SPSSX Discussion

interpreting simple slopes even if interaction effect is not significant?

Classic

List

Threaded

2 messages Options

psy_vm

interpreting simple slopes even if interaction effect is not significant?

Hello,

Can we interpret simple slopes or Johnson-Neyman even if our interaction effect is not significant? If yes, then how? Also, what do we do with Data for visualizing conditional effect of X on Y?
My results are listed below. Thanks in advance!

*********.
Model = 1
Y = qoltot
X = VisaGrou
M = bcope_ma

Sample size
97

**************************************************************************
Outcome: qoltot

Model Summary
R R-sq F df1 df2 p
.5083 .2584 10.9638 3.0000 93.0000 .0000

Model
coeff se t p LLCI ULCI
constant 68.8666 1.7854 38.5717 .0000 65.3211 72.4121
bcope_ma -.7737 .3217 -2.4048 .0182 -1.4125 -.1348
VisaGrou -12.1130 3.5872 -3.3767 .0011 -19.2366 -4.9895
int_1 1.0766 .6546 1.6445 .1034 -.2234 2.3765

Interactions:

int_1 VisaGrou X bcope_ma

*************************************************************************

Conditional effect of X on Y at values of the moderator(s):
bcope_ma Effect se t p LLCI ULCI
-5.9635 -18.5331 4.8543 -3.8179 .0002 -28.1727 -8.8934
.0000 -12.1130 3.5872 -3.3767 .0011 -19.2366 -4.9895
5.9635 -5.6930 5.7143 -.9963 .3217 -17.0406 5.6546

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.

********************* JOHNSON-NEYMAN TECHNIQUE **************************

Moderator value(s) defining Johnson-Neyman significance region(s):
Value % below % above
3.1219 71.1340 28.8660

Conditional effect of X on Y at values of the moderator (M)
bcope_ma Effect se t p LLCI ULCI
-8.7526 -21.5357 6.2472 -3.4472 .0009 -33.9415 -9.1299
-7.5526 -20.2438 5.6176 -3.6037 .0005 -31.3993 -9.0884
-6.3526 -18.9520 5.0320 -3.7663 .0003 -28.9445 -8.9594
-5.1526 -17.6601 4.5077 -3.9178 .0002 -26.6114 -8.7088
-3.9526 -16.3682 4.0683 -4.0234 .0001 -24.4471 -8.2893
-2.7526 -15.0763 3.7440 -4.0268 .0001 -22.5112 -7.6415
-1.5526 -13.7845 3.5662 -3.8653 .0002 -20.8663 -6.7026
-.3526 -12.4926 3.5571 -3.5120 .0007 -19.5563 -5.4289
.8474 -11.2007 3.7178 -3.0127 .0033 -18.5835 -3.8179
2.0474 -9.9089 4.0280 -2.4600 .0157 -17.9078 -1.9099
3.1219 -8.7521 4.4073 -1.9858 .0500 -17.5041 .0000
3.2474 -8.6170 4.4567 -1.9335 .0562 -17.4672 .2333
4.4474 -7.3251 4.9734 -1.4729 .1442 -17.2012 2.5510
5.6474 -6.0332 5.5534 -1.0864 .2801 -17.0612 4.9947
6.8474 -4.7414 6.1790 -.7673 .4448 -17.0117 7.5290
8.0474 -3.4495 6.8377 -.5045 .6151 -17.0279 10.1289
9.2474 -2.1576 7.5208 -.2869 .7748 -17.0926 12.7773
10.4474 -.8657 8.2223 -.1053 .9164 -17.1936 15.4621
11.6474 .4261 8.9377 .0477 .9621 -17.3225 18.1747
12.8474 1.7180 9.6641 .1778 .8593 -17.4730 20.9090
14.0474 3.0099 10.3990 .2894 .7729 -17.6406 23.6603
15.2474 4.3017 11.1409 .3861 .7003 -17.8219 26.4254

**************************************************************************

Data for visualizing conditional effect of X on Y:
VisaGrou bcope_ma yhat
-.5464 -5.9635 83.6066
.4536 -5.9635 65.0736
-.5464 .0000 75.4850
.4536 .0000 63.3720
-.5464 5.9635 67.3635
.4536 5.9635 61.6705

******************** 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:
VisaGrou bcope_ma

NOTE: Some cases were deleted due to missing data. The number of such cases was:
6

NOTE: All standard errors for continuous outcome models are based on the HC3 estimator

------ END MATRIX -----

David Marso

Re: interpreting simple slopes even if interaction effect is not significant?

