Level: Intermediate
Many cases of questionnaire surveys end at the consideration stage, and although there are some cases where proposals are made based on the considerations, the proposals in those cases are merely hypothesis ideas (presenting ideas) and rarely go as far as hypothesis verification (verification experiments). On the other hand, many cases of conjoint analysis, which is a questionnaire experiment, have been published, but most of them are standalone approaches specialized for conjoint analysis. And from the perspective of the engineering application of experimental design, those conjoint analyses have the following three problems: 1) quantitative factors are treated qualitatively, 2) product terms (interactions) and quadratic terms are ignored, and 3) analysis and design are performed by simple condition determination based on variance analysis. These three points are mutually intertwined, and the following three are effective methods to comprehensively solve them: [1] Adopt an optimal design (custom design). [2] Adopt variable selection for multiple regression analysis with the data in [1]. [3] Implement mathematical programming based on the estimated multiple regression equation in [2]. This study proposes a method that combines questionnaire surveys and questionnaire experiments and overcomes the problems of questionnaire experiments.
For nearly 50 years, he has been researching Total Quality Management (TQM), Statistical Quality Management (SQM), and design theory. Since the beginning of the 21st century, he has proposed a new design paradigm called Hyper Design, developed its mathematical theory, HOPE theory (Hyper Optimization for Prospective Engineering), and is co-developing its support software, HOPE-Add-in, with SAS. Through this effort, he has realized a new design method that combines the concept of Hyper Design, the statistical theory of HOPE theory, and the support tool, HOPE-Add-in for JMP. As a social scientific extension of this theory, he has proposed selective two-sided causal analysis (multi-group principal component regression analysis as obverse causal analysis + factor regression analysis as reverse causal analysis), and is currently proposing reverse causal analysis (a causal analysis method equivalent to quasi-SEM), which is an advanced approach following selective two-sided causal analysis.
Go to the collection site
https://community.jmp.com/t5/Discovery-Summit-Japan-2022/Discovery-Summit-Japan-2022-%E8%AB%96%E6%96%87%E9%9B%86/ta-p/568396
*The video will be available until Wednesday, November 30th.
The publication has ended.
