A psychometric approach to the measurement of color preference.

A three-part investigation was conducted to explore the meaning of color preferences. Phase 1 used a Q-sort technique to assess intra-individual stability of preferences over 5 wk. Phase 2 used principal components analysis to discern the manner in which preferences were being made. Phase 3 used canonical correlation to evaluate a hypothesized relationship between color preferences and personality, with five scales of the Personality Research Form serving as the criterion measure. Munsell standard papers, a standard light source, and a color vision test were among control devices applied. There were marked differences in stability of color preferences. Sex differences in intra-individual stability were also apparent among the 90 subjects. An interaction of hue and lightness appeared to underlie such judgments when saturation was kept constant. An unexpected breakdown in control pointed toward the possibly powerful effect of surface finish upon color preference. No relationship to five manifest needs were found. It was concluded that the beginning steps had been undertaken toward psychometric development of a reliable technique for the measurement of color preference.