Conjoint analysis as a new methodology for Korean typography guideline in Web environment

摘要

Statistical significance approach such as the analysis of variance (ANOVA) has been commonly used to analyze the legibility elements of on-line and printed texts. However, few researches have been done with conjoint analysis. Therefore, a new methodology of Korean typography guidelines for Web pages using conjoint analysis is introduced because conjoint analysis has a great merit of its representation of relative importance. Twelve volunteer subjects participated to investigate the effects for font type (Goolim, Batang, and Dodum), font size (10 and 12 pt) and line spacing (100% and 150%) in reading preference tasks. The results showed that conjoint analysis provided quantified customer preference functions for multi-attribute typography alternatives, including the contributions of each attribute and attribute level to the overall typography utility and the relative importance of typography attributes. The analysis results also suggest the following preference structures in selecting Hangul typography at the aggregate level: Line spacing has a relative importance of 53%, the most preferred factor in typography, followed by fonts (35%). The least preferred attribute is size, with an importance value of 12%.