Terrorism, Public Opinion and Topic Salience

For most of us Brussels became again the capital of Belgium and home of the European commission. For a few days after the recent bombings Brussels was the most salient news topic. The cost of these terrorism acts will unfold in the months and years to come. More police, more controls, more time before we can travel etc.

I wanted to get a feeling for the size of such an act on public consciousnesses: how central a topic was it for people? There is no better place to start than google searches. So I looked for the intensity of Google search for a disjunction of  all possible ways to write Brussels in as many languages as I can put together. I used: “Brussels + Brussel + Brüssel + Bréissel + Bruxelles + bryssel + Bruselas + Βρυξέλλες +Brusel+ Bruxelas”. The results ought to capture salience by country as one might realise after a moments thought (some error is introduced by cross-language-country searches but ought to be negligible for our precision level requirements).

In all languages the searches are clearly related to the terror attacks and are most likely mean to locate news about the events. Here are the most prolific searches: “brussels attack”, “attentat bruxelles”, “brussels airport”, “brussels bombing”, “attentats bruxelles”, “where is brussels”, “atentado bruselas”,, “atentado en bruselas”, “attentato bruxelles”, “brüssel anschlag”.

The intensity of search by country radiates out from the epicenter (neighbors search more) with the exception of German speaking countries which are superseded by Spain and Italy:

Belgium     100
Luxembourg     48
Netherlands     21
France     21
Italy     20
Spain     20
Germany     17
Switzerland     17
Austria     15
Ireland     13

The macro picture which shows the long term deviation from the mean intensity world wide looks so:


The 7 day picture looks like below. Notice that the salience of a shock topic is extremely short lived:

7dayFinally comparing topic salience with two other more or less ubiquitously written words (“sport” and “Facebook”) shows that “Brussels” overshadowed “sport” but did not really challenge “facebook”. Also two days after the events “sport” became more salient than Brussels.

sport facebook


This entry was posted in Big Data, comment, nowcasting and tagged , , . Bookmark the permalink.