• Journal of the Korean Statistical Society
• /
• 제29권4호
• /
• pp.423-433
• /
• 2000

In this paper, we consider applying the bootstrap method to the median test procedures for right censored data. For doing this, we show that the median test statistics can be represented by the differences of two sampler medians. Then we review to the re-sampling methods for censored dta and propose the test procedures under the location translation assumption and Behrens-Fisher problem. Also we compare our procedures with other re-sampling method, which is so-called permutation test through an example. Finally we show the validity of bootstrap median test procedure in the appendix.

• 대한불안의학회지
• /
• 제4권2호
• /
• pp.99-103
• /
• 2008

Objectives : The authors sought to examine the differences in the number of P300 detected by subjects in experimental and control groups when exposed to relevant stimuli. In addition, the authors tried to determinethe subjects' reaction time when asked to respond to relevant stimuli. Methods : The subjects were 20 informed female volunteers aged 22-38 years (median age : 24). Relevant stimuli (concealed information) were created by unknowingly exposing the two groups to obscene pictures and semantic questions. The number of P300 detected and the length of reaction time were compared between the two groups. Results : Contrary to theauthors' expectations, there were no significant differences in the number of P300 detected or in the length of reaction time between the two groups. Conclusions : These results indicate that, unlike a polygraph test, concealed information tests adopting P300 and reaction time as major physiological responses ofsubjects exposed to meaningful personal information might not serve as a valid practical tool in the field of criminal psychology. However, further studies involving a greater number of subjects are needed before a final conclusion regarding the validity of this new type of concealed information test can be drawn.