Hi Katharina,

as you use SPSS/Process, you incorporate a manifest product term. This downward biases the interaction effect and reduces the power to find a significant effect (as the product term is very unreliable), see

McClelland, G. H., & Judd, C. M. (1993). Statistical difficulties of detection interactions and moderator effects. Psychological Bulletin, 114, 376-389.

Jaccard, J., & Wan, C. K. (1995). Measurement error in the analysis of interaction effects between continuous predictors using multiple regression: Multiple indicator and structural equation approaches. Psychological Bulletin, 117(2), 348-357.

Busemeyer, J., & Jones, L. (1983). Analysis of multiplicative combination rules when the causal variables are measured with error. Psychological Bulletin, 93, 549-562.

Hence, one of the reasons for an non-sign. moderator effect may simply be "low power". As a remedy, there is a long lasting discusion how to set up those efffects in structural equation modeling, e.g.,

Moosbrugger, H., Schermelleh-Engel, K., & Klein, A. (1998). Methodological problems of estimating latent interaction effects. Psychological Research, 2(2), 95-108.

Little, T. D., Bovaird, J. A., & Widaman, K. F. (2006). On the merits of orthogonalizing powered and product terms: Implications for modeling interactions among latent variables. Structural Equation Modeling, 13(4), 497-519.

Moosbrugger, H., Frank, D., & Schermelleh-Engel, K. (1991). Zur Überprüfung von latenten Moderatoreffekten mit linearen Strukturgleichungsmodellen. Zeitschrift für Differentielle und Diagnostische Psychologie, 12(245-255).

Perhaps, going this road would be promising for you. By using the lavaan-package in R, you can easily set up such a model (see the lavaan.org-page for tutorials. I have a German book about SEM in R for beginners.

But of course, interactions are simply seldom although we constantly think in terms of "it depends" :) That is, you do not find a moderator effect because their is none.

If you "overall model is significant" then I think you mean the F-test of the R-square, right? That only shows that the main effects of the IVs are significant.

HTH

Holger (@HolgerSteinmetz)

Need help

If I am employing moderation in secondary time series analysis can I need to proceed with the methodology of time series?

I need some liturature regarding its methodology

Actually my all variables (quantitative) are of mixed order can I still procedd with ARDL approach irrespective of moderation analysis?

If someone guide me / suggest some relavent stuff I shall be very grateful for this favor.