Influence of Scientific Methodology on the Public’s Willingness to Participate in Citizen Science: A Cross Comparison between Theoretical Deduction and Experimental Intervention

科学方法论对公众科学参与意愿的影响

Abstract

Scientific methodology is an architecture to guide scientific research activities. This paper attempts to explore the influence of scientific methodology on public willingness to participate in citizen science. The methodology of the citizen science project was divided in two dimensions: theoretical deduction and experimental intervention, and the research methods under the two dimensions are summarized as induction and deduction, as well as exploration and exploitation, forming a 2X2 research methodology system. 1,970 citizen science projects were adopted as research corpus, and text mining is employed to classify the narratives. We expanded the keyword dictionary regarding the research methods, resulting in a list of 481 keywords for citizen science research methods, and conduct corpus training to expand keywords based on text similarity. Then we use the constructed keyword dictionary as the corpus for classifying texts. Using stem words and word frequency for text classification. The econometric models with the number of participants and the amount of financing as dependent variables are constructed to estimate the effectiveness of the methodology. The results show that induction method accounts for the largest proportion (33%), while deduction method accounts for the smallest proportion (20%). However, from the effect of methodology, deduction and exploration have significant advantages over induction and exploitation, respectively. From the comparison of the 2X2 quadrant, the deduction has significant advantages over the other three types of methods. Further, researcher identities, disciplinary heterogeneity, funding goals and social engagement play a moderating role. That is, non-professional researchers should focus on the feasibility and scientificity of the methods used in the project, while professional researchers may not have to overly focus on the impact of research methods on project participation. The results indicate that the effectiveness of research methods is mainly caused by graduate students and amateur researchers, and there is no significant correlation with the research methods of university professors and industrial researchers. Relatively speaking, university professors and industry researchers have richer background knowledge, more experience and expertise in matching appropriate research methods for research problems, and stronger professional abilities than the graduate students and amateur researchers. As initiators, professional researchers provide a sense of security to the public participating and are not easily affected by the choice of research methods. Therefore, university professors and industry researchers should not focus primarily on tuning research methods. Also, the project launchers should focus on project attributes. Natural science and high financing target projects should focus on the appropriateness and accuracy of scientific research methods. Furthermore, extensive social participation provides a wider space for information dissemination and resources, easily attracting project enthusiasts. We believe that social participation serves as a "catalyst" between research methods and public participation, and project initiators can gain more public support by increasing social participation. From the comparison of research methods among disciplinary differences, the deductive-exploratory research method has the greatest impact on the willingness of the public to participate under the influence of the social influence of the project initiator. This paper expands citizen science research methods and performance evaluation theory, providing practical reference for citizen science launchers to formulate effective research programs.