Expressions of emotions and social behavior around COVID19 on Twitter in Germany

Methods

• We analysed 89 million tweets from users located in Germany by Crimson Hexagon. Only original tweets (no retweets) in German and from users with a number of followers between 100 and 100.000 (to remove spam and media) were included.
• For anxiety, text is matched for anxiety terms from the LIWC lexicon. LIWC is a standard methodology in psychology for text analysis that includes validated lexica in German.
• Similarly, tweet text was matched to a list of prosocial terms used in previous research, including for example words that describe ways of helping, supporting, cooperating, sharing, volunteering, or donating. We excluded words referring to health and healing (e.g. hospitalize), as wall as the word sharing (“teilen”), to avoid confounds with non-prosocial tweets about Covid-19 and health or about sharing of links/tweets/videos, etc, respectively.
• Marked dates during the Covid-19 outbreak are taken from Wikipedia.

Anxiety

In both figures below, the lines indicate the following events:

• Red:
  • Terrorist attack in Christchurch on 2019-03-15
  • Terrorist attack in Halle on 2019-10-09
  • Terrorist attack in Hanau on 2020-02-25
• Yellow: Christmas
• Green: First Covid-19 case on 2020-01-27 in Bavaria
• Grey: Covid-19 terms and anxiety terms in tweets become more frequent from 2020-02-26 onwards, but this does not seem linked to one specific event (or the event still remains to be identified).
• Orange: First death from Covid-19 on 2020-03-09
• Turquois: Social distancing measures and emergency state in effect in Bavaria on 2020-03-16, later on the same day also announced for entire Germany

Percent of tweets containing anxiety-related terms

Percent change in anxiety-related tweets relative to the average level in 2019

The figure below shows the difference to the average amount of anxiety-related tweets in 2019 (i.e. the baseline) on a logarithmic scale. Taking into account the baseline level makes emotion levels comparable across countries, and shows how much emotions change from their usual level expressed in percent. A value of zero corresponds to the average level in 2019, and everything above or below to an increase or decrease in comparison to this level.

Observations for anxiety

• A typical day in Germany shows a frequency of anxiety terms in Twitter of about 2.7%.
• The highest values in the last year are related to terrorist attacks, with up to 5% of tweets containing anxiety related words (a doubling of the normal frequency). However, these strong increases in anxiety-related tweets never lasted long, but quickly returned back to the previous level.
• Although the increases by about 40% (compared to the 2019 average) related to Covid-19 are less strong, they are still quite high compared to other countries.
• Two of the three peaks of anxiety-related tweets since the outbreak clearly coincide with the first death and the onset of social distancing measures. The first peak on 2020-02-26 (after the grey doted line) does not seem linked to one specific event.

Prosocial behavior since the outbreak of Covid-19

In most countries in which expressions of anxiety have risen since the outbreak of Covid-19, expressions of empathy, willingness to support each other and cooperate (in short, prosocial behavior) have increased as well. The increase begins at different moments in different countries, probably depending on which information people had about Covid-19 in other countries, and on which measures were communicated in what way by their governments.

In the plot below, the lines indicate the following events:

• Red: Terrorist attack in Hanau on 2020-02-25
• Green: First Covid-19 case on 2020-01-27 in Bavaria
• Orange: First death from Covid-19 on 2020-03-09
• Turquois: Social distancing measures and emergency state in effect in Bavaria on 2020-03-16, later on the same day also announced for entire Germany
• Grey: 26 February 2020: Since the outbreak of Covid-19, the day on which levels of prosociality in tweets increase above the average level from 2019 and remain high for several days in a row.

Percent change in prosocial tweets relative to the average level in 2019

In the figure below, a value of zero corresponds to the average level in 2019, and everything above or below to an increase or decrease in comparison to this level. The figure on the left displays the evolution over time since 1 January 2020. The small figure on the right shows the mean difference from the 2019 baseline before and after the day since which the level of prosocial terms on Twitter lies and remains above the baseline.

Observations for prosocial behavior

• Levels of tweets referring to helping, sharing, empathy and similar behaviors or feelings have started to increase since February 26, together with anxiety-related tweets.
• In comparison to the average from 2019, they have increased by 5% on average.
• Prosocial tweets about refugees being stuck at the Turkish border may contribute to the first peak observed around 2 March.
• A later and higher peak starting on 14 March is clearly linked to the onset of first social distancing measures, and lies about 9% above the baseline.

Conclusions

• Interestingly, events around Covid-19 have not yet led to abrupt peaks of anxiety terms as observed in other countries. Still, one can observe an increase from the beginning of March on.
• Expressions of prosocial behavior (helping, sharing, cooperating, …) have increased as well. Since the outbreak of Covid-19 and the implementation of social distancing measures, prosociality expressed in tweets has reached the same level as after a local terrorist attack in February.
• Our results suggest that people in Germany are not only more afraid, but also clearly more willing to support others in need.

Back to main page