@@researchwithfawad Thank you for your answer. Do you have another tip for the visualization? I have tried it with the given Excel sheet, but I am looking for a way to generate a graph in which the metric predictor is placed on the x-axis (M - 1Sd, Mean, M + 1Sd), the dependent variable on the y-axis and two regression lines are generated for the categorical predictor (e.g. 2 training conditions).
@@researchwithfawad thanks for the link. I had the same question above. Kindly wanted to know can if I can do all moderation analysis at once with one dichotomous IV and 1 continuous moderator and 4 continuous DV or separately for each analysis?
I have a question regarding the signs in the interaction term and the effects. At one point you said that the negative sign in the interaction term means that the relationship between IV and DV is weakened by the moderator. How do we interpret the effects then? I have a negative interaction and negative effects. My IV is stress, DV is well-being and M is self-regulation. Could you please explain it further? Thank you!
The negative sign would mean that it weakens the relationship between IV and DV. If your moderator has a negative sign, and the relationship between IV and DV is also negative. This would mean the hte moderator weakens the negative relationship between IV and DV.
@@researchwithfawad Thank you for your reply. Could you please specify what the sign in the effects tells us? I've seen examples, where the interaction term is negative, yet the effects have positive values. In my case, both interaction and effects have negative values, and I'm wondering what it tells me. What is the general rule for interpretation here? Also, my effects decrease mathematically: when my moderator is below 1SD it is -1, then above 1 SD it's -2. So the effects become weaker or stronger with an increase in the moderator value? I'd really appreciate your detailed explanation because I got a bit lost here. Thanks!
Thanks for your great explanation, how could i calculate and decide on model fit using SPSS before model 1 procecess macro for moderation with covariates,?
Thank you for this tutorial. It really helped me. Could I proceed the same way if IV and DV are continuous variables (likert scale), and the moderator is a quantitative variable: length of stay (kind of continuous variable)? Thank you
Sir, very well explained. In my model, the independant variable is categorical (car driven: imported vs. local ) and so is the mediating variable (type of road: city vs. highway). I am planning to use pls-sem for the analysis.. is it ok?
Thank you for the informative video...How to do moderation when we have have multiple ( for example 3 to 4) DVs ? Do we need to test each of them separately?
I am afraid, you can only have 1 dv in Hayes. You have to do it separately. Otherwise you can use IBM SPSS AMOS. Here is a complete Playlist. th-cam.com/play/PLb7vm6tsQ3Kvinj9MFF6wbQFDBdkU1ERP.html
There are different models each based on the type of theoretical framework a researcher would like to test. You can search google for Hayes Process Macro Models
Explained very well. Can you please discuss the sign of interaction value. In current case the moderator (as per literature and calculated correlations) is having a negative relation with the DV and positive relation with IV. However, moderation interaction value turns out to be positive. How can this be addressed??
Thanks. For moderation effect, assess the interaction effect. The sign will guide whether the moderator strengthens or weakens the relationships. Positive mean that moderator strengthens the relationship whereas negative would mean it weaken the relationship. Further, if it is significantly it is strongly recommended to perform slope analysis.
thank you for great explainantion, i am running this process but grapgh i am getting is so much dispursed and having too much values can you help me solving this problem.
Dear sir. What is the interpretation if the interaction effect is significant. And only lower level of moderator is significant and not higher level of moderator. Do we still need to draw the intercation graph.
Hi Sir, this video has been very informative. I really enjoyed and would like to appreciate your professional assistance. However, for the sake of complete understanding, can you please attach the data you have used so that we can try and check out against the results for mastery of it? Thanks
Thank you so much for the great explanation. Was really useful. I wanted to clarify the interpretation of the results using the gaskination package. In the video, you did'nt enter the value of the intercept, and it was kept at 3. Is this what we need to keep it as (3) while doing own analysis or do we need to add the model-specific intercept in the gaskination tool?
Thank you so much for your videos! Quick question: The whole procedure works even if my moderator + DV are continuous and the ID is categorical? Thank you!
