【正文】
=12, 查附表 3F界值表, ,(11,12) = = F ,(11,12) P ,即方差不齊同 =FINV(,11,12) =2*FDIST(,11,12)= Excel方差齊性檢驗 Excel方差齊性檢驗 結果 t’ 檢驗實例(表 ) 2 2 2 21 1 2 22 2 2 2 2 21 1 2 212( / / ) ( 59 / 12 9 / 13 ) 874( / ) ( / ) ( 59 / 12) ( 9 / 13 )12 1 13 111S n S nS n S nnn?? ?? ? ???????12221212 5 38 7 59 912 13XXtSSnn?? ? -= =++H0: ?1??2, H1: ?1??2, a? =TDIST(,15,2) P= 差異有統(tǒng)計學意義,高蛋白喂養(yǎng)的動物體重高于低蛋白組 Excel做方差不齊的 t ‘檢驗 Excel做方差不齊的 t ‘檢驗 結果 SPSS計算方法 SPSS計算結果 方差不齊的結果讀取 正態(tài)檢驗結果 Te s t s of N or m a l i t yb. 2 4 4 12 . 0 4 8 . 8 8 8 12 . 1 1 1V A R 0 0 0 0 1S t a t i s t i c df S i g . S t a t i s t i c df S i g .K o l m o g o r o v S m i r n o vaS h a p i r o W i l kL i l l i e f o r s S i g n i f i c a n c e C o r r e c t i o na . V A R 0 0 0 0 2 = 1 . 0 0b . T e s t s o f N o r m a l i t yb. 1 5 2 13 . 2 0 0 * . 9 5 1 13 . 6 0 6V A R 0 0 0 0 1S t a t i s t i c df S i g . S t a t i s t i c df S i g .K o l m o g o r o v S m i r n o vaS h a p i r o W i l kT h i s i s a l o w e r b o u n d o f t h e t r u e s i g n i f i c a n c e .* . L i l l i e f o r s S i g n i f i c a n c e C o r r e c t i o na . V A R 0 0 0 0 2 = 2 . 0 0b . The t Test formula Onesample t test Two independent samples t test Pairedsample t test D if f e r e nc e in the m e a ns 1st a nda r d e r r ordddtns? ??? ? ? ?t test When variances of the two samples are heterogeneous 2221212139。nSnSXXt???1)/(1)/()//(22222121212222121??????nnSnnSnSnS Statistical Analysis control group mean treatment group mean Is there a difference? Slide downloaded from the Inter What does difference mean? medium variability high variability low variability The mean difference is the same for all three cases Slide downloaded from the Inter signal noise difference between group means variability of groups = 167。 6. Two type error in hypothesis test 假設檢驗中的兩類錯誤 ? 假設檢驗是針對 H0,利用小概率事件的原理對總體參數(shù)做出統(tǒng)計推論。 無論拒絕 H0還是接受 H0,都可能犯錯誤。 假設檢驗的兩型錯誤 樣本假設檢驗 的推斷 結論 真實情況 拒絕0H 不拒絕0H 0H正確 I 型 錯誤 ( t ype I e rr o r ) 且犯錯誤的概率為 a ,即 檢驗 水 準 推斷正確 且正確結論的概率為 1 a? , 即 置信度( c o n fid e n c e le v e l ) 0H不正確 推斷正確 且正確結論的概率為1 ??, 即 檢驗 功效( p o w e r ) II 型 錯誤 ( t ype II err or ) 且犯錯誤的概率為? 演講完畢,謝謝觀看!