This research utilizes the method from research of Romaniuk and Sharp (2000) to discover the owned attributes for each brand color. The concept is to reduce usage bias of data by identifying a second pattern in image responses. The problem of usage bias is that brands with more users tend to score higher on image responses. This study aims to transform collected data from focus group into invisible patterns and distinguish owned attribute for each brand color. First, this research counts the total number of traits (attributes) for each color and presents in order. That is, out-of-doors occurs frequently with 50 times. As the usage effect is reflected in the number of responses for each color, the concept of chi-square can be utilized to estimate the expected number of responses in each cell. The following formula is used to calculate the expected values.