Languages

The first important element of our data story is to know where each language is spoken in Switzerland. It is important to understand the repartition of languages throughout the country in order to fully appreciate the results we will observe later on.

Complementary dataset

The main dataset we use in this project does not contain any linguistic features. We then need to find a complementary dataset. The data comes from the Atlas of the BFS (Federal Statistical Office) and gives the language predominance in each municipality in 2000 (which is quite at a time in the middle of the timeframe of the votations dataset, i.e., in the middle of 1981-2017).

The language predominance can take 9 different values in the dataset. These values represents a strong or weak dominance of one of the four national language or no predominance. The strong predominance is when 85% or more of inhabitants of the municipality speak the language as their first language in at home. Similarly, a weak is between 70% and 84.9%. Below 70% it is considered that there is no net predominance (the classification is defined by BFS and we cannot change it).

The dataset required some cleaning. The municipalities in the initial dataset are the municipalities as in 2000. However, many of them have merged together since, creating new municipalities. We added manually the mergers, plus some values which are copied with a different name in order to match the name of the municipalities in the votation dataset and/or the geojson used to draw the maps. The dataset after cleaning is the file languages_2000.xlsx in the folder data of the project.

Map of the official languages in Switzerland

Let us start by looking at the repartition of languages in Switzerland. It is reported that in 2015, the repartition was the following (Wikipedia):

• 63.0% German
• 22.7% French
• 8.4% Italian
• 0.6% Romansh

Which is quite on par with what we’re seeing on the following map drawn using our complementary dataset with data of 2000. Romansh is quite visible despite being nearly extinct; this is because the communities speaking the language are situated in low populated alpine regions and are thus quite big in size.

See the html map here

Dominance	German	French	Italian	Romansh	None
Strong
Weak

Presentation most and least voted proposition by language

The next thing we will do is to take the five most voted yes and the least voted proposition by language to see if we can discover some trend.

Allemand: forte Allemand: moyenne Français: forte Français: moyenne Italien: forte Italien: moyenne Romanche: forte Romanche: moyenne Pas de dominance nette

First we can see that there is some subject that are unanimously liked, by example : “Transfert de la commune de Vellerat au canton du Jura “ and “Soins médicaux de base” where it is in the most voted list of every languages and on the other side of the spectrum some vote are disliked by all languages by example :”Initiative «Remplacer la taxe sur la valeur ajoutée par une taxe sur l’én”.

But we can also see some vote in which some people loved it and some hated it : by example : “Rues pour tous” or “Loi fédérale sur l’assurance-maladie” which is loved by french and hated by italian . That happens also between commune of the same majority of people by example : “Initiative des petits paysans” is loved by commune with a lot of french but is hated by communes where there is a majority of french but less than the other group. The same can be said about “Surveillance des prix (contre-projet)” for italian. An interesting thing to note is that there is some several times we have a vote that is hated by some of the groups except for the group that has commune with a strong population of german speaking people, it happens in “Initiative «Remplacer la taxe sur la valeur ajoutée par une taxe sur l’én”, “Oui à l’Europe!” and “Initiative des quotas”.

There is also case of vote that are loved by one group but ignore by the rest by example “Formation professionnelle et recyclage” only loved by the group of strong german speaking communes, this case is especially true for the romansch which have several of such.

In the end, even if there is some consensus for all group as a whole we can see that language somehow divides political opinions on subjects, even if for now, with such a simple analysis, we are not sure of how this difference shows, but this allow us to take a first glimpse at this division.