Data Bias Recognition in Museum Settings: Framework Development and Contributing Factors

Research Background and Problem

• What problems or challenges did the authors identify?

  1. Data bias cannot be completely eliminated, and the organization and presentation of data may influence people's judgment of information.
  2. In informal educational settings such as museums, visitors typically do not question the credibility of data, as the authority of museums may suppress critical thinking.
  3. Although data literacy is considered a crucial 21st-century skill, existing methods (e.g., the Inquiry Cycle framework) have limited adaptability to museum contexts due to constraints in resources, time, and guidance.

• Why is this problem important? Data is becoming increasingly important in people's lives. Enhancing the public's critical ability to reflect on data bias and credibility is key to fostering data literacy and reducing the spread of misinformation.

• Research Motivation and Related Work The study aims to investigate the factors influencing visitors' ability to identify data bias through the design of museum exhibitions and to promote critical thinking in informal educational environments. Existing research has explored data literacy in education, but studies on how to design effective interactive exhibits in museums to stimulate critical thinking remain relatively limited.

Solution

• What methods or solutions did the authors propose?

  1. The authors proposed a new "three-layer data bias identification framework" in the museum context to supplement the traditional Inquiry Cycle framework.
  2. The museum exhibition featured an interactive AR (Augmented Reality) data map called "Data Windows," enabling visitors to explore data bias through multiple visualization perspectives.
  3. Multiple data views were designed to help users shift from static, intuitive thinking to analytical and critical thinking.

• What is innovative about this solution? The three-layer framework explicitly outlines the cognitive processes visitors undergo when identifying data bias, providing new guidance for designing data presentations in non-traditional educational environments such as museums. Additionally, the use of augmented reality technology to layer data presentation enhances visitor engagement and critical thinking.

• What are the implementation steps and key technologies used?

  1. Data Selection and Processing: Open, potentially biased iNaturalist data was selected and cleaned to ensure accuracy and relevance.
  2. Interactive Design: A laser-cut city map was used as a visual anchor, combined with AR technology to provide three views ("by season," "by time," and "nearby observations") to showcase biodiversity data to visitors.
  3. User Testing and Iterative Optimization: Pilot studies and interviews with university students were conducted to refine navigation interfaces, icon designs, and information descriptions.
  4. Qualitative Data Analysis: Statements from 15 interviews were analyzed using coding and the theoretical framework (three-layer bias identification process) to assess participants' performance in identifying bias.

Research Findings

• What specific findings were achieved?

  1. The "three-layer data bias identification framework" was proposed, including:
     • Layer 1: Perception of contradictions and mismatched information based on prior knowledge.
     • Layer 2: Critical analysis of data and datasets.
     • Layer 3: Explicit identification of bias.
  2. Key factors influencing visitors' ability to identify data bias were identified (e.g., familiarity with the region).

• What advantages does it have compared to existing solutions?

  1. The framework is tailored to the specific context of museums, making it more suitable for short-duration, self-directed learning environments than the traditional Inquiry Cycle.
  2. AR technology is used to provide multi-layered information displays, helping visitors see the multifaceted nature of data and potential biases.

• What were the experimental or evaluation results?

  • Among 15 participants, 47% were able to identify bias, while 73.3% at least reached the second layer of the bias identification framework.
  • Familiarity with the region was shown to be a significant factor influencing bias identification.

• Limitations and Future Directions

  1. Limitations:
     • Small sample size, with participants primarily being highly educated university individuals, limiting representativeness.
     • The interview environment in the museum may have constrained participants' responses due to time or social expectations.
     • The definition of bias varies among individuals, affecting participants' responses to direct questions.
  2. Future Directions:
     • Explore specific barriers faced by groups unable to identify bias, including visitors' definitions of "bias."
     • Expand the study sample and test the framework's adaptability in more informal learning environments.
     • Conduct deeper evaluations of the specific role of AR technology in bias identification and learning.

Conclusion

This study investigates how to design interactive exhibits using augmented reality technology in a museum setting to help visitors identify data bias. By proposing the "three-layer data bias identification framework," the research demonstrates that familiarity with relevant background knowledge and strong data analysis experience are key influencing factors. These findings are not only practical for improving museum design but also significant for fostering public data literacy and critical thinking.