Research On Clues And Criteria Of Perceived Quality Of Agricultural Scientific Data

The continuous development of the open science movement and data-intensive scientific research paradigm has significantly promoted the development of scientific data on a global scale,and more and more scientific and technological innovation forces have benefited from the sharing of scientific data.However,with the rapid growth of scale and demand,the insufficient quality of scientific data has gradually become prominent in many disciplines such as agricultural science,and has increasingly become an important constraint on the effective reuse and value of scientific data.This phenomenon puts forward the requirements for improving and strengthening the quality control of scientific data in practice,and theoretically reflects the current situation of people’s insufficient understanding of the process and mechanism of scientific data quality perception.Based on this background,the thesis focuses on the quality perception process of agricultural scientific data reusers,and uses empirical research methods to explore and determine the key cognitive elements such as quality perception clues and criteria of users and their interactions,in order to enhance the understanding of the quality perception process and mechanism and provide a theoretical basis for the development and improvement of agricultural scientific data quality control mechanisms.In specific research,the thesis constructs an agricultural scientific data quality perception model based on information processing theory and lens model,and designs observational context experiments based on this.The study comprehensively uses qualitative and quantitative methods such as eye tracking,semi-structured interviews,and questionnaires to triangulate the scientific data quality evaluation process of 82 Ph.D and master students selected from multiple institutes of the Chinese Academy of Agricultural Sciences.By analyzing the gaze point,eye trajectory and questionnaire data in eye movement observations,and using grounded theory methods to code and analyze interview data,the paper identifies quality perception clues and criteria of users,and determines the statistical relationship between clues and criteria.According to the data source,this paper divides the data into two types:article supporting data and online shared data,and discovers the differences in the quality perception process when users face the two types of data.The main research results of this paper are:(1)a agricultural scientific data quality perception model is preliminary proposed,which describes the "clues-criteria-dimensions-quality perception" multi-stage cognitive process of agricultural scientific data quality perception,and the "clues-criteria" cognitive processing link in the model is empirically demonstrated;(2)24 clues of agricultural scientific data quality perception are identified and divided into three types: data content clues,data context clues,and data source clues,and based on the perspectives of different disciplines,this article describes in detail the characteristics of perception clues of subjects in agricultural biology,agro-ecological environment and agricultural economics;(3)10 criteria for quality perception are determined,and are divided into the data inherent quality dimension and the data utility quality dimension based on the objective attributes and value/use of the data.The inherent quality of data in the framework includes criteria such as timeliness,completeness,accuracy,standardization,clarity,authority,and understandability,and the quality of data utility includes criteria such as usefulness,academicity,and novelty;(4)two types of interactions between clues and criteria are preliminarily determined: "single clue activates single criterion" and "single clue activates multiple criteria";(5)the quality perception characteristics of the subjects are summarized when they faced the article supporting data and the online shared data.For the former,the subjects pay more attention to the contextual clues and data source clues when perceiving the quality.For the latter,the subjects pay more attention to the data content clues,and the total number of clues and the number of criteria used by the subjects concerned about the former were more than the latter.