Study for Style Image Evaluation of Scooters in Taiwan and China -A comparative study on formation elements and image analysis of scooters- Jui-Chen KAO*, Mitsuo KAMAIKE**, Toru NAGAO** *Chiba University, Graduate School of Science and Technology, 1-33 Yayoi-cho Inage-ku Chiba 263-8522 JAPAN, takakou@graduate.chiba-u.jp **Chiba University, 1-33 Yayoi-cho Inage-ku Chiba-shi Chiba 263-8522 JAPAN, kamaike@faculty.chiba-u.jp nagao@faculty.chiba-u.jp Abstract: Scooters that resemble each other are used in various areas in Asia. You may wonder users in different areas have what images they have toward scooters. It will be constructive to learn exactly what styles users want scooters to have based on rational analysis. The purpose of this study is to define what images scooters have in different regions and to estimate specific styles they are required to have through scientific research. In short, this study is to comprehend regional characteristics of scooter design images by comparison of evaluation results conducted in different regions. In order to clarify the image of the scooter style, a questionnaire survey had been conducted in China and Taiwan. 24 photographs of typical style images of scooters have been evaluated with 22 adjectives throughout 5 stages. Judging from the results of factor analysis and cluster analysis, regional characteristics are clarified. In Taiwan personality has a priority and in China a factor of stateliness is considered most important. It is detected that there is an explicit difference in images of scooters between Taiwan and China. Thus, when design targets are defined clearly, bases for required scooter design styles become ready to be assumed. These bases can provide suggestions for design development in each region. Key words: Scooter, Style, Regional comparison 1. Introduction Scooters and light vehicles are utilized for everyday transportation in Asia. Especially in Taiwan, their spread rate reaches one to two people and you cannot image streets in Taiwan without them. On the other hand, big waves of bicycles are observed in every street in China as if it is filled up by them. People in China are expected to replace bicycles with scooters dramatically in future. Though Taiwan and China have different backgrounds in many aspects, similar scooters are manufactured and used commonly. What images do these similar scooters have in different regions? It will be helpful to grasp what styles users tend to want for scooters through an objective evaluation method. It will be also valuable for design development to establish a specific survey method for evaluation, because design development of motorization in the Eastern Asia depends on interactions among markets.
2. Purpose and Method of research Beijing The purpose of this study is to clarify what images scooters have in different regions and to Changhou Jintan estimate specific styles they are required to have Shanghai Liyang Shanghai by means of scientific survey, so that it could be Pinghu grasped regional characteristics of scooter design images related to comparison of evaluation Jiaxing N Hangzhou Taipei Haining Kaohsiung results performed in different regions. 0 500 1000km This study used the method described below Fig.2 Investigation place which consisted mainly of questionnaire surveys about scooter design images in Taiwan and China (Shanghai). (1) Selection of scooter models: 24 typical models of scooters were selected and they were classified in styles. (2) Questionnaire surveys in typical areas in Taiwan and China: In order to learn what images users in Taiwan and China have toward the selected 24 models, I conducted questionnaire surveys. That reaeach employed factor analysis to analyze the survey results to learn about differences in the evaluation of users. (3) Statistical analysis: According to the questionnaire results and the factor analysis result, and used Quantification Theory Type I to compare the images of users in Taiwan and China. (4) Consideration and Conclusion: The user of Taiwan and China considers the image of scooters by the above analysis result. Conclusion, the direction of the style image of scooters comes by comparing regional characteristics. 3. Survey of images toward scooters in Taiwan and China 3.1. Selection of scooter models Category forming depends upon past experiences. It is difficult to extract individual categories in a strict sense. Thus, picked up 79 models of scooters that were sold in Japan, Taiwan and China in 2001 and focused on physical conditions to classify them into several patterns by using remembrance. Used the Ward Method for cluster analysis and selected 24 models for the survey(figure 1). 3.2. Subjects of the survey The survey ground: (Figure 2) China (The neighborhood area city in Shanghai, Changhou, Jintan, Liyang, Pinghu, Jiaxing, Haining, Hangzhou). Taiwan (Kaohsiung, Taipei). Period: China : August 29, 2001~September 10 Taiwan : November 5, 2001~end of November Subjects for the survey were selected among the users of scooters living in the regions described above. Effective answers were achieved from 117 users (39 in Taiwan and 78 in Shanghai). 3.3. Image experiment