Thank you for your explanation!! I have a question... If running previous linear regression analysis my X variable is not significant related to Y, does it make sense to carry out moderation analysis? I have been trying to find a reference book that could answer me this question as I'm starting with research :)
Thank you for your tutorial! Can I ask a question? If change in R2 due to interaction is not statistically significant but the interaction is significant, can I still say that there is a moderation effect?
Thank you for the insightful video. Quick question, can one test for moderation even though a correlation test found that there is no significant relationship between two variables?
Hello Mr. Fawad, first and foremost, thank you for all wonderful videos. You are really enhancing the quality of my thesis. Two questions for you. First, when I have two IV's with the same moderator to 1 DV, can I just test them separately? Second, what would you include in a table to visualize the effect?
Pleasure. I am glad you liked the video. No, text the model as a whole and for reporting, include the effects, IV to DV, MOD to DV, and Interaction to DV, R-Sq with and without the include of the interaction effect.
@@MHagen-v3u @ResearchWithFawad Dear Mr. Fawad, thank you for this clear explanation! I have the same question. I have two IV's and now I tested the moderation separately for both IV's. How can we test the model as a whole in SPSS?
Mr Fawad, if i were to put control variables in the model, is it sufficient to put it in the covariate box? or are there anything else I should take into account? Thank you love your videos
Thank you for your video!! Can I ask a question: I did some analysis in model 1, and the interaction term is not significant, but the model summary shows that it is significant, how do I interpret the result? Does it mean that there's no moderation but there is a relationship between the IV and the DV without considering the moderating effect of the moderator variable (MV). In other words, the IV has a main effect on the DV?
This means overall the model is significant however the moderating effect is insignificant. This shows the moderator does not moderate the relationship.
You can download sample data from Data in Brief. Please watch th-cam.com/video/ePY8YAxmoek/w-d-xo.html You can also download the sample data from this link researchwithfawad.com/index.php/download-research-resources/
Thank you Mr. Fawad great explanation. Could I do the same if I have an IV of three categories (small, medium, high)? If I create a dummy variable for Large and Medium and set one as my IV and the other as a covariate may I interpret it then?
Thanks for the informative video:) May I ask, if I have two moderators, would I carry out this analysis two times, or would I have to use model 2 instead? The moderators are not assumed to be related/to moderate each other (IV-MOD1-DV; IV-MOD2-DV)
You can use model 1, the output will change as now you have a binary outcome variable. This video may also help in developing some understanding of the output th-cam.com/video/tlslGucrD_c/w-d-xo.html
Thank you for this! Do you know if PROCESS can test the interaction of one moderator across 8 IVs and one DV simultaneously? Or would it be better to create the mean-centered variables and test together manually? I am using SPSS if that matters.
Although, they have both similarities and differences (Covariates and Controls). You can add control as covariates in Hayes Process Macro but issue is that then the variable is affecting all the outcome variables not just the criterion variable.
May Allah bless you, Dear Professor. Thank you. highly appreciated
You are most welcome
saved my life, thank you so much
Pleasure. I am glad that it helped.
Great explanation, could be please elaborate why you specifically mentioned no zeros in the upper and lower limit of interaction effect?
Thanks. I am glad you liked it. If there is no 0 between upper and lower CIs, this shows that results are significant l.
you’re a legend! Thanks
You're welcome!
That was a great explanation. Thank you prof.
Pleasure. I am glad you liked it.
Thank you a lot for your tutorial. I have a question about the Gaskin sheet. Why the value of the "constant" is 3? Thank you
Please email Prof. Gaskin.
Thank you for the informative video. Is it still possible to interpret a moderation effect (b2) even if the overall model does not become significant?
Thanks for watching. Yes, you can assess it.
@@researchwithfawad Thank you for your answer. Do you have another tip for the visualization? I have tried it with the given Excel sheet, but I am looking for a way to generate a graph in which the metric predictor is placed on the x-axis (M - 1Sd, Mean, M + 1Sd), the dependent variable on the y-axis and two regression lines are generated for the categorical predictor (e.g. 2 training conditions).
