Hallo,
Es geht um folgendes. Meine Haupthypothese ist, dass potentiell traumatische Ereignisse (X = PTEs ges) einen Einfluss auf den subjektiv empfundenen Disstress während der Covid-19 Pandemie hat (y = IESR Sum). Ich hypothesiere also, dass eine Person, die mehr belastende Ereignisse erlebt hat, sensibler auf die Pandemie reagiert. Das ist meine eigentliche Frage.
Eine Unterfrage ist, ob dieser Zusammenhang zwischen belastenden Ereignissen und Disstress durch die Resilienz einer Person ( W = RS Sum) moderiert wird. Wie du in der unten stehenden Tabelle erkennen kannst, gibt es eine signifikante Interaktion.
Meine Frage an dich lautet nun, ob ich den Haupteffekt von Ereignissen (PTEs ges) auf Disstress (IESR Sum) eigenständig und unabhängig von der Moderation interpretieren darf. In deinem Tutorial steht, dass man das NICHT darf, wenn die Mittelwerte nicht zentriert wurden. Hier wurden sie aber zentriert.
Meine Interpretation wäre nun, dass Ereignisse einen eigenständigen positiven Effekt auf Disstress haben (mehr Ereignissen führen zu mehr Disstress). Dieser Effekt ist aber umso stärker, je höher die Resilienz einer Person ist (Moderationshypothese). Somit Wären beide Hypothesen bestätigt. Zusätzlicher Befund wäre, dass Resilienz einen eigenständigen, negativen Haupteffekt zu auf Disstress zu haben (mehr Resilienz führt zu weniger Disstress).
Ist das korrekt?
Glg
Rick
Run MATRIX procedure:
***************** PROCESS Procedure for SPSS Version 4.1 *****************
Written by Andrew F. Hayes, Ph.D.
http://www.afhayes.com Documentation available in Hayes (2022).
http://www.guilford.com/p/hayes3**************************************************************************
Model : 1
Y : IESR_Sum
X : PTEs_ges
W : RS_Sum
Covariates:
sex alter
Sample
Size: 1071
**************************************************************************
OUTCOME VARIABLE:
IESR_Sum
Model Summary
R R-sq MSE F(HC3) df1 df2 p
,3511 ,1233 533,2585 30,5629 5,0000 1065,0000 ,0000
Model
coeff se(HC3) t p LLCI ULCI
constant 33,2379 2,1599 15,3890 ,0000 28,9998 37,4759
PTEs_ges ,4097 ,1180 3,4713 ,0005 ,1781 ,6413
RS_Sum -,5230 ,0735 -7,1154 ,0000 -,6672 -,3787
Int_1 -,0235 ,0105 -2,2479 ,0248 -,0440 -,0030
sex -11,3603 1,5110 -7,5184 ,0000 -14,3252 -8,3954
alter ,0044 ,0459 ,0955 ,9240 -,0856 ,0944
Product terms key:
Int_1 : PTEs_ges x RS_Sum
Test(s) of highest order unconditional interaction(s):
R2-chng F(HC3) df1 df2 p
X*W ,0051 5,0531 1,0000 1065,0000 ,0248
----------
Focal predict: PTEs_ges (X)
Mod var: RS_Sum (W)
Conditional effects of the focal predictor at values of the moderator(s):
RS_Sum Effect se(HC3) t p LLCI ULCI
-11,0188 ,6687 ,1772 3,7729 ,0002 ,3209 1,0165
,0000 ,4097 ,1180 3,4713 ,0005 ,1781 ,6413
11,0188 ,1507 ,1516 ,9938 ,3205 -,1468 ,4483
Moderator value(s) defining Johnson-Neyman significance region(s):