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Fig.1 The sample of 24 types of scooters It was used the SD (Semantic Differential) method for analysis the 24 models mentioned above to subjects and asked them to evaluate the models using 22 pairs of adjectives. Selection of adjectives was based on the past survey of styling of passenger cars and modified it for this study. It was analyzed the result by the factor analysis. The 22 pairs of adjectives are as follows: Showy Humble, Modern Classical, Local Universal, Soft Hard, Feminine Masculine, Common Eccentric, Intimate Cold, Smooth Rough, Large Small, Buoyant Stately, Bright Dark, Refined Unrefined, Complicated Simple, Sturdy Delicate, Straight Rounded, Unique Orthodox, Sporty Familiar, Outstanding Modest, Round Irregular, Look Big Look Small, Thick Thin and Heavy Light. A subject evaluated each model with each pair of adjectives throughout 5 stages. Used the questionnaire data for a factor analysis in order to learn what evaluation axes these pairs of adjectives had. 4. The result and analysis of investigation 4.1. Result of factor analysis In order to the define evaluation axes of the subjects, and processed the data by the SD method. As a result, It was obtained the factor loading of each evaluation item as shown in the Table1, Table2, Table3. Rotation used the varimax method. 4.2. Interpretation of factor (Table1), (Table2), (Table3), The following is the result of factor analysis. When observing the accumulated contribution ratio to the third factor, it was obtained rather high percentages; 76.03% for the entire subjects, 77.78% for the Taiwan group and 76.53% for the China group. Therefore, it was decided to use the
Table1. Each adjective pair of factor loading categories--general Factor Stateliness Softness Commonness Adjectives First factor Second Factor Third Factor Straight 0.9341 0.0354 0.2016 Sturdy 0.6403 0.5304-0.3384 Sporty 0.3414 0.1224-0.5485 Modern 0.2612 0.0647 0.1307 Complicated 0.1987 0.3483-0.4152 Unique 0.0926 0.2540-0.8910 Look big 0.0817 0.9696-0.1341 Heavy 0.0602 0.9656-0.1511 Thick 0.0115 0.9511-0.2380 Large 0.0924 0.9351-0.0672 Showy -0.2664 0.4370-0.2805 Outstanding -0.1583 0.2799-0.8540 Round -0.9302 0.0762-0.1210 Local -0.1029 0.0621-0.9170 Common -0.0937-0.2564 0.9423 Smooth -0.6168 0.0014 0.5556 Intimate -0.8428-0.0833 0.3318 Feminine -0.7086-0.4699 0.2126 Buoyant -0.3723-0.8596 0.2101 Soft -0.8776-0.3025 0.2025 Refined -0.8379-0.1482 0.0550 Bright -0.8509-0.0601-0.2680 Square 6.2964 5.4420 5.0983 Ratio of contriburion 28.6199 24.7365 23.1741 Cumulative ratio of contriburion 28.6199 53.3565 76.5305 Table2. Each adjective pair of factor loading categories--taiwan Factor Rigidity Heaviness Universality Adjectives First factor Second Factor Third Factor Outstanding 0.8916 0.2640-0.1502 Unique 0.8646 0.2359 0.2862 Sporty 0.8157-0.0169 0.2785 Local 0.7821 0.1935-0.0117 Complicated 0.7408 0.1711 0.2144 Sturdy 0.5417 0.4646 0.6105 Showy 0.3568 0.3545-0.1162 Modern 0.2249-0.1354 0.2848 Look big 0.1549 0.9723 0.0969 Heavy 0.2078 0.9654 0.0179 Thick 0.1714 0.9589 0.0250 Large 0.1880 0.9167 0.0491 Round -0.2603 0.1185-0.8604 Straight -0.0284 0.0305 0.9167 Common -0.8596-0.4089-0.0881 Buoyant -0.1257-0.8984-0.2608 Smooth -0.8270 0.0007-0.3789 Feminine -0.5343-0.4028-0.6345 Intimate -0.6247-0.1759-0.6621 Refined -0.2174-0.1838-0.7269 Soft -0.5180-0.2532-0.7698 Bright 0.0164-0.0800-0.8341 Square 6.4425 5.5673 4.7169 Ratio of contriburion 29.2842 25.3063 21.4409 Cumulative ratio of contriburion 29.2842 54.5904 76.0313 Table3. Each adjective pair of factor loading categories--china Factor Personality Stateliness Hardness Adjectives First factor Second Factor Third Factor Large 0.9439-0.0045 0.1205 Look big 0.9369 0.0665-0.0332 Heavy 0.9258 0.0530 0.2590 Thick 0.9152 0.0009-0.2309 Sturdy 0.7117-0.4718-0.1633 Complicated 0.6468 0.0183-0.3221 Sporty 0.3961-0.2332-0.4459 Showy 0.3738 0.4327-0.5024 Intimate 0.0907 0.8930-0.0011 Smooth -0.0212 0.8265 0.2091 Soft -0.4166 0.8181-0.1583 Bright -0.0300 0.7609-0.4615 Refined -0.1626 0.7578-0.4169 Round 0.0938 0.7449-0.5544 Feminine -0.6714 0.5862-0.1431 Buoyant -0.8218 0.4654 0.0733 Outstanding 0.2886 0.2501-0.7834 Unique 0.0791 0.2359-0.9154 Common -0.0031 0.1702 0.9501 Straight 0.1409-0.6196 0.6268 Modern 0.3708-0.0251 0.1486 Local -0.1416 0.1423-0.9069 Square 6.4701 5.4762 5.1661 Ratio of contriburion 29.4096 24.8918 23.4824 Cumulative ratio of contriburion 29.4096 54.3014 77.7838 factors to the third as a base for analysis. These factors can describe images of scooters sufficiently. For the entire group, the first factor showed large factor loading for the items, Straight Rounded and Sturdy Delicate and it was named Rigidity factor. The second factor showed large factor loading for the items, Look Big Look Small, Heavy Light and Thick Thin.it was named Heaviness factor. The third factor showed large factor loading