• Communications for Statistical Applications and Methods
• /
• v.18 no.5
• /
• pp.637-644
• /
• 2011

Partially ranked data refers to the situation in which there are p distinct objects; however each judge specifies only first s (s < p) choices. The group theoretic formulation for partially ranked data analysis was set up by Critchlow (1985). We propose a graphical method for partially ranked data by quantifying objects and judges. In a plot for judges, the interpoint distances can be interpreted as Spearman or Kendall distances between two rankings given by respective judges. Similarly, we also construct a plot for objects with a sensible relationship to the previous plot for judges. This study extends the Han and Huh (1995) quantification method of fully ranked data using Gabriel's (1971) biplot technique for multivariate data matrix.

• The Korean Journal of Applied Statistics
• /
• v.22 no.1
• /
• pp.163-170
• /
• 2009

The purpose of this article is to measure the level of insurance industry's social cognition based on the university students' survey questionnaire. For this, this study analyzes the cognition of students on the level of social reliability, the causes of low reliability, and the measures of enhancing the reliability level about insurance industry. For the analysis of survey questionnaire, partially ranked data analysis was employed. The results show that most students recognize that the reliability of insurance industry is lower than that of other financial industry such as banks and securities firms. The main reason for this low reliability is the lack of sales forces' specialty, the complexity of insurance products and contracts, and negative media release about insurance industry. To enhance the social reliability level of insurance, establishment of the future insurance image is essential, and that can be achieved through strengthening the insurance education and public relations and simplifying insurance products and contracts.