1、簡單效應(yīng)分析Simple main effectv 必要性v 方法：v 改寫EMMEANS語句v 編寫MANOVA語句改寫改寫EMMEANS語句語句Paste到到syntaxv UNIANOVAv grade BY textbook methodv /METHOD = SSTYPE(3)v /INTERCEPT = INCLUDEv /POSTHOC = textbook method ( LSD T2 )v /EMMEANS = TABLES(textbook) COMPARE ADJ(LSD)v /PRINT = DESCRIPTIVE HOMOGENEITYv /CRITERIA = A

2、LPHA(.05)v /DESIGN = textbook method textbook*method .改寫改寫EMMEANS語句語句v檢驗檢驗textbook在在method上的簡單效應(yīng)上的簡單效應(yīng)vUNIANOVAv grade BY textbook methodv /METHOD = SSTYPE(3)v /INTERCEPT = INCLUDEv/EMMEANS = TABLES(textbook*method) COMPARE textbook ADJ(LSD)v/CRITERIA = ALPHA(.05)v /DESIGN = textbook method textbook

3、*method .v檢驗檢驗method在在textbook上的簡單效應(yīng)上的簡單效應(yīng)vUNIANOVAvgrade BY textbook methodv /METHOD = SSTYPE(3)v /INTERCEPT = INCLUDEv /EMMEANS = TABLES(textbook*method ) COMPARE (method) ADJ(LSD)v /CRITERIA = ALPHA(.05)v /DESIGN = textbook method textbook*method .編寫編寫MANOVA語句語句v檢驗檢驗textbook在在method上的簡單效應(yīng)上的簡單效應(yīng)vM

4、ANOVAvgrade BY textbook(1,2) method(1,3)v /DESIGN = textbook within method(1)v textbook within method(2)v textbook within method(3) .v檢驗檢驗method在在textbook上的簡單效應(yīng)上的簡單效應(yīng)vMANOVAvgrade BY textbook(1,2) method(1,3)v /DESIGN = method within textbook (1)v method within textbook (2).v 二要素分析，包括簡單效應(yīng)，多重比較檢驗！v G

5、LMa1b1 a1b2 a2b1 a2b2v /WSFACTOR = A 2 Polynomial B 2 Polynomial/METHOD = SSTYPE(3)/EMMEANS = TABLES(A*B) COMPARE (A) ADJ(LSD) /在B要素的各個程度上A要素的各個程度的多重比較(LSD方法)v /EMMEANS = TABLES(A*B) COMPARE (B) ADJ(LSD) /在A要素的各個程度上B要素的各個程度的多重比較(LSD方法)/PRINT = DESCRIPTIVE/CRITERIA = ALPHA(.05)/WSDESIGN = A B A*B.v v

6、 三要素條件v v GLMa1b1c1 a1b1c2 a1b2c1 a1b2c2 a2b1c1 a2b1c2 a2b2c1 a2b2c2v /WSFACTOR = A 2 Polynomial B 2 Polynomial C 3 Polynomial/METHOD = SSTYPE(3)/EMMEANS = TABLES(A*B) COMPARE (A) ADJ(LSD) /* A 要素在B要素各個程度上的效應(yīng)多重比較采用LSD/PLOT = PROFILE ( A*B) /* 作圖(A as 橫軸, B為橫軸)/EMMEANS = TABLES(A*B*C COMPARE (A) ADJ(LSD) /* A要素在B*C結(jié)合的各程度上的效應(yīng)./PRINT = DESCRIPTIVE/CRITERIA = ALPHA(.05)/WSDESIGN = A B C A*B A*C B*C A*B*C.v v 以上全部是反復(fù)丈量方差分析的例子，其他實驗設(shè)計在PASTE之后作相應(yīng)的修正即可。v 給個混合設(shè)計的程序，和完全被試內(nèi)有一點兒不同：v GLM a1 a2 BY B /WSFACTOR=A 2 Polynomial /A是被試內(nèi)變量，B是被試間變量 /METHOD=SSTYPE(3) /EMMEANS=TABLES(B