Citation

Wang J, Chen J, Lan F, Liu Q, Xiao L. Int. J. Veh. Safety 2020; 11(4): 304-314.

Copyright

(Copyright © 2020, Inderscience Publishers)

DOI

10.1504/IJVS.2020.111522

PMID

unavailable

Abstract

In the in-depth investigation of traffic accident stratified sampling, when the allocation of sample size needs to consider multiple indices, the traditional method treats each index equally, or uses the variance of the indices to evaluate the weights, which ignores the subjective preference of the researchers. In addition, there may be some cases where the variance difference of some important indices is small or missing, resulting in the sample size allocation failing to achieve the desired goal. In view of this situation, Analytic Hierarchy Process (AHP)-Coefficient of Variation (CV) method is adopted as a compromise scheme to comprehensively evaluate the weight of each indicator from both subjective and objective aspects. Then, according to the proportional distribution method, the samples of each index of each layer are determined. Finally, the sample size of each layer is distributed by weighted average method to maximise sampling accuracy.

Keywords: traffic accident investigation; stratified sampling; distribution method; weight; analytic hierarchy process; coefficient of variation method.

Language: en