for the items, Common Eccentric and Smooth Rough and it was named Universality factor. For the Taiwan group, the first factor showed large factor loading for the items, Outstanding Modest and Unique Orthodox and it was named Personality factor. The second factor showed large factor loading for the items, Look Big Look Small and Heavy Light. it was named Stateliness factor. The third factor showed large factor loading for the items, Bright Dark and Soft Hard and it was named Hardness factor. For the China group, the first factor showed large factor loading for the items, Large Small and Look Big Look Small and it was named Stateliness factor. The second factor showed large factor loading for the items, Intimate Cold and Smooth Rough. it was named Softness factor. The third factor showed large factor loading for the items, Common Eccentric and Straight Rounded and it was named Commonness factor. 4.3. Interpretation of factors The figure shows the result of factor analysis. The followings are description of these major groups of factors: (1) Hardness Square headlight shows higher hardness. One-body of the front fender and the front panel shows higher hardness. The handgrip with the side turn signal lamps shows higher hardness. (2) Heaviness
Heaviness Second factor Third factor Universality Rigidity First factor Rigidity Fig.3 The dispersion figure of factor scores(first factor & Second factor, First factor & Third factor) First factor If the body under the seat looks larger, it obtains higher scores in heaviness. The headlight on the front side shows higher heaviness. One-body of the front fender and the front panel shows higher heaviness. (3) Universality One headlight achieved higher scores in universality. If a scooter has the rear carrier, it shows higher universality than that without it. 5. Difference in evaluation between Taiwan and China 5.1. Scatter chart and image map There is an explicit difference between Taiwan and China. The style image map was based on the factor analysis(figure 4), (Figure 5). Furthermore, using the scores of each region and classification by performed cluster analysis for sorting. The result of cluster analysis provides three clusters for Taiwan and four clusters for China. 5.2. Style images and regions 5.2.1. Consideration about Taiwan There are three image groups found with Taiwan samples and they can be described as follows: (1) Sporty This image has a movable front fender that gives mechanical impression and high performances and speed on a whole. (2) Stylish The steering handle has a headlight and the front side has side turn signal lamps. It is simple and easy to operate. (3) Utility The front fender and the front panel make one-body. The body length is rather long and gives stateliness. It is functional, practical and good for transportation and medium-range of trip. 5.3.2. Consideration about China There are four image groups found with China samples and they can be described as follows:
Second factor Stylish Second factor Smart Utility Casual Sporty Utility Family First factor Fig.4 The dispersion figure of factor score and image figure -Taiwan First factor Fig.5 The dispersion figure of factor score and image figure-china (1) Smart This image has a compact 1.Screen 2. Tire size body and the headlight is located on the handle and it is functional 10inch 12inch 16inch and smart. 3. Location of the light 6. Headlight (round, square and others) (2) Family The side turn signal lamps are not integrated with the headlight. The front side and the front mask 4. Height of the side cover 5. Under the seat 7. Side turn signal lamp (round, square and others) 8. Rear carrier have a soft and standard image. (3) Utility Fig.6 The items of the shape of scooter The front fender and the front panel make one-body. The body length is rather long and gives stateliness and sturdiness. (4) Casual It has a movable front fender and the body is compact. It gives an impression you can ride on easily. 5.3. Relation between scooters design elements and images The relation of the design composition elements of the obtained result obtained by image analysis of scooters appearances and and the design elements by using the Quantification Theory Type I. The 24 models of scooters can be classified by design elements as described below: 1. Screen (fixed fender and movable fender), 2. Tire size (10, 12 and 16 ), 3. Location of the light (8 locations in all), 4. Height of the side cover (1/3 or shorter, between 1/3 and 2/3, and 2/3 or higher), 5. Under the seat, 6. Headlight (round, square and others), 7. Side turn signal lamp (round, square and others), and 8. Rear carrier (long, short and none)(figure 6). These design characteristics are described in the matrixes. Then, it was employed the Quantification Theory Type I for the analysis of item/category about the shape of scooters by using the factor scores as criterion valuables. The result of analysis obtained partial correlation coefficients and ranges(table4, Table5). This analysis can be interpreted as follows.
