罗老师您好,我的IV,moderator1&2都是categorical的(1/0)Mediation 和DV是continuous的3 P* Q6 s( p3 u6 b) m
7 d: E$ Z+ D8 W% n; q" D/ y
IV与两个moderator是manipulate的,mediator和DV是measure的。。。现在的结果有小小奇怪的地方是,iv到mediator以及3-way interaction (iv*moderator1*2)到mediator是不显著的. 但是,mediation report显示有作用 (具体请看结果)。$ i/ k5 t' I. ^
请问这样的结果可以去分成八种情况(3-way)Interprate 么???
3 d6 d% B' E" p
W; r9 I9 P# P6 G/ V0 z# |谢谢
" b, @1 t8 O4 @
( c7 \# X& S, [1 QMy model results:
7 N6 V; Y+ Z; o' i/ u" N# d0 ~, @' _
0 z3 \! H( d3 e0 d2 V' i. ^. g/ {Run MATRIX procedure:
& W. f7 q8 c( Q2 E- R0 l
& u3 {8 W* t- K
) U+ Y: {8 n* y9 A) k************************************************************************** Model = 11 Y = mean_be X = similari M = mean_des W = method Z = culture + S" |- c: G- ~5 w* Z2 b, o
Sample size 197 ! W7 D8 m" s+ g, u" G9 ~. d) V
Outcome: mean_des
2 ^$ e5 F9 N0 I6 MModel Summary R R-sq F df1 df2 p .5534 .3063 11.9204 7.0000 189.0000 .0000 & Z! M: I2 ~6 U& L/ D
Model coeff se t p constant 5.9886 .1169 51.2177 .0000 similari .3486 .2339 1.4908 .1377 method 1.9783 .2339 8.4570 .0000 int_1 -.5295 .4678 -1.1318 .2591 culture .4626 .2338 1.9783 .0494 int_2 -.6915 .4677 -1.4785 .1409 int_3 -.3246 .4678 -.6939 .4886 int_4 1.2173 .9357 1.3009 .1949
# ?) C% B3 s+ O/ C0 {2 _Interactions: - S) \! j* Z+ {+ I
int_1 similari X method int_2 similari X culture int_3 method X culture int_4 similari X method X culture # j1 a& U7 [3 D+ p+ i& s
************************************************************************** Outcome: mean_be
" e; m8 {* ^; ^0 y4 X( r( `+ C/ sModel Summary R R-sq F df1 df2 p .2292 .0525 5.3762 2.0000 194.0000 .0053 / L1 {7 N" n0 o* _5 ]
Model coeff se t p constant 3.7189 .4102 9.0661 .0000 mean_des .2136 .0652 3.2757 .0012 similari -.1017 .2516 -.4044 .6863 $ W8 [4 }2 K3 H) w; s/ d
******************** DIRECT AND INDIRECT EFFECTS************************* . _- M* u4 W8 T' E% x- S. F
Direct effect of X on Y Effect SE t p -.1017 .2516 -.4044 .6863 0 c3 t5 \# A: D2 ]9 r
Conditional indirect effect(s) of X on Y at values of themoderator(s) 0 @$ ^6 V' N: d5 N
Mediator method culture Effect Boot SE BootLLCI BootULCI [url=]mean_des -.5127 -.5025 .2737 .1528 .0421 .6538[/url] mean_des -.5127 .4975 -.0073 .1136 -.2685 .2067 mean_des .4873 -.5025 .0299 .0923 -.1381 .2551 mean_des .4873 .4975 .0089 .0849 -.1580 .1887 3 k2 I" u6 A# x( }5 i: [
******************** ***YSIS NOTES AND WARNINGS************************* . _2 P. l/ n( ^) [
Number of bootstrap samples for bias corrected bootstrapconfidence intervals: 1000 # i+ k+ \' n. l) d- R u
Level of confidence for all confidence intervals inoutput: 95.00 * ^9 @4 E6 C# ^$ c( X4 l3 b
NOTE: The following variables were mean centered prior to***ysis: similarimethod culture ( }2 T0 ?/ I+ K5 [' R" \5 a5 ~ _
NOTE: Some cases were deleted due to missing data. The number of such cases was: 4 . e5 {7 e. ?# _5 h! I u: o& b( u
/ Y3 x# S$ e- d L+ I8 A5 L
' n; N; D, L4 O+ R
3 Q' v: k* A. a& A1 w
# C1 ~* T- V K |