Thank you for this -- do you have a video with a dichotomous IV, continuous moderator, and continuous DV?
Pleasure. This video may help, it has categorical IV, But categorical moderator.
th-cam.com/video/TNNK_l3jP0w/w-d-xo.html
Hopefully i can make one soon with Categorical IV, Moderator and Continuous DV.
@@researchwithfawad thanks for the link. I had the same question above. Kindly wanted to know can if I can do all moderation analysis at once with one dichotomous IV and 1 continuous moderator and 4 continuous DV or separately for each analysis?
I have a question regarding the signs in the interaction term and the effects. At one point you said that the negative sign in the interaction term means that the relationship between IV and DV is weakened by the moderator. How do we interpret the effects then? I have a negative interaction and negative effects. My IV is stress, DV is well-being and M is self-regulation. Could you please explain it further? Thank you!
The negative sign would mean that it weakens the relationship between IV and DV. If your moderator has a negative sign, and the relationship between IV and DV is also negative. This would mean the hte moderator weakens the negative relationship between IV and DV.
@@researchwithfawad Thank you for your reply. Could you please specify what the sign in the effects tells us? I've seen examples, where the interaction term is negative, yet the effects have positive values. In my case, both interaction and effects have negative values, and I'm wondering what it tells me. What is the general rule for interpretation here? Also, my effects decrease mathematically: when my moderator is below 1SD it is -1, then above 1 SD it's -2. So the effects become weaker or stronger with an increase in the moderator value? I'd really appreciate your detailed explanation because I got a bit lost here. Thanks!
It is very useful for me. Thank you so much Sir.
Thanks. I am glad you liked it
Thank you for the video. The SmartPLS can't make this? i tried and the numbers seems different and also the T never pop up negative.
Thanks for watching. Negative sign shows inverse relationship.
@@researchwithfawadthank you for ur answer
Thanks for your great explanation, how could i calculate and decide on model fit using SPSS before model 1 procecess macro for moderation with covariates,?
Thanks for watching. For model fit indices, please use AMOS.
Thank you for this tutorial. It really helped me. Could I proceed the same way if IV and DV are continuous variables (likert scale), and the moderator is a quantitative variable: length of stay (kind of continuous variable)? Thank you
Pleasure. I am glad you liked it. Yes, you can.
Sir, very well explained. In my model, the independant variable is categorical (car driven: imported vs. local ) and so is the mediating variable (type of road: city vs. highway). I am planning to use pls-sem for the analysis.. is it ok?
Yes
Thank you for the informative video...How to do moderation when we have have multiple ( for example 3 to 4) DVs ? Do we need to test each of them separately?
I am afraid, you can only have 1 dv in Hayes. You have to do it separately. Otherwise you can use IBM SPSS AMOS. Here is a complete Playlist.
th-cam.com/play/PLb7vm6tsQ3Kvinj9MFF6wbQFDBdkU1ERP.html
Thank you
Thank you very much. Can i ask why it is model 1 chosen?
There are different models each based on the type of theoretical framework a researcher would like to test. You can search google for Hayes Process Macro Models
Thanks for the video. I have a question. WHat about when you have mutiple moderators? let's say five.
Pleasure. I am glad you liked it. In that case, use SEM tools like SmartPLS or AMOS
Explained very well.
Can you please discuss the sign of interaction value.
In current case the moderator (as per literature and calculated correlations) is having a negative relation with the DV and positive relation with IV.
However, moderation interaction value turns out to be positive.
How can this be addressed??
Thanks. For moderation effect, assess the interaction effect. The sign will guide whether the moderator strengthens or weakens the relationships. Positive mean that moderator strengthens the relationship whereas negative would mean it weaken the relationship. Further, if it is significantly it is strongly recommended to perform slope analysis.
@@researchwithfawad thankyou Sir.
thank you for great explainantion, i am running this process but grapgh i am getting is so much dispursed and having too much values can you help me solving this problem.
