回复 1楼 helen526 的帖子6 S: p; ]/ l! q' u1 X6 w( l8 H6 e% e" r
! X: L/ {# c5 ?8 THelen 526,: g3 T5 F: l; i3 m! w
3 i% p' w+ Z+ ~你的模型是:
1 u. R! M! ? Y2 n3 x5 ?1 i" X3 @& v. u, g, @+ ^, k9 D6 L
T% D9 `. T8 \0 _# d- k7 S; }模型假设1:个体层面,A对B有显著的正向影响 假设2:C对A和B之间的关系,具有跨层次的调节作用 假设3:D、C对A和B之间的关系,具有三重调节作用 另,团队层面的控制变量包括,organiztion1, organiztion2; 个体层面的控制变量包括age,sex。A和B和D为个体层面变量,C为团队层面变量。 Model 1:只包含控制变量 Level 1
1 @9 S/ M; f( I- g) \B ij=β0j+β1j(age)+ β2j(sex)+ β3j(C) + β4j(D) +file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gifij Level 2
% L8 G& z9 E+ p, N: D- e& Eβ0j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image004.gif+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif01(organiztion1)+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif02(organiztion2)+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image006.gif β1j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif10+U1j β2j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif20+U2j β3j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif30+U3j β4j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif40+U4j (将调解变量C和D首先引入方程,是为了保证“在分析预测变量A的效果之前,首先控制C和D的作用”。请问,这样做可以吗?) Model 2: 加入预测变量A Level 1
6 o5 ^) @0 O* D; ?B ij=β0j+β1j(age)+ β2j(sex)+ β3j(C) + β4j(D) + β5j(A) +file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gifij Level 2
6 \" Q& ~# U8 _7 f. ^9 kβ0j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image004.gif+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif01(organiztion1)+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif02(organiztion2)+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image006.gif β1j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif10+U1j β2j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif20+U2j β3j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif30+U3j β4j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif40+U4j β5j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif50+U5j(file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif50表示A对B的效应,验证假设1,对吗?) Model 3: 加入所有的二维交互项A*C、A*D、C*D(验证假设2) Level 13 T8 A9 O3 M; |0 D& u2 A' M0 P8 |/ Z
B ij=β0j+β1j(age)+ β2j(sex)+ β3j(C) + β4j(D) + β5j(A) + β6j(A*C) +β7j(A*D) + β8j(C*D) +file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gifij Level 2! X: L3 O- k# ?$ a) P! V0 U
β0j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image004.gif+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif01(organiztion1)+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif02(organiztion2)+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image006.gif β1j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif10+U1j β2j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif20+U2j β3j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif30+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif31(C) +U3j β4j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif40+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif41(mean D)+U4j(由于,D本来属于个体层面的变量,所以涉及到level2的分析,一律用mean D的值代替, 这样合适吗?或者您认为在D是低层面变量时,应当怎样处理?)
; a- p) G) A* I5 k β5j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif50+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif51(A) +U5j β6j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif60+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif61(A*C) +U6j β7j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif70+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif71(A* mean D) +U7j β8j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif80+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif81(C* mean D) +U8j (看到JBP加油提供的文献中,file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif30(C)表示C对A和B关系的调节作用,但是我认为是file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif31表示C对A和B关系的调节作用?) Model 4: 三重交互作用(验证假设3) Level 1$ F- {3 k- E6 |- Q
B ij=β0j+β1j(age)+ β2j(sex)+ β3j(C) + β4j(D) + β5j(A) + β6j(A*C) +β7j(A*D) + β8j(C*D) + β9j(A*C*D)+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gifij Level 2
# [0 }- i& O3 r, h+ ]5 lβ0j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image004.gif+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif01(organiztion1)+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif02(organiztion2)+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image006.gif β1j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif10+U1j β2j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif20+U2j β3j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif30+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif31(C) +U3j β4j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif40+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif41(mean D)+U4j β5j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif50+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif51(A) +U5j β6j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif60+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif61(A*C) +U6j β7j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif70+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif71(A* mean D) +U7j β8j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif80+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif81(C* mean D) +U8j β9j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif90+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif91(A*C* mean D) +U9j (其中,file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif91(A*C*D)表示D对C\A\B关系的三重调节作用,对吗?) 如果写论文,下面表格报告的格式可以吗?