Table4. The degree of influence of item by the range-taiwan First factorpersonality Under the seat 0.9028 6.7563 Location of the light 0.8923 5.5345 Rear carrier 0.8776 4.0149 Multiple correlation coefficient0.9334 Coefficient of determination0.8713 Second factorstateliness Under the seat 0.9257 6.8394 Tire size 0.9408 6.6415 Location of the light 0.9607 5.2922 Multiple correlation coefficient0.9729 Coefficient of determination0.9466 Third factor : Hardness Under the seat 0.8433 7.7504 Location of the light 0.8165 4.8543 Tire size 0.8017 4.7304 Multiple correlation coefficient0.9218 Coefficient of determination0.8497 Table5. The degree of influence of item by the range-china First factor : Stateliness Tire size 0.9032 3.6464 Location of the light 0.9249 3.4088 Rear carrier 0.7885 2.0852 Multiple correlation coefficient0.9619 Coefficient of determination0.9252 Second factorsoftness Under the seat 0.8009 9.3958 Tire size 0.8136 6.4167 Rear carrier 0.7745 6.0018 Multiple correlation coefficient0.8561 Coefficient of determination0.7329 Third factorcommonness Under the seat 0.9025 2.6895 Location of the light 0.8176 2.1319 Headlight 0.7452 1.3876 Multiple correlation coefficient0.94817 Coefficient of determination0.8990 In case of Taiwan, the first factor personality depended largely on Under the seat, Location of the light and Rear carrier s shape. For the second factor stateliness, it was Under the seat, Tire size and Location of the light that determined it. And for the third factor hardness, it was Under the seat, Location of the light and Tire size that determined it. According to the analysis, it was found that the design element Under the seat had a large influence on users. And Location of the light also turned out to be an important element. In case of China, the first factor stateliness depended largely on Tire size, Location of the light and Rear carrier s shape. For the second factor softness, it was Under the seat, Tire size and Rear carrier s shape that determined it. And for the third factor commonness, it was Under the seat, Location of the light and Headlight shape that determined it. According to the analysis, it was found that the design element Tire size had a large influence on users. And Location of the light also proved to be an important element. 6. Consideration This study is to define what images scooter users have toward similar scooters by surveying users in Taiwan and China. The followings are what this study found: (1) The result of the analyses are factors expressing the image of scooters by factor analysis and Quantification Theory Type I. The regions surveyed (Taiwan and China) had their own images toward scooters. As a result of factor analysis of the SD method, it was found that users in Taiwan had images described as personality, stateliness and hardness while users in China had images as stateliness, softness and commonness. (2) There is an explicit difference in image of scooter between Taiwan and China. In Taiwan a scooter is considered as something sporty, stylish and useful, while in China it is regarded as smart, familiar related and useful and casual. When the scooter design is judged according to the cluster classification, the first classification comes by carrying out a specification, it looks like that a scooter shape does not necessarily contribute to a design image as an independent variable. (3) Both in Taiwan and China, the design elements of Location of light and Rear carrier s shape had a large influence among users. It is a possibility that indication is carried out as a part of items (category) used by
Quantification Theory Type I. Therefore, it is thought that it is based on the combination of the items of the scooters such as arrangement of location of the light, tire size, rear carrier etc. 7. Conclusion and Prospect This study can define images that users in Taiwan and China have toward scooters. The survey and analysis method used in this study proved to be one of effective methods to estimate market trends about popular styles by comparing regional characteristics. In Asia the devices for individuals to use for transportation to cover middle and short distances, such as scooters, automobiles and bicycles are indispensable for everyday life. I will continue to study these devices from the viewpoint of ordinary people. References 1. Furuya Shiberu, Morimoto Masao. Transition of Design Policies and Basic Form for Japanese Passenger Cars. Bulletin of JSSD, 73, 57-64(1989). 2. Chang Yu-ming: The system which reasons form specification using an image concept set of car, Chiba University, 27-29, 1994.