Resident Income And Expenditure Data Quality Multidimensional Evaluation Based On User Perspective

Resident income and expenditure data is the important basis of national policy decision,enterprise production decision,and resident life decision.Therefore,the data quality is of vital importance.However,the current research on resident income and expenditure data quality evaluation is not only limited,but also mainly about accuracy analysis and qualitative analysis.So it can be developed.In addition,statistics department internal self-evaluation is the main form of resident income and expenditure data quality evaluation in practice so that public is not involved and social credibility is poor.Therefore,as a kind of multidimensional evaluation methods,user satisfaction survey is adopted to evaluate resident income and expenditure data quality,in order to develop related research and improve social credibility.Firstly,the theoretical logic of user satisfaction survey is discussed.Secondly,in regard to the definition of statistical data quality by international organizations and Chinese scholars and the characteristics of resident income and expenditure data,resident income and expenditure data quality dimensions are determined and resident income and expenditure data quality evaluation index system is established.Thirdly,user satisfaction survey is conducted,and three methods of descriptive statistics,entropy weight fuzzy comprehensive evaluation and structural equation model are applied for quantitative empirical analysis.There are three results as below.(1)There is a significant gap between the level of user perception and expectation of resident income and expenditure data quality.Objectivity,integrity and applicability should be the key points to improve the data quality.(2)The level of user perception of resident income and expenditure data quality is only on average,whether for quality dimension,quality aspect,or the overall quality.In addition,applicability,integrity and objectivity should be the key points in the data quality improvement.(3)Except for the two quality dimensions of comparability and applicability,other quality dimensions all have a significant positive effect on user satisfaction.The effect of methodological soundness and objectivity,institutional normalization and availability is evident,and the impact factors of these quality dimensions are all over 0.5.Fourthly,based on the empirical analysis results,improvement suggestion is provided from three points of institutional environment,productive process and release process.In addition,some research prospects are put forward.