| The increasingly prominent water problem has become one of the important challenges restricting high-quality economic and social development and meeting people’s needs for a better life.Knowing and understanding water issues,especially water knowledge,is the core element and important basis for solving water-related issues,and is an important part of water activity development,water environmental protection and water literacy development.Therefore,mastering the necessary water knowledge is crucial to the sustainable use of water resources and the improvement of citizens’ water literacy.This research focuses on the measurement of citizens’ water knowledge stock and incremental dynamic and proves the measurement indicators of citizens’ water knowledge stock.Understanding and grasping the awareness and growth law of citizens’ water knowledge can provide a certain theoretical basis and decision-making reference for relevant departments to evaluate the effectiveness of water knowledge popularization and publicity work and formulate more targeted popularization and publicity priorities and strategies.This study proposes a new scientific measurement method by reviewing basic theories such as knowledge measurement methodology,knowledge base theory,knowledge growth theory,etc.,as well as sorting out the influencing factors of water knowledge stock and related measurement literature.Based on the measurement method,a more objective measurement model is constructed,and the measurement of the water knowledge stock and its dynamic increment is truly and effectively completed.This provides new ideas and methods for reference in the field of knowledge measurement and expands the method system in the field of knowledge measurement.The main research contents and conclusions are as follows:(1)Based on the grounded theory and event system theory,a grounded-system analysis method is proposed,and an exploratory study on the measurement index of water knowledge stock is carried out.Based on the knowledge-based theory,based on the collected public information and semi-structured interviews and other research materials,and using the method of grounded-system theory,4 secondary indicators,10 tertiary indicators,and 43 basic measurement indicators are summarized.The key category chain is constructed by retrospectively tracing the interview data,and its influence path and degree are analyzed,to construct the indicator representation model of citizens’ water knowledge stock measurement.In addition,the formation logic of the water knowledge stock index system is analyzed,which lays a theoretical foundation for the subsequent evaluation of water knowledge stock measurement indicators.(2)The research adopts the multi-level item response theory to screen and evaluate the basic measurement indicators of water knowledge stock and selects a basic measurement indicator system for measuring water knowledge stock,which is convenient to accurately measure the respondents with different water knowledge levels.Based on the obtained survey data,marginal maximum likelihood estimation was used to estimate the project parameters of 43 indicators,and the unreasonable indicators were eliminated by combining the results of the model-data fitting indicator characteristic function.According to the feedback results of the indicator information function and the test information function,the relationship between the tested indicators and the characteristics of the water knowledge stock level of the respondents is verified.It can be seen from the discrimination degree of the indicators and the display results of the characteristic curve that three indicators that are not suitable for the basic measurement of water knowledge stock have been deleted.From the results of the indicator information function and the measurement error function,the scale composed of the remaining measurement indicators is universal and reliable.(3)The research proposes a measurement method based on fuzzy identificationBayesian network,which is based on the basic index system of water knowledge stock measurement to scientifically measure citizens’ water knowledge stock.Fuzzy Recognition-Bayesian Network is a new knowledge measurement method proposed by the fusion of fuzzy set theory and Bayesian network model.Probabilistic reasoning is performed on the test sample set to calculate the probability value of the citizen’s water knowledge stock and convert it into a percentile measurement value.It can be seen from the results that 60.38% of the respondents have water knowledge above the level of "understanding” and can master and identify most of the key points of water knowledge,reaching the intermediate level or above.There are 164 samples whose measurement value is less than 60 and greater than 46.48,and the corresponding respondents’ water knowledge stock level is "not very clear",and their understanding of the relevant knowledge points of the water knowledge stock measurement index is relatively weak.The sample size with a measurement value of less than 46.48 is 23,indicating that these respondents basically did not master water knowledge points,and did not know or have never touched most of the knowledge points at all,resulting in a low overall water knowledge stock level.Judging from the measurement results,it conforms to the reality of citizens’ water knowledge level and confirms the feasibility of the measurement method.(4)Based on knowledge absorptive capacity and knowledge growth theory,the basic measurement index of original water knowledge stock is updated and optimized to realize the measurement of water knowledge increment.Compared with the basic measurement indicators in Chapter 4,a total of 7 indicators have been revised,16 indicators have been added,and 3 indicators have been deleted over time,resulting in54 basic measurement indicators of water knowledge increment.Then,through the selection of the measurement index optimization model based on the information contribution rate,a total of 42 basic measurement indicators including “geographical location of drinking water source and water supply area” were retained,and a total of9 basic measurement indicators including “water meter functions and reading data”were removed.The research results provide important theoretical support for the realization of scientific water knowledge increment measurement.The dynamic rough water knowledge increment algorithm based on the attribute probability set value constructs a dynamic measurement model of water knowledge increment and uses the measurement data of two time points in 2019 and 2021 to dynamically measure the water knowledge increment of citizens.The results show that in the decision attribute equivalence class attribute values 3(general),4(basically possessing water knowledge),and 5(having water knowledge),the corresponding cardinality increases.The cardinality of attribute values corresponding to 1(no water knowledge)and 2(basically no water knowledge)is significantly reduced.This shows that citizens’ water knowledge has undergone dynamic changes in the corresponding condition attributes,and the decision-making attributes with above-average water knowledge level have increased significantly,realizing the dynamic measurement of citizens’ water knowledge increment,and verifying the feasibility of the method. |
