在統計研究上，資料收集以及P值(P-value)是常見的兩個問題。本研究針對這兩個問題分別去探討:
1.同一問卷若設計成不同尺度，所得的結果是否會不同?
2.在統合分析(Meta-Analysis)中，如何簡單且近似的估計P值。
第一個問題主要探討的問題是，如何把不同尺度資料都轉化到同一尺度、哪種尺度是最好的評量尺度都是值得探討的問題。本研究一開始透過給同一批受試者填寫同樣題目不同尺度的問卷，觀察他們的填答狀況變化並且找出對應方式並且從3個尺度中找出問卷評量最適合的尺度。結果顯示，為當尺度改變的時候，填答者填答的狀況也更為左偏。一般來說，尺度提升的時候，填答者給的分數平均也會變高，所以如果想要明確區分出使用者喜好，使用5分量表是較好的；如果單純希望填答結果偏向樂觀，則使用100分量表是較好的。本研究使用模糊理論建構模糊歸屬函數，建立了一個對照表來顯示不同尺度資料轉化時的狀況。
第二個問題裡，當使用統合分析時，如何估計P值。傳統上來說都是使用Peto法或是MH法，此兩種方法都是將未計算到的資料直接忽視，或是使用某些數值插補運算 (Peto-CC法、MH-CC法等) ，但插補前後所得P值均與真實值仍有落差。Liu et al. (2014) 提出了一種稱為CDF法 (Confidence Distribution Functions) 的統合分析，把前面的資料整合在一起，透過數學運算整合稱以得出更近似的P值。此方法提供了P值更近似的估算法，但由於使用時需要大量的數學運算，使得其他非數理統計背景的研究學者較不易使用，因此本研究在此提出簡單的估計方法，並驗證此方法與CDF法的估值結果差距不大。在閱讀相關文獻時，可以直接使用此簡單的估計法來針對文獻中給的Peto法或是MH法自行估計出一個近似真實P值的數值。

Through the Development of statistical study, data collection and analysis are very important topics. With the era of big data, statisticians developed many ways about data mining, the situation of data insufficient will be less and less in the future, the new issue is, how to get useful information from those data? On medical research, Meta-Analysis is a common statistical method. First, to explore the theme of the same or similar topics in the past papers. Second, to choice data which can be used, Third, to analysis them. This way offer a very big help in some rare diseases which are difficult to collect. In order to fully use all data, we need to find a way to change dimension to merge Questionnaires of different dimensions and find a ways to support people of different backgrounds to estimate P-value easily. That’s the two points of this thesis.

林原宏、楊慧玲 (2002)‧模糊語意量表與傳統量表計分之模擬比較分析‧第四屆華
人心理學家學術研討會暨第六屆華人心理與行為科際學術研討會，台北市。
徐村和 (1998)‧模糊德菲層級分析法‧模糊系統學刊，第四卷第一期，59-72。
Armitage, P., Berry, G., & Matthews, J. N. S. (2008). Statistical methods in medical research. Example 15.6, Page516, John Wiley & Sons.
Berndt, D. J., & Clifford, J. (1994). Using Dynamic Time Warping to Find Patterns in Time
Series. In KDD workshop, Vol. 10, No. 16, 359-370.
Bradburn, M. J., Deeks, J. J., Berlin, J. A., & Russell Localio, A. (2007). Much ado about
nothing: a comparison of the performance of meta‐analytical methods with rare events. Statistics in medicine, Vol. 26, No. 1, 53-77.
Fleiss, J. L. (1993). Review papers: The statistical basis of meta-analysis.Statistical methods in medical research, Vol. 2, No. 2, 121-145.
Haldane, F. D. M. (1980). General relation of correlation exponents and spectral properties
of one-dimensional fermi systems: Application to the anisotropic s= 1/2 Heisenberg chain. Physical Review Letters, Vol. 45, No. 16, 1358-1362.
Liu, D., Liu, R. Y., & Xie, M. G. (2014). Exact meta-analysis approach for discrete data
and its application to 2×2 tables with rare events. Journal of the American Statistical Association, Vol. 109, No. 508, 1450-1465.
Mantel, N., & Haenszel, W. (1959). Statistical Aspects of the Analysis of Data From
Retrospective Studies of Disease. Journal of the National Cancer Institute, Vol. 22, No. 4, 719-748.
Nissen, S. E., & Wolski, K. (2007). Effect of rosiglitazone on the risk of myocardial
infarction and death from cardiovascular causes. New England Journal of Medicine, Vol. 356, No. 24, 2457-2471.
Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1988). SERVQUAL: A multiple-item
scale for measuring consumer perceptions of service quality. Journal of Retailing, Vol. 64, No. 1, 12-40.
Peto, R., & Peto, J. (1972). Asymptotically Efficient Rank Invariant Test Procedures.
Journal of the Royal Statistical Society. Series A (General), Vol. 135, No. 2, 185-
207.
Zadeh, L. (1965). Fuzzy sets. Information and Control, Vol. 8, No. 3, 338-353.