Contents
Goals
- Learning how amplification algorithms work
- Understanding how users can influence the algorithmic curation of content
- Understanding the unequal competition between socio-politically important information and other information on social media platforms
Steps
Preparation
Facilitators prepare a slide or feed with typical social media contents. These posts should be mixed, just as they are on real platforms – for example, about three posts from the areas of celebrity, sports, politics, and music, shown in random order.
- Download the templates and prepare. Download as ZIP file
If participants should vote by cards, the card set should be printed out for each participant (template-amplify-love-anger-1.pdf in the ZIP folder). Otherwise, it can be printed out once for the person leading the activity. This person can then hold up the corresponding emoji that is currently being voted on.
Each participant receives a voting sheet (in the open document text format .odt in the ZIP folder).
One facilitator should prepare the documentation and calculation sheet and use it during the session (in the open document spreadsheet format .ods in the ZIP folder).
Introduction (10 minutes)
We will now simulate how we consume our daily news on social media. But we don't just consume, because social media is a lot about responding too. That's why you should now respond to the content shown by the facilitator. We have the following options:
Either you point to the cards in front of each participant, or you present them in a presentation.
Reaction Types
Journey through the media stream (20-40 minutes)
Now, different posts are presented to which the participants should react.
The facilitator asks for reaction.
- Participants hold up cards: Each participant receives cards with the following icons, which they hold up for voting.
- Or: raise hands: Each participant raises hand when facilitators asks for the votes for a reaction type.
The facilitator notes the scoring on the documentation and calculation sheet (in the open document spreadsheet format .ods in the ZIP folder): If 2 participants reacted with a heart, then a “2” is put in the respective field,....
Each participant crosses the field in their chart (page 2 of material) too. If one reacts with a heart, crossing the “heart” field:
Template for participants
Comment: The amplification factors below the table are left blank during this step! They play a role during the next step - sum up.
Post by post the facilitators guide through the news stream, always in the same order: 1.) Facilitator shows posting 2.) Each participant crosses their reaction in their template. 3.) The facilitators puts the reactions in the spreadsheet protocol ...
Sum up (15 minutes)
At the end the reactions are summed up by participants.
Therefore, they receive the following information: Like in real social media some reactions have more value than others. Here is the table with the factors. A heart weights 4 times of a thumb up. No reaction has no value.
Amplification Factors:
Participants are invited to:
- Note the amplification factor for each column under the table (row “Factor”)
- Then they could calculate for each column: Add up all the crosses in a column and multiply by the factor.
In example: 2💡x factor 2 = 4 - The results inform them on their most frequent style of reaction.
Result Presentation
The co-facilitator sums up columns and rows and shares the results: The highest ranking card is: _____ The highest ranking emotion is: _____
Present the score to the participants, explain what the row and column results mean: Row results: inform how diverse the reactions of a post was. The overall score says, how successful it was. Column result: shows eventually dominant reaction patterns of the group independent of specific posts – a general more positive or negative word view, preference for reactions...
Reflection
- How to react on postings? Is no reaction an option? What consequences have positive and negatiove responses?
- Did you feel, that your reactions are treated fairly? If not, please explain.
- Did your feelings change when you realised that there are different amplification factors?
Transfer: Amplification & datafication (10 minutes)
- Amplification is a common technology in order to decide about what content is shown, in what row, context and to what audience. It is based on the inputs of users – which then co-determine what goes “viral” and what will drown in the deep ocean of content soon.
- Users are also objects of datafication. Since their response data is stored a platform can also conclude on their favourite way of response and favourite content – psychologic and intellectual conclusions. In this sense, what we do shows who we are.
- Amplification and datafication of the self have an impact on the society when something like that what was experienced during this task is used in large and if even the different data of our selves can be merged.
- Your platforms of choice: EU data from 2024 suggest that the most important platforms for sociopolitical information for youth are
Instagram (47 %),
TikTok (39 %) and
YouTube (37 % ).
How do you think amplification works on these platforms? How can you find this out?
Explanatory Data
The following empirical data may support reflection and transfer. For more information, you can also consult the freely available sources:
Transfer: Questions with a socio-political focus (20 minutes)
Platform interests: Why do platforms rank different kinds of content? Find examples of technical, economic, or political reasons.
Hate speech: Why do platforms find it difficult to block or delete hate messages, or to restrict the authors of such messages?
Political content: Which political content is particularly emphasised? Young people assume that they will come across important news anyway – because it will be shown to them on social media. Which important news do you think might escape their attention this way?
Example: low-emotion information about an important upcoming legislative change regarding compulsory schooling, compulsory military service… Independent media: What must serious and well-researched authors do so that their stories are read? What must they do in order to make a living from their work on social media?
Governance: What conclusions should we draw for the governance of social media? What can you personally contribute to a less polarised discourse?
References
European Union (2024). European Parliament Eurobarometer. Youth Survey 2024. FL013EP DGCOMM https://europa.eu/eurobarometer/surveys/detail/3392, accessed 24/03/2025
Eurostat: Evaluating data, information and digital content (2021 onwards)” Online data code: isoc_sk_edic_i21 https://doi.org/10.2908/ISOC_SK_EDIC_I21
Gagrčin, E., Schaetz, N., Rakowski, N., Toth, R., Renz, A., Vladova, G., & Emmer, M. (2021). We and AI - Living in a Datafied World: Experiences & Attitudes of Young Europeans. Weizenbaum Institute; Goethe-Institut. https://doi.org/10.34669/WI/1
Handbook: More than Go with the Flow
- A handbook on Digital Citizenship Education, created in the frame of the project DIYW-ROAD/Competendo. Digital Youth Work - rights-sensitive, open, accessible, democratic.
- Unless otherwise stated, authors and editors of the methods published in the project are Elisa Rapetti, Markus José Plasencia Kanzler, and Nils-Eyk Zimmermann
