The Experiential Assessment Of Chinese Provincial Population Data Quality

The scientificalness of decision-making depends largely on the quality of statistical data. From the macroscopic level, statistical data is the foundation of states’ macro decision-making; From the micro level, the enterprises’ correct decision-makings also can’t do without the high-quality statistical data. Therefore, the statistical data quality has received increasing attention from the society and been important research subject. Among it, because of the special effect of the demographic data in the states’ macro-control, the quality of demographic data has received increasing attention from the society. However, the complicacy of the demographic data has also lead to the query of demographic data quality. So, the quality assessment of demographic data is relatively meaningful. And this paper’s assessment criteria of statistical data is accuracy.Based on the summary of demographic data quality’s assessment methods, this paper applies the method that based on the evaluation index itself’s law (which can be called individual index evaluation method) and the method that based on related Indexs (which can be called correlation index evaluation method) in our country’s provincial population data quality evaluation research. That is the paper attempts to use BENFORD law and spatial panel data model to evaluate the provincial population data quality. The BENFORD law is used to inspect the accuracy of Chinese31provinces’ population data. The analysis result shows that, under the significance level of0.05, the first and the second number’s distribution is consistent with the expected distribution of BENFORD law. That is to say there is no abnormal condition in Chinese provincial population data and the data quality meets the requirements. The paper uses spatial panel data model to analyze Chinese provincial population data from2007to2011, the result shows that except for the4high error rate provinces in China, the remaining27provinces population statistics data’s error rate is low.BENFORD law and spatial panel data model obtains the consistent result, which fully illustrates that Chinese provincial population data has high data quality. Combines BENFORD law and spatial panel data model when assessing Chinese provincial population data quality can avoid the the problem that the assessment method is too simplex and the problem that the evaluation criteria is too subjective to lead to the ambiguous results.