Here is the URL link to the original dataset. This is my personal project about ParcourSup data, the French process to get to higher education and which type of students go to which type of higher education which will be published in multiple parts. I wanted to look at this data because higher education (also called post-bac education in France) is a huge deal and can also shows a lot about our society. It's also interesting to compare reality with what we usually view higher education. Last but not least, that's also a good exercice of data visualization and story telling which I hope you will enjoy.
Most figures are interactive: you can select an item in the legend (usually on the right side) to remove this item from the figure. You can also double-click on it to show this item only and remove the others. Another double-click will display back all items. If you have any comment or problem, reach me out on my LinkedIn!
Back to school! Or rather, its analysis. Last time, we learned a bunch of information about French higher education. This background was necessary to understand the available formations: their size but also who are the students that will elect this formation in particular. If you want a reminder, here is the link
Before delving deeper into the data, a quick geography lesson! As we will explore how formations are geographically distributed in France, a map is in order to understand how the country is administratively divided. The main division that will be used are the thirteen regions, the biggest subsections of the country. Compared to similar divisions in other countries (the States of the USA, German Länder), regions have much less powers. You may notice 4 islands (and Guyane) to the left of the map. These are overseas territories also called DOM-TOM. I have chosen to lump all (along with Corse) in one category: "Not Metropole". As a result, I will not talk much about it as this category is the sum of several isolated and sparsely (compared to the other regions) areas which do not have much in common otherwise. However it will make the figures more readable as I would not have been able to draw much conclusion from these territories anyway.
There are some big differences between them (Nouvelle-Aquitaine is huge, Ile-de-France is very dense) but there is at least one major city hosting many high educations institutes (universities, engineer and business schools etc.) in every region. Regions hold departments, the other main administrative division. However we will use regions for our analysis as department are smaller and much more variable. Most departments lack a university or hosts a very small one: our analysis would overlook them and focus too much on the biggest education centers. Using regions will allow us to get look at the broader picture without having to look at every single of the 101 departments.
But there is one last division that could be useful to us: Academies, shown in the below picture (note that Caen and ROuen have now merged). Academies are a strictly educative division. Each academy holds a rector which oversees the academy education (to make it very simple) and execute the government strategy. It would make sense to use them for this analysis, wouldn't it?
Well, that is not entirely true. For starters, academies often follow the previous regions frontiers with the same caveat than departments: some academies have much less smaller campuses (and overall higher education possibilities) than the others. That is less the case with regions. But the other and in my eyes main issue is that I want to show how some regions are disproportionally chosen... and some academies have been created exactly for the purpose to divide the workload on small but attractive regions (Paris and Lyon mainly, Marseille to another degree). Choosing academies will blur this information as Ile-De-France is composed of 3 academies, the smallest in size yet comparable to any other academy in the number of students. The rest of this analysis will show this phenomenon.
We will start with a map of every institution available in the data. The dataset provides coordinates which will put to good use. Each point will represent a formation (so there will be multiple points if one institution hosts multiple formations). The larger the point, the more students the formation can receive. The provided data is not always exactly on point but it is more than enough there. The initial zoom is enough to get a view across the whole country but you will need to zoom up if you wish to look at a specific area.
As expected, the biggest formation clusters are in the main cities. Small formations (especially BTS as they can be hosted by high schools) are available all over the country though. Another noteworthy information to see is the size of the formations which here directly visible through their dot's size. Licences and PACES are the formations with the most students. However interesting this map is, it still doesn't give us an aggregated view of the data, which is what we will look at now.
Ile-de-France is the region of Paris which explains why there are so many formations there. And Ile-de-France is dense: this first place is achieved over an area of 10 000km², smaller than any of the USA states except Rhode Island and Delaware. Note that Paris (and its surroundings) not only hosts many institutions but these often are the most prestigious ones. To call back from what I said about Academies and Regions, we can easily see how Ile-de-France is far above any other region. If we looked at the academies instead, each of the 3 academies composing Ile-de-France would still be in the upper half.
These formations aren't equally distributed though. Ile-de-France has a staggering lead in the number of CPGE places (far more than twice the closest other region) but is actually behind Auvergne-Rhône-Alpes in the number of BTS available places. BTS are not hosted in a university so it is not that surprising to see the larger region have an advantage there. Occitanie has more places in Licence than Auvergne-Rhone-Alpes. That also tells us that regions also shape what students will chose: the most prestigious CPGE are often in Paris, Occitanie with two big student-cities is logically ahead in Licence etc. Finally, the lowest regions besides DOM-TOM are regions close to Paris: students can go to the ever-so attractive capital more easily.
Another information to keep in mind: French regions can have very different sizes (even without Ile-de-France): Occitanie is larger than 84 000 km² whereas Hauts-de-France is more than two times smaller, and yet they attract roughly the same number of students.
