When algorithms create personas from social media data, the personas can become noxious via automatically including toxic comments.
To investigate how users perceive such personas, we conducted a 2 × 2 user experiment with 496 participants that showed participants toxic and non-toxic versions of data-driven personas.
We found that participants gave higher credibility, likability, empathy, similarity, and willingness-to-use scores to non-toxic personas. Also, gender
affected toxicity perceptions in that female toxic data-driven personas scored lower in likability, empathy, and similarity than their male counterparts. Female participants gave higher perceptions scores to non-toxic personas and lower scores to toxic personas than male participants.
This study offers three main takeaways:
- An increase in toxicity in a persona’s comments results in a decrease in all the measured persona perceptions.
- The increase in toxicity affected likability, similarity, and empathy the most.
- The increase in toxicity affected willingness to use and credibility to a lesser degree.
We find it logical that the affective dimensions of persona perception (likability, similarity, and empathy) decreased the most with the toxic quotes since these perceptions deal with how warmly a person thinks about a persona.
For full set of implications from our research for designing data-driven personas, read the complete research article.
Salminen, J., Jung, S. G., Santos, J. M., and Jansen, B. J. (2021) Toxic Text in Personas: An Experiment on User Perceptions. AIS Transactions on Human-Computer Interaction. 13(4), Paper 4.