Adaptive Design of Dynamic Surveys and Data Quality Assessment
The massive influx of data from online surveys has significantly reduced the response rates for traditional sampling surveys.However,the data quality and inferential validity of web samples have been questioned.Adaptive sampling survey design has emerged as an effective means to improve data quality in online survey practice.This paper,based on a review and comparison of existing literature,systematically introduced the construction principles and estimation methods of the R statistic and the partial R statistic.Then the paper deeply explored their critical roles in adaptive survey design,and designed numerical simulations to demonstrate the application process of these statistics,thereby verifying their practical effectiveness in enhancing data collection efficiency and quality control.The results indicated that,when auxiliary variable information for the target sample was available,using the R statistic and the partial R statistic to guide the data collection process in adaptive surveys could significantly enhance the representativeness of the response sample and the accuracy of estimations,substantially improving the inferential validity of web samples and the efficiency of future sample integration.
NETL
NSTL
万方数据
