服務創新類型及影響因素分析

Classification and Influence Factor of Service Innovation

10.6338/JDA.200802_3(1).0010

徐怡(Yi Hsu)；張志俊(Chih-Chun Chang)

服務創新 ； 集群分析 ； 類神經網路 ； 多次項迴歸模式 ； Service Innovation ； Cluster Analysis ； Artificial Neural Network ； Polynomial Regression Model

Journal of Data Analysis

3卷1期（2008 / 02 / 01）

145 - 165

繁體中文

研究目的除了解高度已開發國家服務業占GDP比率高達70%以上之重要性，並透過集群分析方式對服務業創新進行模式構建，以了解不同服務業下的不同解釋。本研究所使用的方法是調查法，依據商業週刊（2006）對1000大公司及上市上櫃服務業公司進行調查分析。在統計方法上，使用因素分析萃取因子，透過集群分析將樣本進行分類，根據各類別集群建構創新回歸模式。在問卷方面，總共郵寄出1380份問卷，回收89分問卷，有效問卷共83份。在統計結果方面，從64個變數中經由因素分析萃取為6個因子，因子名稱分別為組織溝通、以顧客為導向的服務、創新的外部因素、行銷策略、創新資訊來源、員工訓練，再依據集群分析分為兩大類，並透過類神經網路加以驗證其分類的正確性，最後進行多次項迴歸模式構建。在理論上的貢獻，將服務創新樣本公司分類，並比較不同模式間差異。在管理的意涵上，提供服務業廠商找出自己的集群，以利於投入服務創新的配置。對於後續研究者建議，可以根據不同集群對服務創新做出最佳化模式。

The purpose of this study is to categorize the sample companies for service innovation through cluster analysis, and then to establish a successful service innovation model for clusters of service innovation companies using polynomial regression model in order to find out the differences between the input of service innovation and the results due to different type of service innovation. For the survey questionnaires, 1,380 copies in total were sent out by mail, and 89 were returned, in which 83 were effective. Statistically, factoring analysis is used to extract the factors and the samples are categorized with cluster analysis. The polynomial service innovation regression model is then established based on each category of clusters. For statistical results, 6 factors are extracted from 64 variables using factoring analysis, and the 6 are structural communications, customer-oriented service, external innovation factors, marketing strategies, innovative information sources and employee training. Then, two major categories are developed based on cluster analysis and by means of Artificial Neural Network to demonstrate the effect of two major categories. Finally, a polynomial regression model is established. In theoretical contributions, service innovation company samples are categorized and the difference in polynomials between different clusters of samples is discovered through comparison. In the implication of management, service business owners are able to find out which clusters they belong to, and by looking into the result from polynomial regression, they are allowed to distribute their service innovation resource more properly. As for the suggestions for future studies, a optimization model can be developed for the compromise between the input and generation of service innovation based on various clusters.

1. Arvanitis, S.(2000).Explaining Innovative Activity and its Impact on Firm Performance in Service Industries: Micro Data Evidence for Switzerland.Proceedings of the paper presentation at the 25th CIRET conference,Paris:
2. Barcet, A.,Bonamy, J.,Mayere, A.(1987).Modernization et Innovation dans les Services aux Enterprises.Lyons:CEDES.
3. Cooper,Edgett(1996).Critical success factors for new financial services.Marketing Management,5(3),26-37.
4. Crozier, M.,Normann, R.,Tardy, G.(1982).L'innovation dans les Services.Paris:Mission a l'innovaton.
5. Evangelista, R.(2000).Sectoral Patterns of Technological Change in Services, Economics of Innovation.Economics of Innovation and New Technology,9,183-221.
6. Griffin, A.(1997).PDMA Research on new product development practies: updating trends and benchmarking best practices.Journal of product innovation management,14(6),429-458.
7. Hollenstein, H.(2003).Innovation Modes in the Swiss Service Sector: A cluster analysis based on firm-level data.Research Policy,32,845-863.
8. Kelly, D.,Storey, C.(2000).New Service Development: Initiation Strategy.International Journal of Service Industry Management,11(1),45-62.
9. Kleinknecht, A.(Ed.)(1996).Determinants of Innovation and Diffusion.London:Macmillan.
10. 昌金河譯、Shama(2005)。多變量分析與應用。台中:滄海書局。
11. 陳順宇(2005)。多變量分析。台北:華泰書局。
12. 龔明鑫、吳家豪(2006)。先進國家服務業發展策略對我國之啟示。台灣經濟研究月刊，11，14-28。