What is the issue?
Dear sir. What is the interpretation if the interaction effect is significant.
And only lower level of moderator is significant and not higher level of moderator. Do we still need to draw the intercation graph.
This shows significant moderation.
wonderful and informative. Can you please tell the book you are using for teaching Process Macros.
Well i am not using any. Doing it in my own.
Hi Sir, this video has been very informative. I really enjoyed and would like to appreciate your professional assistance. However, for the sake of complete understanding, can you please attach the data you have used so that we can try and check out against the results for mastery of it? Thanks
Thanks for watching. The data is part of an ongoing study and cannot be shared.
Thank you so much for the great explanation. Was really useful. I wanted to clarify the interpretation of the results using the gaskination package. In the video, you did'nt enter the value of the intercept, and it was kept at 3. Is this what we need to keep it as (3) while doing own analysis or do we need to add the model-specific intercept in the gaskination tool?
Thanks. I am glad you liked it. You can change but it does do much for the interpretation of slope. Only the location of the gradients is changed.
I have model that is more similar to model 1 but i have 2 IVs can i use this model
For more than 1 IV, use SEM tools like SmartPLS or AMOS.
Can the Hayes macro solve problems with more than 1 independent model?
No. for that please use SEM tools.
Thank you so much for your videos!
Quick question: The whole procedure works even if my moderator + DV are continuous and the ID is categorical?
Thank you!
Yes it will
Thank you for your explanation!! I have a question... If running previous linear regression analysis my X variable is not significant related to Y, does it make sense to carry out moderation analysis? I have been trying to find a reference book that could answer me this question as I'm starting with research :)
Yes, please check moderation.
Sir! What if the interaction term has p value of .046? Do we consider the moderation significant or nonsignificant? Kindly address this query.
It is significant
@@researchwithfawad thank you Sir!
Thank you for your tutorial! Can I ask a question? If change in R2 due to interaction is not statistically significant but the interaction is significant, can I still say that there is a moderation effect?
Yes, you will check for the significance of interaction effect.
Sir why in my output theres no neyman Johnson results inspite of checking it before performing the analysis?
Check if you are getting any warning when running the analysis.
I don't see any warning messages, is this necessary Sir?
If the moderation effect isn't significant, so, should we still need to draw the interaction graph? Please advise.
No, interaction shall only be plotted if the moderation is significant.
Thank you for the insightful video. Quick question, can one test for moderation even though a correlation test found that there is no significant relationship between two variables?
Pleasure. Yes, give it a go.
Hello Mr. Fawad, first and foremost, thank you for all wonderful videos. You are really enhancing the quality of my thesis. Two questions for you. First, when I have two IV's with the same moderator to 1 DV, can I just test them separately? Second, what would you include in a table to visualize the effect?
Pleasure. I am glad you liked the video. No, text the model as a whole and for reporting, include the effects, IV to DV, MOD to DV, and Interaction to DV, R-Sq with and without the include of the interaction effect.
@@researchwithfawad Thank you for coming back to me. How could I test two IVs, with one common moderator in Process?
@@MHagen-v3u @ResearchWithFawad Dear Mr. Fawad, thank you for this clear explanation! I have the same question. I have two IV's and now I tested the moderation separately for both IV's. How can we test the model as a whole in SPSS?
sry sir i am unable to download the stats tool package from the link
You can google it and visit statwiki
Great job sir
Thanks. I am glad you liked it.
My version of macro is 4.1, I've followed your steps but this part "conditional effects of focal predictor " doesn't show up
Make sure you have selected the right model and have added the variables where they should have been.
Mr Fawad, if i were to put control variables in the model, is it sufficient to put it in the covariate box? or are there anything else I should take into account? Thank you love your videos
Yes you can put in covariates.
