近年來在QS9000及TS16949等國際品保標準之推動下，從事品質實務者均開始重視量測變異分析（Gauge Variation Study）的研究，由於量測之精確與否在工業品質改善上扮演一個相當重要的角色，唯有精確之量測技術，才能保證對產品及製程品質的判定無誤。國際上對於品質保證之要求亦日趨嚴格，因此如何確保量測品質，已成為從事品質實務工作者的當務之急。

在量測變異分析研究中，依不同的資料型態可分為計量型與計數型兩部份，其中計數型包含順序型（ordinal）資料、二元型（binary）及服從卜瓦松分配（Poisson distribution）等資料，本研究則著重於順序型資料之探討。

食品界常使用官能品評作為判定食品可否被顧客接受的重要指標，而品評員本身及品評員間評分的一致性會影響食品官能品評之可信度。故本研究乃針對QS9000中所建議之兩種順序型資料評估方法（即Kappa及Kendall一致性係數的估算方法），進行比較分析並進一步探討品評員個數b、樣本大小n及重覆品評次數r之最佳組合。研究成果可作為進行食品官能品評等順序型資料量測變異分析時之參考。

Recently, gauge variation study has been highly regarded by the quality practitioners when QS9000 and TS16949 become fashionable requirements for manufacturing industries. Measurement plays a significant role in helping organizations improve their product quality. Good quality of products is the key factor of business success. Therefore, how to ensure the quality of measurement becomes an important task for the quality practitioners.

Gauge variation study can be divided into two parts by data types; continuous data and attribute data. Especially, attribute data also can be divided into three parts; ordinal data, binary data and the data follows Poisson distribution. Therefore, the main purpose of this research lays stress on ordinal data.

In food industries, sensory evaluation is increasing in importance because of the present economic emphasis on consumer needs and wants. However, the consistency within panelists and between panelists may affect the reliability of sensory evaluation. Therefore, in performing the gauge variation study for ordinal data, most food industries are using Kappa and Kendall concordant coefficient stipulated by QS9000. A comparative analysis has been conducted for evaluating the accuracy of gauge variation study among two methods (Kappa and Kendall). Moreover, the rationale for a proper choice of the number of panelists (b), the sample size (n) and replicate measurement (r) is discussed. Hopefully, it can provide a useful reference for the food industries.

1. AIAG Editing Group, Measurement Systems Analysis, Automotive Industry Action Group (1998).

2. Bartko, J. J., “The Intraclass Correlation Coefficient as a Measure of Reliability,” Psychological Reports, 19, 3-11 (1966).

3. Bower, K. M., “Measurement System Analysis with Attribute Data,” Minitab Inc., KeepingTAB, 35, 10-11 (2002).

4. Boyles, R. A., “Gauge Capability for Pass/Fail Inspection,” Technometrics, 43, 223-229 (2001).

5. Cicchetti, D. V. and Allison, T., “A New Procedure for Assessing Reliability of Scoring EEF Sleep Recordings,” American Journal of EEG Technology, 11, 101-109 (1971).

6. Cohen, J., “A Coefficient of Agreement for Nominal Scales,” Educational and Psychological Measurement, 20, 37-46 (1960).

7. Cohen, J., “Weighted Kappa: Nominal Scale Agreement with Provision for Scaled Disagreement or Partial Credit,” Psychological Bulletin, 70, 213-220 (1968).

8. Cox, E. P. Ⅲ, “The Optimal Number of Response Alternatives for a Scale: A Review,” Journal of Marketing Research, 17, 407-422 (1980).

9. Fleiss, J. N., “Measuring Nominal Scale Agreement among Many Raters,” Psychological Bulletin, 76, 378-382 (1971).

10. Friedman, M., “The Use of Ranks to Avoid the Assumption of Normality Implicit in the Analysis of Variance,” Journal of the American Statistical Association, 32, 675-701 (1937).

11. Pan, J. N., “Determination of the Optimal Allocation of Parameters for Gauge Repeatability and Reproducibility Study,” International Journal of Quality and Reliability Management,” 21, 672-682 (2004).

12. Pan, J. N., “Evaluating the Gauge Repeatability and Reproducibility for Different Industries,” Quality & Quantity, 40, 499-518 (2006).

13. Jeroen D. M. and Wessel V. W., “Measurement System Analysis for Bounded Ordinal Data,” Quality and Reliability Engineering International, 20, 383-395 (2004).

14. Kendall, M. G. and Babington-Smith, B., “The Problem of M Rankings,” The Annals of Mathematical Statistics, 10, 275-287 (1939).

15. Kendall, M. G. and Gibbons, J. D., Rank Correlation Methods, Oxford University Press, 119 (1990).

16. Landis, J. R. and Koch, G. G., “The Measurement of Observer Agreement for Categorical Data,” Biometrics, 33, 159-174 (1977).

17. Mandel, J., “Repeatability and Reproducibility for Pass/Fail Data,” Journal of Testing and Evaluation, 25, 151-153 (1997).

18. Montgomery, D. C. and Runger, G. C. “Gauge Capability Analysis and Designed Experiments. Part I: Basic Methods,” Quality Engineering, 6, 115-135 (1993).

19. Montgomery, D. C. and Runger, G. C. “Gauge Capability Analysis and Designed Experiments. Part II: Experimental Design Models and Variance Component Estimation,” Quality Engineering, 6, 289-305 (1993).

20. Stevens S. S., “On the Theory of Scales of measurement,” Science, 103, 677-680 (1946).

21. 潘浙楠、李文瑞，品質管理，華泰書局，(2003)。

22. 潘浙楠、江巧玉，“量測系統重複性與再現性的分析研究”，品質學報，第二期， 121-154頁 (2002)。

23. 台灣光電校正實驗室，“量具之重覆性與再現性討論”，台儀資訊，第二十一期，(2001)。

24. 呂執中、陳盈全，“離散型量測系統分析之研究”，成功大學，工業與資訊管理研究所碩士論文 (2003)。

25. 呂執中、陳銘男，“離散型量測系統重複性與重現性之研究”，品質學報，第一期， 31-49頁 (2004)。

26. 中華民國經濟部商品檢驗局，食品官感檢查手冊，經濟部商品檢驗局編印，(1984)。

27. 天坂格郎、長澤伸也，官能評價的基礎與應用，五南圖書，(2003)。

28. 魏夢麗、呂秀英，“官能品評資料的統計分析方法之正確使用”，農業試驗所技術服務季刊，第三期， 32-37頁 (2002)。

29. 魏夢麗、呂椿棠、呂秀英，“官能品評資料分析之SAS程式應用及例釋”，科學農業，第三、四期， 97-105頁 (2004)。