没做变量中心化之前
Model = 1
Y = TSA
X = avg外
M = avgsart
Sample size
30
**************************************************************************
Outcome: TSA
Model Summary
R R-sq MSE F df1 df2 p
.5314 .2824 4.7666 3.4104 3.0000 26.0000 .0323
Model
coeff se t p LLCI ULCI
constant -152.7415 76.0772 -2.0077 .0552 -309.1255 3.6425
avgsart 7.3360 3.4875 2.1035 .0452 .1671 14.5049
avg外 16.7972 7.3466 2.2864 .0306 1.6956 31.8988
int_1 -.7225 .3380 -2.1374 .0421 -1.4173 -.0276
Product terms key:
int_1 avg外 X avgsart
R-square increase due to interaction(s):
R2-chng F df1 df2 p
int_1 .1261 4.5683 1.0000 26.0000 .0421
*************************************************************************
Conditional effect of X on Y at values of the moderator(s):
avgsart Effect se t p LLCI ULCI
19.3690 2.8035 .9666 2.9004 .0075 .8166 4.7904
21.3939 1.3405 .5825 2.3012 .0296 .1431 2.5380
23.4188 -.1224 .8254 -.1483 .8832 -1.8192 1.5743
Values for quantitative moderators are the mean and plus/minus one SD from mean.
Values for dichotomous moderators are the two values of the moderator.
******************** ANALYSIS NOTES AND WARNINGS *************************
Level of confidence for all confidence intervals in output:
95.00
------ END MATRIX -----
做了中心化以后:
Model = 1
Y = TSA
X = avg外
M = avgsart
Sample size
30
**************************************************************************
Outcome: TSA
Model Summary
R R-sq MSE F df1 df2 p
.5 .3 4.8 3.4 3.0 26.0 .0
Model
coeff se t p LLCI ULCI
constant 17.7 .4 40.3 .0 16.8 18.6
avgsart .1 .2 .3 .8 -.4 .5
avg外 1.3 .6 2.3 .0 .1 2.5
int_1 -.7 .3 -2.1 .0 -1.4 .0
Product terms key:
int_1 avg外 X avgsart
R-square increase due to interaction(s):
R2-chng F df1 df2 p
int_1 .1 4.6 1.0 26.0 .0
*************************************************************************
Conditional effect of X on Y at values of the moderator(s):
avgsart Effect se t p LLCI ULCI
-2.0 2.8 1.0 2.9 .0 .8 4.8
.0 1.3 .6 2.3 .0 .1 2.5
2.0 -.1 .8 -.1 .9 -1.8 1.6
Values for quantitative moderators are the mean and plus/minus one SD from mean.
Values for dichotomous moderators are the two values of the moderator.
******************** ANALYSIS NOTES AND WARNINGS *************************
Level of confidence for all confidence intervals in output:
95.00
NOTE: The following variables were mean centered prior to analysis:
avg外 avgsart
------ END MATRIX -----
不懂为何选择变量中心化以后小数只有一位了,这个位数在哪来能改变吗。而且从结果来看,原来有的负向调节关系好像没了???