The next session at the 5th Data Donation Symposium at the Weizenbaum-Institut in Berlin starts with Jessica Daikeler, whose interest is in data donation studies focussing on WhatsApp. It remains unclear what drives people’s willingness to donate data to such studies, however, and this study tests these patterns. Who donates very directly influences the results of such studies, of course: data donors represent only a very small subset of the overall population, and are usually not very representative of it.
Such participation bias begins with the consent process: this introduces frictions, and loses participants in the process. Further, those who are willing to donate data from a platform like WhatsApp differ systematically from those who do not; what what they are asked to donate further shapes what a research project can observe.
This study engaged in two experiments: first, it asked survey participants whether hypothetically they would be willing to donate one of their WhatsApp chats; second, it actually stepped survey participants through the WhatsApp data donation process. In the first study, motivational framing, data content, and privacy frictions were all tested as potential factors affecting data donation willingness.
Only about 11% of participants were willing to donate; societal motivational appeals, and emphasis on aggregate data, and lower consent requirements were all stronger motivators. In the second study, while some 25% indicated their willingness to donate, only around 11% ended up donating actual data. Specific design factors in both studies might also have played a role in affecting these results, however.