Thank you for your video!! Can I ask a question: I did some analysis in model 1, and the interaction term is not significant, but the model summary shows that it is significant, how do I interpret the result? Does it mean that there's no moderation but there is a relationship between the IV and the DV without considering the moderating effect of the moderator variable (MV). In other words, the IV has a main effect on the DV?
This means overall the model is significant however the moderating effect is insignificant. This shows the moderator does not moderate the relationship.
Hi, how can I interpret the main effect (IV --> DV) with this output?
Please check for the p value. A p value less than 0.05 would mean there is significant impact.
from where can we access this data set you are currently using?
You can download sample data from Data in Brief. Please watch
th-cam.com/video/ePY8YAxmoek/w-d-xo.html
You can also download the sample data from this link
researchwithfawad.com/index.php/download-research-resources/
@@researchwithfawad this sample data is not avalible in both links ,can you provide the name of research you used for this video
Thank you for this video :) Can I ask a question? The results of the analysis of some variables error. "Error encountered in source line #110359
>Error # 12417
>Source operand is singular for INV. What is the cause?
Thanks for watching. Did you the error in SPSS. Is your INV binary, double checknif it has one value
Thank you so much Sir@@researchwithfawad
Dear Fawad..how do we find the "composite/consolidated variable' value from the given values of different items within each construct
Thanks for your comment.
Take the mean of the items in a construct.
hello, i need to report my results in apa style. which symbol should i denote effect? is it d? i couldnt find anywhere thank you!
In APA style, the common symbol used to denote effect size is "d".
Can u please make a video on disordinal moderation
Hopefully soon. Thanks for the suggestion.
Thank you Mr. Fawad great explanation. Could I do the same if I have an IV of three categories (small, medium, high)? If I create a dummy variable for Large and Medium and set one as my IV and the other as a covariate may I interpret it then?
Thanks. I am glad you liked it. No need to create dummy variables for ordinal variables.
Pls will I be wrong if I mean center my continous variables (iv & moderator) wen using process macro?
No need to do it manually, as there is an option to do it from the Process Macro.
What can I do if my moderator and IV are categorical? Only my DV is continuous
Then you may opt for Multi group analysis to test effect of Moderator and the independent is dummy coded.
Thanks for the informative video:) May I ask, if I have two moderators, would I carry out this analysis two times, or would I have to use model 2 instead? The moderators are not assumed to be related/to moderate each other (IV-MOD1-DV; IV-MOD2-DV)
Pleasure. I am glad you liked it.
Please use Model 2. Here is a link to a video session
th-cam.com/video/vKTf95s5DGY/w-d-xo.html
Kindly tell me what does it mean if there is 0 in LLCI and ULCI.
In simple terms if there is a 0 between LLCI and ULCI, this means that your hypothesis is not supported.
@@researchwithfawad Thanks a lot.
what kind of moderation analysis can I use if my IV and Moderator are continuous and my DV is categorical?
You can use model 1, the output will change as now you have a binary outcome variable.
This video may also help in developing some understanding of the output
th-cam.com/video/tlslGucrD_c/w-d-xo.html
@@researchwithfawad thank you!
Hi Fawad. Is Hayes Process Macro only for latent variables? Or is it possible to use it with observed variables as well?
Hi. Yes, Hayes Macro is only for composite variables computed through the observed variables.
Dear sir, add some content on fsQCA and combined PLS-SEM & fsQCA
Hopefully soon
@@researchwithfawad 🙏
Thank you for this! Do you know if PROCESS can test the interaction of one moderator across 8 IVs and one DV simultaneously? Or would it be better to create the mean-centered variables and test together manually? I am using SPSS if that matters.
Thanks. I am glad you liked it.
Hayes Process Macro only allows one IV. If you have more, please use SMART-PLS or AMOs
But can we do it separately like if we have 7-8 IV's can we individually enter the data nd find out the link of that IV with moderator?
where can we include control variables?
Although, they have both similarities and differences (Covariates and Controls). You can add control as covariates in Hayes Process Macro but issue is that then the variable is affecting all the outcome variables not just the criterion variable.
@@researchwithfawad thank you for your response