조절된 매개효과 봐주실 수 있으신가요?
X, M은 Y에 유의미한 영향이 확인 되었고, W는 Y에 영향을 미치지 않았습니다.
근데 여기서 제가 궁금한 점은 M*W 상호작용항이 유의하긴 하나 B값이 매우 작다는 것입니다.
매크로 모델에서 유의확률만 유의하면 사용은 가능하다고 하나, B값이 작아 실제로 논문 결과에 적용할 수 있을 지 의문입니다.
밑에 결과 적어두겠습니다..!
Model : 14
Y :
X :
M :
W :
Covariates:
SEX age edu
Sample
Size: 221
**************************************************************************
OUTCOME VARIABLE:
M
Model Summary
R R-sq MSE F df1 df2 p
.3448 .1189 16.8776 7.2864 4.0000 216.0000 .0000
Model
coeff se t p LLCI ULCI
constant -.7537 3.8180 -.1974 .8437 -8.2791 6.7716
X .1982 .0491 4.0387 .0001 .1015 .2949
SEX -1.8968 .6229 -3.0451 .0026 -3.1245 -.6690
age .0234 .0478 .4891 .6253 -.0709 .1177
edu -.1140 .0848 -1.3450 .1800 -.2812 .0531
**************************************************************************
OUTCOME VARIABLE:
Y
Model Summary
R R-sq MSE F df1 df2 p
.5287 .2795 6.6362 11.8034 7.0000 213.0000 .0000
Model
coeff se t p LLCI ULCI
constant -2.7218 2.4760 -1.0993 .2729 -7.6024 2.1588
X .1595 .0324 4.9182 .0000 .0956 .2234
M .1113 .0434 2.5649 .0110 .0258 .1968
W .0024 .0028 .8663 .3873 -.0030 .0078
Int_1 .0012 .0005 2.2521 .0253 .0001 .0022
SEX 1.2996 .4018 3.2343 .0014 .5075 2.0916
age .0574 .0322 1.7838 .0759 -.0060 .1208
edu .0164 .0594 .2758 .7830 -.1006 .1334
Product terms key:
Int_1 : M x W
Test(s) of highest order unconditional interaction(s):
R2-chng F df1 df2 p
M*W .0172 5.0719 1.0000 213.0000 .0253
----------
Focal predict: (M)
Mod var: (W)
Conditional effects of the focal predictor at values of the moderator(s):
W Effect se t p LLCI ULCI
-80.8455 .0182 .0646 .2816 .7785 -.1091 .1455
.0000 .1113 .0434 2.5649 .0110 .0258 .1968
80.8455 .2044 .0549 3.7221 .0003 .0962 .3127
Moderator value(s) defining Johnson-Neyman significance region(s):
Value % below % above
-18.2126 46.6063 53.3937
Conditional effect of focal predictor at values of the moderator:
W Effect se t p LLCI ULCI
-163.9910 -.0776 .1004 -.7725 .4407 -.2756 .1204
-151.1410 -.0628 .0945 -.6641 .5073 -.2491 .1236
-138.2910 -.0480 .0887 -.5408 .5892 -.2229 .1269
-125.4410 -.0332 .0830 -.3996 .6898 -.1968 .1305
-112.5910 -.0184 .0775 -.2373 .8127 -.1711 .1343
-99.7410 -.0036 .0721 -.0497 .9604 -.1457 .1385
-86.8910 .0112 .0669 .1677 .8670 -.1207 .1431
-74.0410 .0260 .0620 .4196 .6752 -.0962 .1483
-61.1910 .0408 .0574 .7107 .4780 -.0724 .1540
-48.3410 .0556 .0533 1.0439 .2977 -.0494 .1607
-35.4910 .0704 .0497 1.4182 .1576 -.0275 .1683
-22.6410 .0852 .0467 1.8259 .0693 -.0068 .1772
-18.2126 .0903 .0458 1.9712 .0500 .0000 .1807
-9.7910 .1000 .0445 2.2490 .0255 .0124 .1877
3.0590 .1148 .0432 2.6598 .0084 .0297 .1999
15.9090 .1296 .0428 3.0253 .0028 .0452 .2141
28.7590 .1444 .0435 3.3183 .0011 .0586 .2302
41.6090 .1592 .0452 3.5260 .0005 .0702 .2482
54.4590 .1740 .0477 3.6522 .0003 .0801 .2680
67.3090 .1888 .0509 3.7117 .0003 .0885 .2891
80.1590 .2036 .0547 3.7226 .0003 .0958 .3115
93.0090 .2184 .0590 3.7013 .0003 .1021 .3348
Data for visualizing the conditional effect of the focal predictor:
Paste text below into a SPSS syntax window and execute to produce plot.
DATA LIST FREE/
M W Y .
BEGIN DATA.
-4.3367 -80.8455 3.4235
.0000 -80.8455 3.5024
4.3367 -80.8455 3.5813
-4.3367 .0000 3.2128
.0000 .0000 3.6955
4.3367 .0000 4.1782
-4.3367 80.8455 3.0020
.0000 80.8455 3.8885
4.3367 80.8455 4.7751
END DATA.
GRAPH/SCATTERPLOT=
M WITH Y BY W .
****************** DIRECT AND INDIRECT EFFECTS OF X ON Y *****************
Direct effect of X on Y
Effect se t p LLCI ULCI
.1595 .0324 4.9182 .0000 .0956 .2234
Conditional indirect effects of X on Y:
INDIRECT EFFECT:
X -> M -> Y
J9B Effect BootSE BootLLCI BootULCI
-80.8455 .0036 .0121 -.0199 .0287
.0000 .0221 .0113 .0031 .0469
80.8455 .0405 .0173 .0103 .0782
Index of moderated mediation:
Index BootSE BootLLCI BootULCI
W .0002 .0001 .0000 .0005
*********************** ANALYSIS NOTES AND ERRORS ************************
Level of confidence for all confidence intervals in output:
95.0000
Number of bootstrap samples for percentile bootstrap confidence intervals:
5000
W values in conditional tables are the mean and +/- SD from the mean.
NOTE: The following variables were mean centered prior to analysis:
W M
------ END MATRIX -----
Existing replies
이일현 (2025-07-29 14:39:44)
상호작용 효과의 B 값
Int_1 .0012
조절된 매개효과의 B 값입니다.
Index of moderated mediation:
Index BootSE BootLLCI BootULCI
W .0002 .0001 .0000 .0005
B 값이 .0012, .0002 로 매우 작게 나옵니다.
이는 두 변수의 단위와 관계가 있습니다.
평균중심화해서 분석을 했는데
M = -4.3367~4.3367
W = -80.8455~80.8455
입니다.
이를 바탕으로 예측하면 W 는 최대값이 160 이상이죠.
즉 W 의 단위가 매우 크므로 W 값이 1 커질 때 X --> Y 에 미치는 간접효과의 B 값은 0.0002 만큼 변화하는 것입니다.
X, M, W, Y 등의 값을 합으로 하지 말고 평균으로 계산해서 분석을 하면 납득 가능한 B 값이 나올 것입니다.
아래 링크를 확인해 보세요.
Legacy document_srl: 307490 / Legacy URL: http://www.statedu.com/QnA/307490
댓글