That may not be eye-catching but if we compare this figure with the previous one, we can observe discrepancies between the two. For example, Occitanie is the third region by the number of applications, but the forth by the number of formations. And as expected, Ile-de-France trumps over any other region, having more than twice applications than any other regions. It is an impressive ratio so in order to understand why so many applicants choose Ile-de-France, we will now look at the number of applications by available places. Not metropolitan territories dropped to the last place in the number of applicants: my guess is that while locals may apply there, they may also apply elsewhere. However, few metropolitan will choose to go there.
The x_axis is maxed at 60: otherwise some formations would stretch the data too much (looking at you IFSI). You can manually zoom to see the difference in IFSI. It is basically the combination of the two previous figures.
The discrepancies shown above is explained by the BTS receiving 26 applications per available place in Ile-de-France which is much more than any other metropolitan region. It is not that surprising: if you remember the previous figures, there were about as many available BTS places in Ile-de-France than in Auvergne-Rhône-Alpes whereas we know the overall number of applications in the former is much higher. In PACES as well, as there are twice as much applications per place than in other regions. Regions with overall low-ish applicants to available places ratio include Normandie, Bourgogne-Franche-Comté or especially Hauts-de-France. These are neighbors of Ile-de-France so students are easily inclined to go there as stated before. Hauts-de-France is still odd though as Lille is an important city. I suppose the region is not seen as attractive enough by the applicants (it does have the reputation of having a rainy gray region).
IFSI also play a role in some of theses differences: PACA (Provence) and Occitanie have a staggering ratio, especially compared to Hauts-de-France or Grand-Est. I have been told southern IFSI schools have a much better reputation but I don't know why.
I want to test my hypothesis on whether students from regions neighboring Paris go there instead of their own region... let's use the "Same academy" data! It represents the number of student applying to a formation in their own academy I will cheat a bit because I group the data by region and a student could go from Grenoble to Lyon, which are in the same region but not the same academy. But that should be good enough.
Overall, academies welcome students from their academy. The degree of which vary (and we will explore it) but this is a reliable rule of thumbs.
Most noticeably, not metropolitan regions mostly welcome students from their own region. Students from the metropole for example usually don't apply to these territories. Ile-de-France has a high same-academy rate as well because of how populous the region is and because nearly every higher education formation is available. The neighboring regions also have a rather high rate but I don't think for the same reason as their population is much lower than in Paris but rather than students from other regions don't apply there (much like not metropolitan regions). However this effect is clearly not as noticeable as I would have thought except in Bretagne and Normandie.
On the other hand, Auvergne-Rhône-Alpes has low-ish rates compared to other regions, meaning that higher education there is attractive enough to encourage students from other academies to come (like Ile-de-France) and form a group nearly as large as students from the region itself (unlike Ile-de-France, most likely because it is not as populous). Interestingly enough, PACES always bears a high same-academy rate except in one region: Occitanie. As far as I know, the university of Montpellier (in this region) is particularly coveted by the students. On the other hand, Autre formation are
An odd case there: IFSI and EFTS in some regions (Nouvelle Aquitaine, PACA, Bourgogne-Franche-Comté) is amazingly low: students from other regions far outnumber the ones from the region itself.
That's it for where students go... but what about which student go where? Do students with better distinctions go in the same regions than the others?
The answer is: "Not quite". Excluding Autre formation as it's too diverse for analysis, the bars are more or less homogeneous but there are exceptions. For example, the left bar (TB distinctions) is usually the shorter except in one case: CPGE. And it is even more notable in two regions : Ile-de-France and Auvergne-Rhones-Alpes. It looks like a massive centralization of CPGE (which, as we note here, attract a large part of the students with a Very Good distinction).
Which lead us to the next figures...
Half of students in CPGE who have a Very Good distinction are in Ile-de-France or Auvergne-Rhone-Alpes, with a third just in Ile-de-France (while it represents ~20% at best of France population). These graphs share the same X-axis so it's telling to see how the Ile-de-France bar shrinks this much. If we exclude CPGE the repartition is still skewed towards Ile-de-France and Auvergne-Rhône-Alpes but noticeably less so in the former case.
It means that in what is seen (rightly or wrongly) one of the most elitist formations, the best students very often go to Paris or Lyon, which is not the case in others formations. These cities (especially the former) are the location of many CPGE, especially the most prestigious ones, which is not helping. If you are unable to go these cities as a CPGE student, you may have an harder time finding a formation.
This concludes this second post about exploring ParcourSup data. Overall, students like to stay in their academy even if some academies (especiall Auvergne-Rhône-Alpes and Ile-de-France) attract students from elsewhere. But in some cases, like CPGE for Très Bien distinctions or IFSI, these rules shatter as some regions (Paris in the first case, Occitanie in the second one) appeal to a much larger of applicants than expected.
While we investigated a lot of information about the students (where do they go, what are their distinctions etc.), we don't know a lot about the formations themselves. In the next post we will look at each formation types and analyze their main application indicators.