"Are we all in the same boat?" Customizable and Evolving Avatars to Improve Worker Engagement and Foster a Sense of Community in Online Crowd Work

Title of the Paper

"Are we all in the same boat?” Customizable and Evolving Avatars to Improve Worker Engagement and Foster a Sense of Community in Online Crowd Work.

Paper Information

• Research Domain: Human-Computer Interaction and Interaction Design for Crowdsourced Work
• Keywords: Crowdsourced microtasks, worker avatars, customization, group identity, engagement, community

Research Background and Problem Statement

• Problems or Challenges:

  • The decentralized and non-interactive nature of online crowdsourcing platforms makes it difficult for workers to develop a sense of group belonging.
  • The repetitive and monotonous nature of microtasks leads to a lack of motivation, burnout, and even task abandonment.
  • Some workers seek a sense of belonging through external forums, but this is not a universal choice or need for all workers.

• Significance:

  • Improving worker experience can enhance task quality and platform sustainability, addressing long-term issues like high attrition rates and low-quality work.

• Research Motivation and Related Work:

  • Studies show that digital avatars in gaming and educational environments can enhance users' self-identity and engagement.
  • Research on fostering a sense of belonging through task progress-driven avatar evolution and customization, combined with lightweight community spaces, remains scarce.

Proposed Solution

• Methods or Solutions:

  • Customizable and evolving worker avatars: Allow workers to personalize their avatars using an avatar editor and unlock additional features as they complete more tasks, creating a gamified experience.
  • Worker community space: Design a virtual community page displaying worker avatars and their emotional expressions post-task completion, exploring its potential to enhance group identity.

• Innovations:

  • Linking gamification elements to worker progress, pioneering research on the impact of avatar evolution on worker experience.
  • Proposing a lightweight community space as a substitute for group identity, without requiring complex social interactions.

• Implementation Steps and Technology:

  1. Develop task interfaces, avatar editor, and community pages using Vue.js and Flask.
  2. Design five experimental conditions, including a control group, basic avatars, evolving avatars, and their community variants.
  3. Recruit 680 workers to complete two types of microtasks: information search and credibility analysis.

Research Findings

• Specific Outcomes:

  • Customizable and evolving avatars significantly improved task retention rates, with a more pronounced effect in credibility analysis tasks.
  • Avatar evolution did not have an overall impact on workers' subjective engagement but enhanced the aesthetic appeal of the interface in credibility tasks.
  • The worker community space did not significantly foster group identity, though some workers felt connected through shared goals or emotional expressions.

• Advantages and Comparisons:

  • Compared to traditional task designs, the improved retention rates from evolving avatars make long-batch crowdsourcing tasks more cost-efficient.
  • Explored the impact of different task types on group identity and gamification interventions, revealing the moderating role of task attributes.

• Experimental and Evaluation Results:

  • Workers with lower group identity reported higher workload, while those with higher group identity demonstrated greater motivation and engagement.
  • High-identity workers tended to spend longer execution times on more complex and monotonous tasks.

• Limitations and Future Directions:

  • The lack of significant impact from the community space may be due to the absence of direct interaction among workers; future research could integrate real-time interaction designs.
  • Gamification designs may be influenced by novelty effects, and their long-term impact requires further validation.
  • Investigate cross-platform adaptability, examining gamification needs and effects across different crowdsourcing platforms.

These insights provide theoretical and practical references for designing sustainable crowdsourcing tasks and improving worker experiences, while also pointing to directions for further optimization and research.