Value % below % above
6,7587 68,2540 31,7460
Conditional effect of focal predictor at values of the moderator:
RS_Sum Effect se(HC3) t p LLCI ULCI
-47,9533 1,5369 ,5326 2,8855 ,0040 ,4918 2,5820
-44,8533 1,4640 ,5010 2,9220 ,0036 ,4809 2,4471
-41,7533 1,3911 ,4696 2,9626 ,0031 ,4698 2,3125
-38,6533 1,3183 ,4382 3,0081 ,0027 ,4584 2,1782
-35,5533 1,2454 ,4071 3,0593 ,0023 ,4466 2,0442
-32,4533 1,1725 ,3761 3,1172 ,0019 ,4345 1,9106
-29,3533 1,0997 ,3455 3,1831 ,0015 ,4218 1,7776
-26,2533 1,0268 ,3152 3,2581 ,0012 ,4084 1,6452
-23,1533 ,9539 ,2853 3,3438 ,0009 ,3941 1,5137
-20,0533 ,8811 ,2560 3,4411 ,0006 ,3787 1,3835
-16,9533 ,8082 ,2277 3,5499 ,0004 ,3615 1,2549
-13,8533 ,7353 ,2005 3,6673 ,0003 ,3419 1,1288
-10,7533 ,6625 ,1752 3,7821 ,0002 ,3188 1,0062
-7,6533 ,5896 ,1525 3,8661 ,0001 ,2904 ,8889
-4,5533 ,5167 ,1339 3,8583 ,0001 ,2539 ,7795
-1,4533 ,4439 ,1213 3,6589 ,0003 ,2058 ,6819
1,6467 ,3710 ,1166 3,1816 ,0015 ,1422 ,5998
4,7467 ,2981 ,1208 2,4691 ,0137 ,0612 ,5351
6,7587 ,2508 ,1278 1,9622 ,0500 ,0000 ,5017
7,8467 ,2253 ,1329 1,6949 ,0904 -,0355 ,4861
10,9467 ,1524 ,1512 1,0082 ,3136 -,1442 ,4490
14,0467 ,0795 ,1736 ,4582 ,6469 -,2611 ,4202
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/
PTEs_ges RS_Sum IESR_Sum .
BEGIN DATA.
-6,5950 -11,0188 30,9982
,0000 -11,0188 35,4084
6,5950 -11,0188 39,8185
-6,5950 ,0000 26,9439
,0000 ,0000 29,6460
6,5950 ,0000 32,3480
-6,5950 11,0188 22,8896
,0000 11,0188 23,8836
6,5950 11,0188 24,8775
END DATA.
GRAPH/SCATTERPLOT=
PTEs_ges WITH IESR_Sum BY RS_Sum .
**************************************************************************
Bootstrap estimates were saved to a file
Map of column names to model coefficients:
Conseqnt Antecdnt
COL1 IESR_Sum constant
COL2 IESR_Sum PTEs_ges
COL3 IESR_Sum RS_Sum
COL4 IESR_Sum Int_1
COL5 IESR_Sum sex
COL6 IESR_Sum alter
*********** BOOTSTRAP RESULTS FOR REGRESSION MODEL PARAMETERS ************
OUTCOME VARIABLE:
IESR_Sum
Coeff BootMean BootSE BootLLCI BootULCI
constant 33,2379 33,3151 2,1519 29,0846 37,4279
PTEs_ges ,4097 ,4127 ,1186 ,1854 ,6620
RS_Sum -,5230 -,5208 ,0706 -,6606 -,3837
Int_1 -,0235 -,0235 ,0104 -,0430 -,0027
sex -11,3603 -11,3830 1,5112 -14,2613 -8,3741
alter ,0044 ,0031 ,0460 -,0864 ,0932
*********************** 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: A heteroscedasticity consistent standard error and covariance matrix estimator was used.
NOTE: The following variables were mean centered prior to analysis:
RS_Sum PTEs_ges
WARNING: Variables names longer than eight characters can produce incorrect output
when some variables in the data file have the same first eight characters. Shorter
variable names are recommended. By using this output, you are accepting all risk
and consequences of interpreting or reporting results that may be incorrect.
------ END MATRIX -----