Comparison Results

From the comparison results of each participant, I sorted all 39 spoon designs and ranked them using the method described below.

With the sorting result as a tree structure, the spoon designs in the bottom “more kawaii” group were ranked as the most kawaii or the 1^st rank. The spoon designs in the bottom “less kawaii” group were ranked as the least kawaii. Two or more spoon designs in the “equally kawaii” group had the same rank.

From the rankings, I calculated the scores of all the spoon designs. The score of the 1^st rank was 39. If the rank was worse, the score was lowered. I used the scores to analyze the comparisons of the spoon designs and to describe the analyzed results in the following sections. Example of the method to rank and score the spoon designs is shown in Figure 4.6.

Figure 4.6 Example of the method to rank and score the spoon designs

4.5.2.1 Comparison among Shapes, Genders, and Nationalities

I performed 3-factor ANOVA to examine the effect of shape, gender, and nationality on the scores of spoon designs. The results were as follows:

 There was a significant main effect of shape (p<0.01).

 There was a significant 2-factor interaction effect between shape and nationality (p<0.01).

 There was a significant 3-factor interaction effect among shape, gender, and nationality (p<0.05).

For each shape, Tukey post hoc tests between participant groups showed significant differences in average scores between Japanese males and females for flower and smiley groups (p<0.05) (Figure 4.7).

Figure 4.7 Comparison of average scores among participant groups for each shape of spoon designs For each participant group, Tukey post hoc tests between shapes showed significant differences in average scores (Figure 4.8) as follows:

 Thai males: flower and heart (p<0.01), smiley and heart (p<0.05)

 Japanese males: flower and heart (p<0.01), flower and smiley (p<0.01)

 Japanese females: flower and smiley (p<0.01), smiley and heart (p<0.05)

 For Thai females, there were no significant differences

Figure 4.8 Comparison of average scores among shapes of spoon designs for each participant groups

These results indicate that each participant groups considered the following shapes as more kawaii spoon designs.

 Thai males preferred flower and smiley shapes.

 Japanese males preferred flower shape.

 Japanese females preferred smiley shape.

 Thai females preferred all shapes equally.

4.5.2.2 Comparison of Individual Spoon Designs

I performed 1-factor ANOVA of the average scores among 39 spoon designs. The result showed a significant main effect of spoon designs (p<0.01). Next, I performed Tukey post hoc tests to compare the average scores of each spoon design between the pairs of four participant groups. Figure 4.9 shows the average scores of the 39 spoon designs of four participant groups.

Figure 4.9 Average scores of 39 spoon designs among four participant groups. The spoon designs with bracket indicated that there was significant difference in score.

As the results, there were 25 spoon designs which had no statistically significant differences in average scores between any pairs of the four participant groups. Each of these spoon designs had similar ranking tendency for all participants. On the other hands, there were 14 spoon designs had significant differences in average scores between pairs of participant groups as follows:

 Thai males vs. Japanese males: H4 and S12

 Thai females vs. Japanese females: F7, F10, and H9

 Thai males vs. Thai females: F9, F11, F13, H7, and S3

 Japanese males vs. Japanese females: F8, H9, S8, S10, and S11

The spoon designs with different average scores can be used to suggest more kawaii spoon designs for specific participant groups. For example, F8 and H9 were more kawaii for Japanese males, while S8, S10, and S11 were more kawaii for Japanese females (Figure 4.10).

Finally, I compared the average scores among 39 spoon designs for all participant groups. The spoon designs which had 10 highest average scores (Figure 4.11) can be suggested as kawaii spoon designs in general.

Figure 4.10 More kawaii spoon designs for Japanese males (left) and females (right)

Figure 4.11 Top 10 kawaii spoon designs in general ranked by average scores of all participants

4.5.2.3 Correlation Analysis

Since I found the similarities and differences of average scores among the four participant groups, I performed further statistical analysis to clarify the correlations between the groups. I performed a correlation analysis of the average scores of the 39 spoon designs among the four participant groups using Spearman’s rank-order correlation. Spearman’s

correlation coefficient (rs) measures the strength and the direction of the monotonic relationships between two ranked variables. The rs ranges from -1 to +1, where -1 indicates a perfect negative association of ranks, +1 indicates a perfect positive association of ranks, and 0 indicates no association of ranks.

I compared the scores between four pairs of participant groups to clarify the similarities and differences between genders and nationalities. The rs value of the four comparison are shown in Table 4.1.

Table 4.1 Comparison results between participant groups by Spearman’s rank-order correlation

Thai Japanese

Male Female Male Female

Thai Male 0.329* 0.649** 0.294

Female 0.104 0.342*

Japanese Male 0.051

Female

** Correlation is significant at 0.01 level (2-tailed).

* Correlation is significant at 0.05 level (2-tailed).

For the comparison between genders, I compared the rs of the Thai vs. Japanese males (rs=.649, p<0.01) and the Thai vs. Japanese females (rs=.342, p<0.05). The rs of the males was higher, which indicated stronger correlation between males than females.

For the comparison between nationalities, I compared the rs of the Thai males vs.

females (rs=.329, p<0.01) and the Japanese males vs. females (rs=.051). The rs of the Thais was higher, which indicates stronger correlation between Thais than Japanese.

From the correlation analysis results, Thai males vs. Japanese males had a strong correlation, which indicates that they had a similar ranking tendency. These combinations had moderate correlations: Thai females vs. Japanese females, Thai males vs. Thai females, and Thai males vs. Japanese females. The combinations of Thai females vs. Japanese males and Japanese males vs. Japanese females had very weak or no correlations.

These results indicate that the tendencies of kawaii spoon designs between Thai and Japanese males are similar, while those between Japanese males and females are different.