We agree with Bayesian statisticians, who are becoming the mainstream in statistics, that the traditional method of data analysis, i.e., Null Hypothesis Significance Testing (NHST), contains many statistical and logical defects, and that we should, therefore, employ Bayesian statistics in our data analyses, which is statistically and logically sound. One of the difficulties in employing Bayesian Data Analysis used to be that the posterior probability distribution cannot be analytically calculated in most of the cases. But thanks to the discovery of Markov Chain Monte Carlo methods (MCMC) and the rapid progress of computer power, good approximation of posterior distribution can now easily be obtained, even for ordinary social scientists. We employ the Bayesian Data Analysis methods in the analyses on our data from the experimental research, which uses four versions of scenarios and gathers data through internet questionnaire survey (N = 4 versions x 200 samples = 800 samples). The scenarios are based upon a Supreme Court Case, where an elderly man with senile dementia wandered into railroad track, was hit by a train, and died. The accident caused a delay in railroad operation and damages of about JPY 7 million (ca. US$70,000) to the railroad company (the plaintiff). The family members of the elderly man (the defendants) had been taking care of him and were sued by the railroad company for the compensation. The Supreme Court ruling concluded that the care-taking family members in this particular case were not liable to compensate the damages, but suggested as a general principle that the care-takers of elderly persons with senile dementia might be liable for accident damages caused by such old person if the care-taking activities could be evaluated as assumption of such risks. Many news papers covered the ruling, expressing some concerns that it might give disincentives for family members to take good and intensive care of their old parents and that people taking care of elderly person with dementia may be forced to take such drastic preventive measures as tying an elderly person to the bed with rope or locking him/her in a room (chilling effects). The main objectives of our study are; (1) whether and to what extent such ruling causes the chilling effects; (2) whether and to what extent people"s attitude/evaluation toward such drastic preventive measures as tying a parent to bed changes when they are informed about the ruling; and (3) whether and to what extent the difference in forums (ADR or court), conclusions (either care-taking family members win or lose), and care-taking experience (with or without such experience) yield different reactions from the people. The author asks the respondents to answer the degree of willingness to take specific types of cares, such as living together and take care of the elderly person by himself/herself, after reading the scenarios (Objective 1). We also asked the respondents whether and to what extent they deem specific types of preventive measures (such as tying a person to bed with rope) as blameworthy or permissible (Objective 2). The scenario manipulation is; (1) about information (Scenario A as control group with no case information, Scenario B with similar conclusion by court, Scenario C with similar conclusion by ADR, and Scenario D with court judgment for plaintiff) and (2) about experience (Group 1 with experience of care-taking and Group 2 without such experience; we used the filter function of internet survey system) (Objective 3). The results are as follows. We do not find any chilling effect on the willingness to live together and to take care by himself/herself if the conclusions are similar to the Supreme Court ruling, while the willingness to take care becomes weaker if the conclusion is the judgment for the plaintiff affirming the legal liability. In the case that the type of care-taking is to live separately and to visit to take care of the old parent by himself/herself, we do not find any chilling effect either. Those who have experience of care-taking are consistently more willing to take care of a parent with senile dementia than those without such experience in both types of care-taking. As a whole, people deem the preventive measure of tying a person to bed with rope as blameworthy, but compared to the control group, the other three groups deem the preventive measure less blameworthy. Another type of preventive measure we asked is locking the elderly person with dementia in a room. As a whole, people deem this type of preventive measure as slightly permissible. Compared to the control group, which deems the preventive measure slightly blameworthy, the other three groups deem it rather permissible. In the both types of preventive measures, the group with care-taking experience deems the measures more blameworthy than the group without care-taking experience.