standardisations I managed to get all my data along the same scope. That way each can contribute to the overall country score.

Radar Charts

I’ve never made a radar chart before. As it turns out, there’s enough tutorials online, even one made by our very own Chris Ktenas. So building it was as simple as copying some calculated fields in Tableau. I actually really do like the radar chart but I do think the barriers to entry in terms of understanding could scare people off. Either way, dashboard week is a perfect time to try creative ventures.


In the end, what let me down was my timeboxing. I spent too long looking for additional data, and spent too much time trying to make fancy parameter sliders. I took on too much for such a small amount of time, and as a result, my viz wasnt as polished as I liked. But still, with a little more time, this viz would really look great.

What I wanted to do was create a viz that would recommend a country to live in based on your specific requirements. These requirements would be manually inputted through parameter sliders. And since I didn’t want 12 ugly looking parameter sliders, I tried to make fancier ones but that was a huge time investment.

Afterwards, you could see how well your requirements matched up with that specific country (how well the radar charts overlapped).

What I did realise afterwards though is that the calculations required to recommend a country are actually more complicated than just adding everything together. Really there should be a penalty if a country doesnt meet a requirement and a bonus if it exceeds it. So for now, although the dashboard looks cool, it needs a bit more time to really polish.


The Data School
Author: The Data School