연속적 간격법의 타당성

Abstract

The current study aimed to confirm the validity of the method of successive intervals (SI) as a scaling method for stimuli and attributes. In order to achieve this research objective, the measurement of stimuli and attributes calculated by SI and item response theory (IRT) were compared. The blameworthiness of 33 behaviors (stimuli) was measured, and based on the measurement, the psychological attribute of ‘representation of innocence’ was measured. The correlations between the measurements calibrated by SI and IRT were estimated for the blameworthiness and the ‘representation of innocence.’ The theoretical relationship among ‘representation of innocence,’ legal threshold for a guilty verdict (beyond a reasonable doubt), the individual practical threshold for a guilty verdict, and the difference between the two thresholds estimates. These analyses were conducted on the 300 participants (survey sample) collected for this study and replicated on the 360 participants (cross-validation sample) obtained from previous research data. As a result, it was found that the two lists of measures for the blameworthiness of behaviors measured by SI and IRT were positively correlated. However, the correlation of measures of ‘representation of innocence’ from the two methods was observed only in the survey sample. In addition, the ‘representation of innocence’ measured by SI showed a theoretically predicted relationship with the legal conviction standard, and the measurement by the IRT showed the relationship predicted by theory with the actual conviction criteria only in the cross-validation sample. In the discussion, the author demonstrated the validity of SI as a scaling method for stimulus and attribute and suggested the direction for further research.