; n- a% W, E; c$ ?3 m4 b" f$ b 1 f8 z B8 x5 d& K1 n) ]
|
# V' @9 G3 E6 |2 l" ZModel 1
7 H, q) t" R" ^) w7 j# ~ |
9 R5 ]6 l8 b/ HModel 2
4 H0 U! U) R/ x3 v+ m# a. ^ |
7 d% a$ M3 ]6 i! v8 IModel 3
% `8 b. n9 p ]9 ^ |
% L' s& ^5 l0 `- A/ {4 j H$ LModel 4
^8 V( l* c0 c- Q' h | Interceptfile:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image004.gif | " V' m+ |- j$ p3 F
数值
. \ _- Z* y3 u( |2 w |
! R+ P- d# W+ v4 p" a. H数值
8 ]& \5 q9 l7 m# e5 J |
b/ B- I. U2 j- s. i3 B0 ~) g数值 ! G. x4 ]8 e' x( G8 z
| % t$ p7 M3 |6 U+ K, M4 {2 J
数值 8 b. c" B5 W+ k! M9 |5 @ Y
| control variables |
' D. S3 P7 z) H ^2 B" L: a |
( u- t) |$ m0 c! F! h. n# z |
/ m1 G( e/ X, z4 b1 v0 n3 Z) k4 P | 6 Z; y- t/ s! C9 F W
| Organiztion1 (file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif01) | 5 r: J/ \ d' Y1 d" K1 ?% U
数值 8 e0 y W- x y5 M5 j f
|
) J) c d( s, s" j4 U数值
. ^0 t! k- Q7 W" i% I& m4 S- q |
9 ~: \, h, y) r/ b1 q" q* S数值
8 v" ^; ^7 ^/ P3 t | ) D5 \* T6 ?# C
数值 4 o9 Y* B% e2 j& d& h
| Organiztion2 (file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif02) | - V6 q3 r% b4 \6 P$ h
数值 5 B$ p, e8 z! H9 ^0 f# W
| 0 T* i" w; Z* ?6 t
数值 / r# J* D5 M2 B5 c5 R- W
|
! v! q' P ~: }4 l- c7 q数值 + N# _' O; C/ `! \7 ^% Q- C! ?
|
3 B( K# v; d0 `1 K数值 , d) H& f& O3 o* O6 y. z( b
| Age (file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif10)
2 f" H# ?) W4 g1 q | 6 ]" W. j1 z! E3 |2 r- `% z( A( B
数值
: o" O3 O. T0 T& j0 v( \ |
0 e3 c/ A& C: t) x+ z! V( O数值 1 }+ o6 L1 {" `* y; B8 v
| % Z6 f; V ~* O- U2 s4 V) z, a: C
数值
# `& K* |- p$ D4 Q: K | ' d8 ]- @5 N/ J+ Z* E& b
数值 * V2 q+ n3 u# k6 P! s, ]
| Sex (file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif20) |
+ p5 E, `8 G5 S5 ^0 I数值
2 ?3 D5 ?2 Q7 {* n6 } | % Q# C* F: d; d+ p! T1 t" k
数值
8 I, u0 A: D3 R | 6 G2 y2 l5 d* a0 l; a: L$ p. _. ]
数值 - ^1 ~" {8 ^" b) l& L! |' O
|
4 Y, j. q- Z& n4 f" Y8 i5 y# ]数值
. ]) O( d0 J* U- n3 B% R | C(file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif30) |
- V- n0 N, Q: M, `. W& x数值 ) M, P- A7 p4 i, G( a3 \
|
s y$ e. C0 s3 ]( Z数值 D/ a- K0 [3 U- d) [: t# A
| 5 [4 f: ?3 A+ E/ K, O6 E. p
数值 9 V$ [" q% | N1 w h8 e \
| . C4 f$ ~ i9 q9 T9 e* [4 b6 Z
数值 8 j4 U# J+ T. l( c2 e% d
| D (file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif40) | , F4 n. N2 O" H: O* E# u
数值 - j9 P$ ~; o9 r) f0 J4 @. E
|
; Z" ^9 ^4 d, @: ~5 K数值 3 h0 {" N4 }5 D
| 5 G$ r- g7 f4 u# n# m3 D
数值 + |# I0 c& N1 z' X4 m) j1 V) b% ~
| 1 a4 V+ l- V L
数值 0 k. j) u0 w! }' y
| A (file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif50) | % k; \( @* l- n; B1 W( L
| & ~# x# K; _( [+ L, G; [
数值
3 P9 h; w- i; r- B$ |; Z | 3 @2 M" X; O, L6 I, y" E- T$ g
数值
) p# k) T" U. i0 @5 d! X | 9 Q- y+ _6 [. M5 A C! `2 t
数值 6 y; y7 D+ e5 _9 @& r+ B. u, j
| A*C(file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif60) |
. |1 l, O- L' Q6 f8 U9 B7 H) M | # f" @/ k& B: A; ]
| ) u6 r' s. _/ G7 g$ p
数值 @2 {0 v* X0 c& o! d* F, J
| % |0 k! U6 t v9 j+ T& _! J9 h( H
数值
6 m# w2 q9 R( b: M/ N+ c' q* g | A*D(file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif70) | 2 u9 h( y) W! N) ~4 |
| , ~2 h4 m4 o, O4 }
|
( g2 R) M v+ z8 K J+ R8 h# J数值
+ q5 P7 {- f! O0 L6 A& i0 r5 d |
1 n" G9 q; S6 U# N/ b# I数值 7 x# Z% f0 L' I+ k7 @
