楊薏蓓 董秀娟 郝躍云 何茹

Abstract：Present study collected 109 data of self-concept，brand personality and brand loyalty selected by preliminary research on consumers from mainland China.The result shows that the male score is much higher than female on brand personality.Using different brands of mobile phones produces different degree of brand loyalty，specifically，Apple users exhibit higher loyalty than Samsung users.

Key words：consumer；self-concept；brand personality；brand loyalty

METHOD

We randomly surveyed 120 participants，a total of 109 valid questionnaires were recovered.Brand Dimensions Scale （BDS）.Brand Loyalty Scale （BLS）.

PROCEDURE

Firstly，the participants completed the Brand Personality Scale to measure their ideal-self，actual-self and brand personality respectively.And then，they were required to finish the Brand Loyalty Scales.In our research，we chose the most wildly used method，named semantic differential means，to investigate the self and brand personality congruency （Kuenzel & Halliday，2010）.It represented the perceived congruency between self image and the personality of the brand used by consumers.Specifically，the participants completed the Brand Personality Scale first.And then，the self-concept was assessed with the same scales.After that，the discrepancy between brand personality and self image was calculated by the score difference for each item.Lastly，the sum of the differences of each item represents the incongruence between brand personality and self-concept.

RESULTS

Independent sample t tests are conducted to examine the gender difference on self-image，brand personality and brand loyalty，results of which indicates that males scored significantly higher than females on brand personality，t（1，107）=2.63，p=.01.Table 1 indicates the mean score and standard deviations of the five dimensions of brand personality.Compared with female，male scored significantly higher on confident，t（1，107）=2.24，p=.03，practical，t（1，107）=2.63，p=.01，precise，t（1，107）=2.57，p=.01，and brave t（1，107） =3.64，p <.001.

One way ANOVA is conducted to figure out whether the four brands differentiate from each on the four variables.Result shows a significant difference on the brand loyalty，F(xiàn)（3，105）=4.78，p=.004.Using the Post hoc Turkey tests to examine the difference，and it is indicated that Apple users exhibit higher loyalty than Samsung users.The means score and standard deviations on five dimensions of brand personality for different brands are illustrated.There is a significant difference on confident，F(xiàn)（3，105）=3.17，p=.03.Post hoc Turkey tests are also followed.Its found that Apple scored significantly higher than Samsung.Moreover，a significant difference are also found on practical，F(xiàn)（3，105）=3.76，p=.01.Post hoc Turkey tests shows，Samsung have significantly lower score than Xiaomi.

DISCUSSION

By result analysis of the score of brand loyalty，it seems the evaluation of male is much higher than female.Female is more sensitive and delicate to details than male guys and they will pay more attention to the defective part.However，male users will pay more attention on the stability of the mobile phone and pay very little attention to the faulty details.

The more intense that the sense of identity of consumer is，the consumer is more likely to form a long term relationship with the brand，which is called brand loyalty.Apple gets higher score than Samsung.“Confident” dimension shows fashionable optimism，and Apple，widely used by young people and focusing on fashion and innovation，is identical with the personality of confident.As for the dimension of “practical”，Samsung has lower score than Xiaomi.“Practical” is the characteristics of material benefit，and Xiaomi mobile phone just highlights its affordable features with its lower prices.

CONCLUSIONS

The male score is much higher than female on brand personality.Apple users produce higher brand loyalty than Samsung users.

REFERENCES

[1]Kuenzel，S.，Halliday，S.V.The chain of effects from reputation and brand personality congruence to brand loyalty：The role of brand identification[J].Journal of Targeting，Measurement and Analysis for Marketing，2010，18（3）：167-176.