Administrator

What does this question have to do with SPSS?
That is not standard SPSS output (aside that somebodies macro spit it out of MATRIX).
Off hand I would say NO. If the effect is NOT sig then you shouldn't be fishing around in the dung looking for a pearl.

psy_vm wrote

Hello,

Can we interpret simple slopes or Johnson-Neyman even if our interaction effect is not significant? If yes, then how? Also, what do we do with Data for visualizing conditional effect of X on Y?
My results are listed below. Thanks in advance!

*********.
Model = 1
Y = qoltot
X = VisaGrou
M = bcope_ma

Sample size
97

**************************************************************************
Outcome: qoltot

Model Summary
R R-sq F df1 df2 p
.5083 .2584 10.9638 3.0000 93.0000 .0000

Model
coeff se t p LLCI ULCI
constant 68.8666 1.7854 38.5717 .0000 65.3211 72.4121
bcope_ma -.7737 .3217 -2.4048 .0182 -1.4125 -.1348
VisaGrou -12.1130 3.5872 -3.3767 .0011 -19.2366 -4.9895
int_1 1.0766 .6546 1.6445 .1034 -.2234 2.3765

Interactions:

int_1 VisaGrou X bcope_ma

*************************************************************************

Conditional effect of X on Y at values of the moderator(s):
bcope_ma Effect se t p LLCI ULCI
-5.9635 -18.5331 4.8543 -3.8179 .0002 -28.1727 -8.8934
.0000 -12.1130 3.5872 -3.3767 .0011 -19.2366 -4.9895
5.9635 -5.6930 5.7143 -.9963 .3217 -17.0406 5.6546

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.

********************* JOHNSON-NEYMAN TECHNIQUE **************************

Moderator value(s) defining Johnson-Neyman significance region(s):
Value % below % above
3.1219 71.1340 28.8660

Conditional effect of X on Y at values of the moderator (M)
bcope_ma Effect se t p LLCI ULCI
-8.7526 -21.5357 6.2472 -3.4472 .0009 -33.9415 -9.1299
-7.5526 -20.2438 5.6176 -3.6037 .0005 -31.3993 -9.0884
-6.3526 -18.9520 5.0320 -3.7663 .0003 -28.9445 -8.9594
-5.1526 -17.6601 4.5077 -3.9178 .0002 -26.6114 -8.7088
-3.9526 -16.3682 4.0683 -4.0234 .0001 -24.4471 -8.2893
-2.7526 -15.0763 3.7440 -4.0268 .0001 -22.5112 -7.6415
-1.5526 -13.7845 3.5662 -3.8653 .0002 -20.8663 -6.7026
-.3526 -12.4926 3.5571 -3.5120 .0007 -19.5563 -5.4289
.8474 -11.2007 3.7178 -3.0127 .0033 -18.5835 -3.8179
2.0474 -9.9089 4.0280 -2.4600 .0157 -17.9078 -1.9099
3.1219 -8.7521 4.4073 -1.9858 .0500 -17.5041 .0000
3.2474 -8.6170 4.4567 -1.9335 .0562 -17.4672 .2333
4.4474 -7.3251 4.9734 -1.4729 .1442 -17.2012 2.5510
5.6474 -6.0332 5.5534 -1.0864 .2801 -17.0612 4.9947
6.8474 -4.7414 6.1790 -.7673 .4448 -17.0117 7.5290
8.0474 -3.4495 6.8377 -.5045 .6151 -17.0279 10.1289
9.2474 -2.1576 7.5208 -.2869 .7748 -17.0926 12.7773
10.4474 -.8657 8.2223 -.1053 .9164 -17.1936 15.4621
11.6474 .4261 8.9377 .0477 .9621 -17.3225 18.1747
12.8474 1.7180 9.6641 .1778 .8593 -17.4730 20.9090
14.0474 3.0099 10.3990 .2894 .7729 -17.6406 23.6603
15.2474 4.3017 11.1409 .3861 .7003 -17.8219 26.4254

**************************************************************************

Data for visualizing conditional effect of X on Y:
VisaGrou bcope_ma yhat
-.5464 -5.9635 83.6066
.4536 -5.9635 65.0736
-.5464 .0000 75.4850
.4536 .0000 63.3720
-.5464 5.9635 67.3635
.4536 5.9635 61.6705

******************** 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:
VisaGrou bcope_ma

NOTE: Some cases were deleted due to missing data. The number of such cases was:
6

NOTE: All standard errors for continuous outcome models are based on the HC3 estimator

------ END MATRIX -----

Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"