PASS 2008

張貼者:2010年9月30日 上午2:59未知的使用者   [ eddie liu 已於 2012年8月9日 上午12:50 更新 ]

抽樣分析實驗設計各種數據含t-tests、F-tests、z-tests、proportions、survival curves、correlations、、analysis of variance、log rank tests、multiple regression等。

開發商: NCSS
採購正式版、大量授權報價、技術支援、軟體諮詢、委託採購、詢問報價請來電 02-29299388 分機16 , 

統計檢定的力量是概率的測試程序將導致在統計意義。 電力是關係到樣本大小,大小I型(α)錯誤,實際大小的效果,實驗誤差的大小。 所有這些都必須考慮,以計算的統計力量。

由於統計意義通常是理想的結果,規劃和運行研究,實現了高功率是頭等重要的研究者。 由於複雜的計算,測定的權力往往被忽視,或只是偶然的嘗試是在採取一些計算,所謂的“原則進行的拇指。”

Power Analysis
A statistical test's power is the probability that the test procedure will result in statistical significance. Power is related to the sample size, the size of the type I (alpha) error, the actual size of the effect, and the size of experimental error. All of these must be considered in order to calculate statistical power.

As statistical significance is usually the desired outcome, planning and running a study to achieve a high power is of prime importance to the researcher. Because of the complexity of the calculations, the determination of the power is often ignored or only a casual attempt is made at its calculation by adopting some, so-called, "rule-of-thumb."