| C*D(file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif80) |
s* s" h, r% s$ q3 _ | 4 s0 }. }* [% ~3 w3 N5 U
| - A5 A; C1 X0 }) v
数值
1 M; X9 f3 t1 j9 M% k, [ |
3 P" Z2 L+ ?3 y0 D; R$ ?! {: J数值
# c6 r2 ~- a2 V$ e5 p3 e | A*C*D(file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif90) |
! p+ ?9 m ~" S2 |9 K9 @ |
' V$ s) Z9 J; v8 E& t! |1 o |
, l$ @9 |0 {6 \9 t4 G | 2 P1 `: t& n% ]6 F9 P: S9 s
数值
5 Z6 D, d2 N) K# v' K! F |
3 I9 A) Q* W6 K) I! z% x' z- } |
8 ]" j$ y& J' I, h+ `) l! S( @ |
- L$ R H& M1 D, S |
, i5 O8 B2 `4 F! }7 I% h |
) i! H4 b+ |) P( @ | X2 | $ }* [& [: O8 B( H
数值 3 K0 e, U0 F1 {5 j8 x) e& ]; m6 \
|
* x+ z. y( R* C, D数值 5 O4 N" u( T0 K. z- r% t0 D) u
|
X& e( t: b+ m- |5 X3 o2 F数值
1 k" R* _9 Q( V) }) L/ ^7 { |
5 ]( ^& |- _1 ?9 t( c/ i* w数值 + @4 _4 k+ [; k( ^4 D
| Deviance | ) F4 {( q6 ~7 M1 a" `; O* E
数值
) X1 b2 E P- u! U6 Q% N | # D8 d9 B" v5 p4 v6 G
数值
9 }+ _+ N7 |6 B, M) U |
0 r% r4 x5 o( I! M+ `! E9 D7 g数值
( J* Y7 u5 q. H$ [# S | / j% c1 B2 a5 E
数值 4 L" G6 P% F9 G4 |
| Decrease in Deviance |
' [$ L+ F/ Q% {: Wm
3 E j3 d# H! s! o( Z | 6 `1 t: I4 x' p. Q
n : v# D. n7 g; P+ ]; ^
|
N+ |- [0 S; l& O% R0 Ql 6 M- K$ P* g: v3 q3 v
|
$ e4 ?4 z% _9 U. @2 T5 r/ fk " J) c u* I. ?; n3 w$ l* v+ u
| m代表 Decrease in Deviance in comparison to the null model n 代表Decrease in Deviance in comparison to the model 1 l代表 Decrease in Deviance in comparison to the model 2 k代表Decrease in Deviance in comparison to the model 3 |
: I, _5 T* l7 u7 R7 NHelen,) f7 p% m8 G. a
我怀疑你的模型根本就不可能,所以你的问题我不知道如何回答。6 ^- j" X, z$ ~% g0 B
凡是一个高层阶的变量影响(或是调节)一个低层阶的变量,就表示这个高层阶的变量在低层阶中是不变的。所以在你的模型中,在每一个层阶中,C对A→B的影响是不变的。那么D何来再调节这个不变的影响呢?如果要有一个三阶的调节,而第一个是高层的话,另外一个一定也是高层的。你的问题只有当反过来说时,才会成立。就是D(低层)的变量调节A→B的影响,而这个调节再被C(高层变量)来调节。9 t6 B0 @; t1 {$ [5 w6 t1 }: U: p' M" ^
在这个情形下,我会这样模拟:4 k+ s, r/ v: [$ F& Q7 i
Level 1
0 X4 C( j+ o+ b( v h; WB ij=β0j+β1j(age)+ β2j(sex)+ β3j(Aij) + β4j(D) + β5j(D*A)+file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gifrij Level 2
, c! b* [8 d" U) H' ^β0j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image004.gif=gfile:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif10 +U1j (organization 是grouping variable,不用模拟的)
9 n- ~. Y* f1 {8 ^β3j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image004.gif=gfile:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif30 +gfile:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif31*C +u3j
% y7 Q$ t- R/ u$ L* y) b" T β4j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image004.gif=gfile:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif40 +gfile:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif41*C +u4j 6 Q" U9 Q; g9 h, y" O$ S' e
β5j=file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image004.gif=gfile:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif30 +gfile:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif51*C +u5j
$ d0 B8 \ P+ D7 z, o4 p& z. s 参数gfile:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image002.gif51就是你要的three-way interaction了。 " O; a- t& R$ h; W- ~
Kenny
4 J+ Q! s( |6 c X- o2 v1 }. z 9 i9 ]/ |! D8 J" u
file:///C:/DOCUME%7E1/mnlaw/LOCALS%7E1/Temp/msohtmlclip1/01/clip_image006.gif 本帖最后由 Kenneth 于 2011-6-12 23:50 编辑
' P, t5 d$ b: s% z/ V. N
. |$ g1 J: l3 c |