# Help to interpret bootstrap and sobel results from spss

#### gi.yair

##### New Member
Hi,
I'm trying to understand the meaning of these results -
my hypothesis is that EA mediates between CNFU and PI.
Is the mediation full or partial?
What does the insignificance of CNFU mean?
What else can I learn from these results?

VARIABLES IN SIMPLE MEDIATION MODEL
Y ave_pi
X ave_CNFU
M ave_EA

DESCRIPTIVES STATISTICS AND PEARSON CORRELATIONS
Mean SD ave_pi ave_CNFU ave_EA
ave_pi 3.8500 1.5745 1.0000 .3320 .6130
ave_CNFU 2.9624 1.2504 .3320 1.0000 .4234
ave_EA 3.1118 1.4924 .6130 .4234 1.0000

SAMPLE SIZE
152

DIRECT And TOTAL EFFECTS
Coeff s.e. t Sig(two)
b(YX) .4180 .0970 4.3100 .0000
b(MX) .5054 .0883 5.7245 .0000
b(YM.X) .6074 .0750 8.0993 .0000
b(YX.M) .1110 .0895 1.2407 .2167

INDIRECT EFFECT And SIGNIFICANCE USING NORMAL DISTRIBUTION
Value s.e. LL 95 CI UL 95 CI Z Sig(two)
Effect .3070 .0660 .1776 .4363 4.6512 .0000

BOOTSTRAP RESULTS For INDIRECT EFFECT
Data Mean s.e. LL 95 CI UL 95 CI LL 99 CI UL 99 CI
Effect .3070 .3064 .0669 .1826 .4436 .1396 .4932

NUMBER OF BOOTSTRAP RESAMPLES
5000

POINT ESTIMATES OF EFFECT SIZE FOR INDIRECT EFFECTS
ab .3070
P_m .7343
R_m 2.7641
R2_45 .1038
ab_ps .1950
ab_cs .2438

Thanks!

#### CB

##### Super Moderator
I'm trying to understand the meaning of these results -
my hypothesis is that EA mediates between CNFU and PI.
Is the mediation full or partial?
What does the insignificance of CNFU mean?...

DIRECT And TOTAL EFFECTS
Coeff s.e. t Sig(two)
b(YX) .4180 .0970 4.3100 .0000
b(MX) .5054 .0883 5.7245 .0000
b(YM.X) .6074 .0750 8.0993 .0000
b(YX.M) .1110 .0895 1.2407 .2167
I haven't used this procedure before so the output is a little unfamiliar, but this looks like you have a case of partial mediation per David Kenny's revised criteria for mediation. The relationship between X (CNFU) and Y (PI) is reduced after controlling for M (EA), and in fact becomes non-significant, but is not reduced to zero (which would be required for a claim of complete mediation).