【正文】
Sat t er t hw ai t e 法 t 39。 H1: ???0,該地難產(chǎn)兒與一般新生兒平均出生體重不同 。 1. z test z檢驗(yàn) ? 167。 ? Testing sample X should be a sample of a normal random variable. 檢驗(yàn)樣本是來(lái)自正態(tài)總體的隨機(jī)樣本 ? If X is not normal, t will have an unknown distribution and, strictly speaking, the ttest is inapplicable. However, according to the central limit theorem, as the sample size increases, the distribution of t tends to be normal. Therefore, if the sample size is big, we can use the ttest even if X is not normal. But there is no way to find out what value is big enough. This value depends on how X deviates from the normal distribution. Some sources claim that n should be greater than 30, but sometimes even this size is not enough. Alternatively, we can use nonparametric test: Wilcoxon ranksign test.(見 p79,第九章) 00 3. 42 3. 30 1. 770. 40 / 35XXXtSSn???? ?? ? ? ?3. 確定 P值,做出推斷結(jié)論 本例自由度 ??n1?351?34,查附表 2,得 ,34=。 應(yīng)用條件 ? Random and independent samples ? Normality ? Homogeneity of variance 兩組變量值分別來(lái)自隨機(jī)、獨(dú)立的正態(tài)分布總體 兩獨(dú)立樣本 t檢驗(yàn)計(jì)算公式 ? 稱為合并方差 (bined/pooled variance) 11221 212 121212221 2 1221 1 22 221( 1 ) ( 1 )(2( ) ( )11)2X X X XCCXXXX XXt n nSSSSSnnnSnnnSnn??????? ? ? ?? ? ? ? ?? ? ? ?? ? ? ?? ? ? ????? ?????1 21212 2XXXXt n nS ???? ? ? ?表 5 2 25 名糖尿病患者兩種療法治療后二個(gè)月血糖值 ( m m o l / L ) 編號(hào) 甲組血糖值 ( X2) 編號(hào) 乙組血糖值 ( X2) 1 8. 4 1 5. 4 2 10. 5 2 6. 4 3 12. 0 3 6. 4 4 12. 0 4 7. 5 5 13. 9 5 7. 6 6 15. 3 6 8. 1 7 16. 7 7 1 1. 6 8 18. 0 8 12. 0 9 18. 7 9 13. 4 10 20. 7 10 13. 5 11 21. 1 11 14. 8 12 15. 2 12 15. 6 13 18. 7 例 25例糖尿病患者隨機(jī)分成兩組,甲組單純用藥物治療,乙組采用藥物治療合并飲食療法,二個(gè)月后測(cè)空腹血糖 (mmol/L)如表 52 所示,問兩種療法治療后患者血糖值是否不同? 由原始數(shù)據(jù)算得 : n 1 =12 , ? X 1 =18 2. 5 , ? X 1 2 =29 53 . 43 , n 2 =13 , ? X 2 =14 1. 0 , ? X 2 2 =17 43 . 16 , 1X =Σ X 1 / n 1 =18 2. 5/ 12 =1 5. 21 , 2X = Σ X 2 / n 2 =14 . 16 / 13 =1 0. 85 代入公式,得 : H0: ?1=?2, H1: ?1??2, a? 1 5 9 =1 2+ 13 2= 23 2t ????,有統(tǒng)計(jì)學(xué)差異拒絕, 023,2/ , HPtt ???Effect size: Cohen’s d ? Cohen’s d: ? Represents mean difference in standard deviation units 122 218= = 7 331 1CXXdS??Effect size: Cohen’s d ? Same guidelines for interpreting Cohen’s d Effect size d Small ? .20 Medium .20 – .80 Large ? .80 ? Percent of variance explained ? Symbol: r2 222