以聯合分析法探討高中職女生夏季校服設計之研究

本研究以探討高中職女生夏季校服設計為研究主題，建立高中職女生夏季校服的屬性與水準，研究目前就讀於高中職女學生對於校服款式之喜好的屬性與水準為研究目的。本研究第一階段首先收集到臺灣地區517所高中職女生夏季校服圖片作為研究樣本，透過5位有校服產業相關經驗25年以上的專家溝通討論後，找出目前學校重視的校服屬性及水準，瞭解校服設計元素的各項因子與款式分佈情形，分別對於各校校服款式進行分析，以編碼方式分析圖片內容的各個屬性及各個水準。第二階段以階層式集群分析華德法進行統計分析，測量以歐基里得直線距離平方區間做為運算方式，經由集群分析法獲得四大群組。分別為:(一)優雅型校服、(二)清新型校服、(三)傳統型校服、(四)活潑型校服；再依據集群分析方式，分別由各群體中選取一個具代表性的樣本形態屬性及水準編碼，再從形態屬性及水準編碼中取得該群組代表之樣本圖片，再將圖片組成四組校服代表性設計選擇方案。第三階段將四組校服代表性設計選擇方案以直交排列法加以重新組合，產生八組受測體圖卡，再製作受測圖卡問卷，透過問卷調查受測者對整體輪廓受測體的喜好分數，分數表達方式為0分 - 100分，分數越高則表示喜好度越高。發放對象為現在正在就讀高中職1-3年級之學生，並採行滾雪球抽樣問卷，以紙本問卷方式一共發出1800份問卷，共回收1716份問卷，問卷回收率為95.33%；其中有效問卷1470份，有效樣本率85.66%。本研究發現就讀高中職女學生大多數穿著意願而言：學生是不期望廢除服儀的，認為穿著校服比較能提升個人人身安全的，對於穿著校服是能增添學生對於學校的認同感，學生是喜歡各校校服的，認為穿著校服上學，是省時與方便的且節省治裝費，認為校服款式是有需要再改變的空間，在款式方面能更新穎，增添短褲或長褲的款式，更能提高女學生穿著校服的意願。第四階段以利用SPSS 21進行聯合統計分析，產生受測者之成分效用值來獲得喜好的權重產生研究結果。產生校服之重要屬性權重如下： B上身領型(31.021%)> E下身款式(21.193%)> D下身顏色(16.612%)> A上身顏色(15.704%)> C搭配配飾(15.471%)。B上身領型相對重要性即高達31.021%，遠遠大於E下身款式、D下身顏色、A上身顏色和C配飾領結四個屬性。本研究得到屬性間的相對重要性喜好程度最低與喜好程度較高的組合結果如下：喜好程度最低的組合分別為：翻領款式素色上衣、褶子款式素色裙、上衣有配飾。喜好程度最高的組合分別為：國民領款式白色上衣、百褶款式格子裙、上衣無配飾。本研究在款式方面提出高中女學生喜好的建議為:上身為國民領白色上衣無配飾、下身為格子百褶裙的款式建議。本研究的結果有助於服裝設計師與學校行政人員設計校服時的參考依據。

關鍵字

集群分析；聯合分析；夏季校服

並列摘要

The purpose of this study is to understand the design of girls’ summer uniforms in high schools, in order to establish the attributes and levels of the uniforms. This study investigates the attributes and levels of school uniforms favored by girls currently studying in high schools. The first stage of this study collects pictures of girls’ summer uniforms in 517 high schools of Taiwan as the study samples. Through the consultation and discussion with 5 experts who have over 25 years of experience in school uniform industry, the attributes and levels of school uniforms that schools put emphasis on for now can be found. Furthermore, various kinds of factors and the distribution of each style can also be realized. In this way, an analysis of the styles of uniforms in different schools can be individually conducted, the result of which is sorted and coded with each attribute and level of the pictures. The second stage is to carry out the statistical analysis by Wards’ Hierarchical method using Euclidean algorithm as the formula. Through cluster analysis, four groups can be defined accordingly, which are 1. Elegant style, 2. Fresh style, 3. Traditional style and 4. Active style. Next, according to the cluster analysis, the attribute and level coding number of one representative sample are selected from each group. A representative sample picture of such group can be obtained from the its attribute and level coding number. After that, the pictures are combined to form representative design solutions for four groups. The third stage is to by orthogonal arrays, re-combine the representative design solutions of four groups and produce eight sets of picture cards and the questionnaires for the cards. These cards are distributed to subjects and they score based on their preference over the profile of the pictures. The score ranges from 0 to 100, with the higher score indicating stronger preference over such picture card. By means of snowball sampling, subjects are students who are studying in the first to third grade of high schools. A total of 1800 questionnaires are distributed, and 1716 questionnaires are recovered with a recovery rate of 95.33%. The numbers of valid questionnaires are 1470, and the valid sample rate is 85.66%. This study discovers that in terms of high school girls’ willingness of wearing school uniforms, they do not wish to abolish this convention as personal safety is considered to be relatively enhanced if they dress in school uniforms. For those who would like to increase the sense of identity with the schools in way of wearing school uniforms, the students show interest in having distinct school uniforms. They think wearing school uniforms to school is not only time-saving and convenient but also reducing cost on clothes. However, it is suggested that there is a need for improvement in the styles of school uniforms. The styles can be more innovative, and shorts or pants can be included, which may increase female students’ willingness of wearing school uniforms. The fourth stage is to SPSS 21 software to conduct conjoint analysis. The part-worth of the subjects is retrieved for comparing with their preference weighing to determine the result. The weighing of important attributes for school uniforms are as follows: B. Collared top (31.021%) > E. Bottom style (21.193%) > D. Bottom color (16.612%) > A. Top color (15.704%) > C. Accessories (15.471%). The relative significance of B. Collared top accounts for up to 31.021% which is far higher than other four attributes, E. Bottom style, D. Bottom color, A. Top color and C. Accessories and neckties. In this study, the combination of higher and lower preference among the relative significance of the attributes is illustrated below. The styles with lower preference are lapel plain tops, pleated plain skirts and tops with accessories, while those with higher preference are white tops with national collar type, pleated plaid skirts and tops without accessories. With regards to style, this study proposes to match a top with national collar type without accessories with a pleated plaid skirt as the bottom based on the preference of high school girls. The result of this study can be referred to for clothes designers and school administrative personnel when they design school uniforms. Key words: Hierarchical method, conjoint analysis, summer uniforms