aTag: Stats, A User Interface to Warn of Statistically Fallacious Claims in the News
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Background : Weak statistical literacy and biases in human cognition (e.g., denominator neglect) impede the citizen’s abilities to recognize statistically fallacious claims in the news. Such claims can be fact-checked by human experts, but it takes at least a few hours during which news readers can be exposed to less-than-valid claims. Methods: As an interim solution, this study proposes aTag:stats, a tag attached to mobile news articles to point out claims that are based on incomplete statistical data. The aTag:stats visualizes the data–claim structure of an argument, while providing rebuttal comments on the omission in the statistical data. The effects of aTag:stats were tested using an online questionnaire, with two mock economy news articles: Car Insurance and Greenhouse gas (GHG). Participants responded to the same questions twice, with the control condition (news text only) and with the experimental condition (news text with aTag:stats) on what they can infer from the given information and how they perceive the validity of claims. Later, they rated the perceived information sufficiency of the news text and aTag:stats. Results: In both Car Insurance and GHG data collected with the experimental condition, the inference scores were higher, the perceived validity ratings were lower, while the information sufficiency of aTag:stats were rated higher than that of the news text. The aTag:stats, by translating word problems to numerical expressions of fractions and equations, brought the participants’ attention to the absence of denominator and base data, while reducing the participants’ cognitive load. Conclusions: We concluded that the aTag:stats presented the participants opportunities to critically evaluate the statistical claims in the news and informed them on the incompleteness of data. In the future studies, statistical fallacies where the argument structure can be better explained and exposed in formats other than fractions will be experimented.
Abstract
Background : Weak statistical literacy and biases in human cognition (e.g., denominator neglect) impede the citizen’s abilities to recognize statistically fallacious claims in the news. Such claims can be fact-checked by human experts, but it takes at least a few hours during which news readers can be exposed to less-than-valid claims. Methods: As an interim solution, this study proposes aTag:stats, a tag attached to mobile news articles to point out claims that are based on incomplete statistical data. The aTag:stats visualizes the data–claim structure of an argument, while providing rebuttal comments on the omission in the statistical data. The effects of aTag:stats were tested using an online questionnaire, with two mock economy news articles: Car Insurance and Greenhouse gas (GHG). Participants responded to the same questions twice, with the control condition (news text only) and with the experimental condition (news text with aTag:stats) on what they can infer from the given information and how they perceive the validity of claims. Later, they rated the perceived information sufficiency of the news text and aTag:stats. Results: In both Car Insurance and GHG data collected with the experimental condition, the inference scores were higher, the perceived validity ratings were lower, while the information sufficiency of aTag:stats were rated higher than that of the news text. The aTag:stats, by translating word problems to numerical expressions of fractions and equations, brought the participants’ attention to the absence of denominator and base data, while reducing the participants’ cognitive load. Conclusions: We concluded that the aTag:stats presented the participants opportunities to critically evaluate the statistical claims in the news and informed them on the incompleteness of data. In the future studies, statistical fallacies where the argument structure can be better explained and exposed in formats other than fractions will be experimented.
참고문헌 (23)
- Chinn, C. A., "Evolution Challenges: Integrating Research and Practice in Teaching and Learning about Evolution" (2012), Oxford University Press
- Daroczy, G., "Word problems: A review of linguistic and numerical factors contributing to their difficulty", Frontiers in Psychology, 6(APR), pp.1~13 (2015)[DOI]
- Hahn, Y., "aTag:half_truth, a user interface to warn of the fallacy in the claims reported in the news", The Proceedings of the IASDR2021 (2021)
- Hoffmann, M. H. G., "Stimulating Reflection and Self-correcting Reasoning Through Argument Mapping: Three Approaches", Topoi, 37(1), pp.185~199 (2018)
- Kahan, D. M., "Ideology, motivated reasoning, and cognitive reflection", Judgment and Decision Making, 8(4), pp.407~424 (2013)
- Klein, T., "An Indicator for Statistical Literacy based on National Newspaper Archives", Promoting understanding of statistics about society. Proceedings on the Round Table Conference of the International Association of Statistics Education (IASE), pp.1~12 (2016), ISI[DOI]
- "2020 Media Consumer Report" (2020)[URL]
- "More than half of Korea’s Labor force has received a higher education" (2021)[URL]
- "Free or discount to Chinese on health insurance premiums paid by Korean?" (2020)[URL]
- Mercier, H., "Why do humans reason? Arguments for an argumentative theory", Behavioral and Brain Sciences, 34(2), pp.57~74 (2011)
- "Addressing Hoaxes and Fake News" (2016)[URL]
- Nyhan, B., "When Corrections Fail : The Persistence of Political Misperceptions", Political Behavior, 32(2), pp.303~330 (2010)
- "Government raises greenhouse gas reduction target by 26.3% → 40% in 2030... Twice the average annual reduction rate of advanced countries" (2021)[URL]
- "Is the government’s 2030 GHG reduction target excessively high?" (2021)[URL]
- Reyna, V. F., "Numeracy, ratio bias, and denominator neglect in judgments of risk and probability", Learning and Individual Differences, 18(1), pp.89~107 (2008)
- Sbarski, P., "Visualizing argument structure", Lecture Notes in Computer Science, 5358, pp.129~138 (2008)
- Stenning, K., "A cognitive theory of graphical and linguistic reasoning : Logic and implementation", Cognitive Science, 19(1), pp.97~140 (1995)[DOI]
- Thawani, A., "Representing Numbers in NLP: a Survey and a Vision", Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 5, pp.644~656 (2021)
- Toulmin, S., "The Uses of Argument" (2003), Cambridge University Press
- van Amelsvoort, M., "Show me your opinion. Perceptual cues increating and reading argument diagrams", Instructional Science, 44(4), pp.335~357 (2016)
- Vraga, E. K., "Testing Logic-based and Humor-based Corrections for Science, Health, and Political Misinformation on Social Media", Journal of Broadcasting and Electronic Media, 63(3), pp.393~414 (2019)
- Wang, L., "Template-Based Math Word Problem Solvers with Recursive Neural Networks", Proceedings of the AAAI Conference on Artificial Intelligence, 33(01), pp.7144~7151 (2019)