Article ID: 95.23211
In qualitative data analysis, comprehensive knowledge collection is considered important, and the degree of saturation serves as one of the indicators. The Zipf distribution method can be used for the quantitative estimation of the degree of saturation in question-and-answer format qualitative data. However, this method may overestimate the degree of saturation. To address this issue, this paper proposes a new approach using the approximate Bayesian computation (ABC) method, which reduces the bias in saturation estimation. Furthermore, through simulations, we compared the bias between this new estimation method and the existing one, demonstrating that the estimation using the ABC method has a higher accuracy. Moreover, we compared the differences and similarities between the two methods through case studies in practical operational settings.
