When running a simple moderated regression, the statistical output of Process macro is very much the same with 'Step 1 & Step 2' hierarchical regression using SPSS's linear regression tool (regression coeff, p & t-values, mean squared error, R2 & R2 change). The tricky part is to interpret the interaction effect visually so that readers will understand better.
According to Hayes (2013), there are two methods in probing the interaction visually. The Process output tabulates a set of data that needs to be plotted into a graph manually in order to visualize the interaction.
1. Observing the slope of the regression lines using Pick-a-Point approach. The effects of X on Y at -1SD, 0SD and +1SD of the moderator (or using 10th, 25th, 50th, 75th & 90th percentile) are plotted and compared visually. In the Process output, search for the heading titled "Data for visualizing conditional effect of X on Y", you may find a set of data that are divided into three columns. Create a new SPSS dataset, copy-paste the data from the output into the new dataset, setup the variable names, and use the available graph options to plot the data.
Perhaps this YouTube link is very useful:
https://www.youtube.com/watch?v=0vcUtzPDGrw
2. Observing the Johnson-Neyman's region of significance. This is done to determine at which point of M does its interaction effect on the link between X and Y becomes significant.
In the Process output, search for the heading titled "Johnson-Neyman Technique: Conditional effect of X on Y at values of the moderator (M)", you may find a set of data that are divided into seven columns. Copy the data that corresponds to these four columns: ModVar, Effect, LLCI and ULCI. Create a new SPSS dataset, paste the data from the output into the new dataset, setup the variable names, and use the graph builder to plot the data.
======
I recommend you to read the book titled "Introduction to Mediation, Moderation and Conditional Process Analysis: A regression based approach" by Andrew Hayes (2013) and published by Guilford Press.
Hope this helps :)
If you download spss v22 in your pc. You go to help spss and you clic satistic basic and go in cases study. You find many exempls. You can downald a book for Andy field (DISCOVERD STATISTIC WITH Spss).
Hi,
Good Evening, Kindly go through the following links:
http://mres.gmu.edu/pmwiki/uploads/Main/process.pdf
http://imaging.mrc-cbu.cam.ac.uk/statswiki/FAQ/SobelTest?action=AttachFile&do=get&target=process.pdf
http://www.statisticalhorizons.com/seminars/public-seminars/medandmod14
Leila,
Try this link
http://bit.ly/1HPEsyH
It contains several articles which used PROCESS as the analytical frame.
Silburn
When running a simple moderated regression, the statistical output of Process macro is very much the same with 'Step 1 & Step 2' hierarchical regression using SPSS's linear regression tool (regression coeff, p & t-values, mean squared error, R2 & R2 change). The tricky part is to interpret the interaction effect visually so that readers will understand better.
According to Hayes (2013), there are two methods in probing the interaction visually. The Process output tabulates a set of data that needs to be plotted into a graph manually in order to visualize the interaction.
1. Observing the slope of the regression lines using Pick-a-Point approach. The effects of X on Y at -1SD, 0SD and +1SD of the moderator (or using 10th, 25th, 50th, 75th & 90th percentile) are plotted and compared visually. In the Process output, search for the heading titled "Data for visualizing conditional effect of X on Y", you may find a set of data that are divided into three columns. Create a new SPSS dataset, copy-paste the data from the output into the new dataset, setup the variable names, and use the available graph options to plot the data.
Perhaps this YouTube link is very useful:
https://www.youtube.com/watch?v=0vcUtzPDGrw
2. Observing the Johnson-Neyman's region of significance. This is done to determine at which point of M does its interaction effect on the link between X and Y becomes significant.
In the Process output, search for the heading titled "Johnson-Neyman Technique: Conditional effect of X on Y at values of the moderator (M)", you may find a set of data that are divided into seven columns. Copy the data that corresponds to these four columns: ModVar, Effect, LLCI and ULCI. Create a new SPSS dataset, paste the data from the output into the new dataset, setup the variable names, and use the graph builder to plot the data.
======
I recommend you to read the book titled "Introduction to Mediation, Moderation and Conditional Process Analysis: A regression based approach" by Andrew Hayes (2013) and published by Guilford Press.
Hope this helps :)
Could any one please recommend a research paper that reports Hayes model 3 output.
first your need to install the extension provided Hayes either the process or the modprobe. Its important to know the concept of interaction with the type or nature of variables. you might use continuous variables and they categorized into dichotomous variable (i.e. Hi-Low group) or you might use naturally occurring dichotomous data (i.e. Gender). the interpretation of interaction should depend on the nature of the type of variables. Modprobe is basically designed to carry out the interactions. however, you can also use process and model 1 to perform a simple interaction. once you specify the variables then results would generate to include the interaction term. if you are running interaction between continuous variable then make sure to tick the mean centering option. you can also tick for generating data for interaction plot to be produced using the spss syntax. it is very important to provide the interaction plot in moderation analysis. also choose J-N method for continuous variables and pick a point method for categorical variables. i have found the J-N method very useful as it provide additional insight beyond the interaction plot. it might also tell you where actually the interaction occurs. it is also important provide the effect size for the interaction based R sq. you can look into Cohen' work on size effect.
Please I need help, i used Haye's process tool to call out a moderation analysis on my variables, but on the graph i got a downward slop of interaction at the low, average and high level of the moderator. Please how do i interpret this?
Ihave a sample of interpretation but in frensh language,if you need this contact me
http://doctorants-chercheurs.blogspot.com/2018/02/solved-how-do-you-interpret-moderation.html
Hello
Please refer to,
Landau, S. and Everitt, B. S.(2004). A Handbook of Statistical analyses using SPSS.
Regards,
Zuhair
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4469902/
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis