Here's a fun pet project I've been working on: udreamed.com/. It is a dream analytics app. Here is the TH-cam channel where we post a new video almost three times per week: th-cam.com/channels/iujxblFduQz8V4xHjMzyzQ.html Also available on iOS: apps.apple.com/us/app/udreamed/id1054428074 And Android: play.google.com/store/apps/details?id=com.unconsciouscognitioninc.unconsciouscognition&hl=en Check it out! Thanks!
Literally the single best explanation of mediating vs. moderating variables. I've struggled through a whole year of my PhD trying to get this. And within 5 minutes, now I get it! THANK YOU THANK YOU
James, you will be the reason I will be able to finish writing my Master's thesis proposal today. No video or teacher or lecture slides have been able to help me learn this ever. I wish you many many subscribers, likes and whatever you need to be happy in life. Thank you. For others, if there are the following variables: 1. motivation to learn about moderators and mediators 2. ease of learning about moderators and mediators 3. efforts in searching for James' YT channel 4. presence of other crappy videos on the same topic Then, "motivation to learn about moderators and mediators" is the IV, "ease of learning about moderators and mediators" is the DV, "efforts in searching for James' YT channel" is the mediator, and "presence of other crappy videos on the same topic" is the moderator.
Mr Gaskin, thank you for this explanation. As narrow as the focus of the video may seem, it has influenced my entire life in altering how I understand cause and effect, and will benefit every aspect of every function in which I serve. I cannot thank you enough for this.
I have spent numerous hours, read several books, and watched many videos trying to figure this out. This was by far the most beneficial explanation. Thank you.
After years of trying to find a simple way to understand these two concepts I finally get it! Thanks a bunch. If I got it correctly this time, the satisfaction with my learning after watching your video was the mediator that made me sign up to your channel ;)
FINALLY get this. The value of just explaining something in simplistic, understandable terms can’t be overstated. Reading the comments I know I’m not the only one! Thank you! 🙌🏻
James, I appreciate this explanation more than you know. While I've had 20 different explanations of this concept and do understand what they mean, this video helps me understand the criteria. Great explanation!
Thank you so much for this. After hours of googling, reading, and watching numerous explanations about the topics, I can now have a break because of your video. Very much appreciated :)
This helped me so much, THANKS to you I can finally wrap my hand around this and get cracking . I was all over the place constantly switching variables, so lost. This video clarified everything!
Sooooo glad someone still knows that you either have man parts or woman parts. Thanks for the video as well. I am working a research class for Clinical Mental health and the book was less than helpful in explaining this. Very much appreciated.
years staying in academia, had to keep visiting these two concepts since always get confused, now finally get it n think won't get confused anymore. thank you James!!!
Finally I could find a perfect explanation! Thank you for sharing this video. It is extremely important for me to understand the difference as I will be dealing with moderation analysis soon. Thank you, James! :)
Very clear explanation and lessens my headache in understanding their difference 😂 Kidding aside, thank you, very helpful as I am doing my dissertation paper now.
I had a problem actually keeping the words themselves straight. I think because for me they kind of bring up a third thing in relation to two things, especially in reference to people. (A moderator in a meeting, A mediator in a debate) So what I'm going to associate it with is "moderation" as in the saying "everything in moderation." What is this everything? It's a situation that tips the scale in a certain direction. The context. The condition. When and when not? When one drinks too much. When one exercises too much. or Whom. A person who is too restrictive in their diet.
Your videos helped me so much when I was a PhD student. I am now a lecturer and I still come here to refresh on stats. Thanks so much for these videos it’s so appreciated.
Dear Professor Gaskin, I have been learning about your videos since a few years ago, and they are all beneficial! Lately, I came across a mediation model (simple mediation model, model 4). My variables are as follows: IV (emotional intelligence), MV( resilience), and DV (perceived stress). In my study, emotional intelligence is believed to positively predict resilience (MV), and resilience (MV) is believed to negatively predict perceived stress (DV). I have an IV-DV hypothesis in which emotional Intelligence is believed to negatively predict perceived stress (DV). However, I got confused after running PROCESS macro for my test. Here are my results: I found a significant indirect effect (presence of a mediation), a negative sign, a nonsignificant direct effect (positive sign), and a significant Total effect (negative sign). I don't know why the direct effect has the opposite sign to the indirect and total effect, even if it is nonsignificant. Is this common in mediation and what will be the implications if I have an opposite direct effect sign (nonsignificant)? I would greatly appreciate your reply.
For the formative variable p-value, you mentioned taking the liberal approach with a p-value of .10 when assessing bootstrap weights. Question: Do you use the same liberal p-value of .10 when explaining moderation or mediation and using the indirect effects or total effects assessment tables in SmartPLS V3.3?
Can a variable be both a mediator and a moderator. For example if Gender has an effect on Income, it can both be mediated and moderated by employment-status?
Employment status can be a mediator and a moderator because gender can have an effect on employment status (assuming there is gender discrimination), and employment status can change the effect of some variable on some other variable (not sure about gender to income relationship moderated by emploment status, since income should be zero for male or female if unemployed...)
Could there be a grey area? Caloric-intake could be a moderator: Exercise leads to weight loss, but how much depends on (conditional) caloric intake . Just as the amount of distractions moderates curiosity > learning .
I agree that caloric intake could actually be either since exercise may also lead to eating more (because you are hungrier). It is not always clear cut whether it should be a mediator or a moderator, but the rules in the video provide guidance when a variable might clearly violate one of the rules. I guess the strictest rule is that a mediator must be a causal outcome and causal antecedent of the IV and DV. If it is not, it cannot be a mediator. However, a moderator might have a mediating and an amplifying effect...
@@Gaskination I'm dealing with this exact situation in my research at the moment. I'd like to investigate the effects of a variable as both a mediator and moderator (because it really is both) but I don't believe it can be done in a single SEM model? Is that right?
Hi James-Thanks for this video. Question-could caloric intake be a MODERATOR if it's not exercise that's causing the change in intake? Like, what if how much you eat is independent of exercise and for people who choose high calorie foods, exercise doesn't really impact weight loss, but for people who choose low calorie foods, it does. (I'm not saying this is actually true, just an example to see if I understand the concept...)
Thanks for the video. I have a question: Can a variable be a mediator between A and C; and at the same time be a moderator of B and C? Does such relationships exist?
Hi James, When we have constructs with varying items say Construct Attitude has 4 items and Construct Behaviour has 6 items, Should we average out the total score of each construct before we carry out any statistical test like t-test, anova and so on? Please clarify
For t-test and anova, these will look at the average anyway, so you will have to combine the items in some way. An average is fine. As for whether there is an issue with having four and six items, no, it is fine.
thanks for making it simple. So, from your explanation, its evident that a variable can't be both mediator and moderator in a given causal relationship between other two variables? More over, are they distinguished at only one level/reason of causality, not any other dimensions on which they can be distinguished?
@@ramaduguhiranmayakaparthi6003 There are some rare cases where a mediator might be considered as a moderator (in a separate model), but not at the same time. The variable either: 1. is the reason for the effect X has on Y (i.e., it's a mediator) 2. or has an augmenting (strengthening, weakening, reversing, nullifying) effect on the relationship between X and Y (i.e., it's a moderator)
Hi James thanks for the great video! However, I got a question regarding the effect of exercising on caloric intake. In the video, you mentioned it is a mediator, since the more you exercise the higher your caloric intake will be. But I don't think that is correct per se, since one theoretically can take a fixed 1500 kcal per day, regardless of exercising or not (for those who try to lose weight). Hence, I would think that caloric is a moderator variable. I was wondering what your opinion is about this?
ctrl+f'd to look for a comment like this, I thought the same thing. Also, if you look at it in the way he explained, it would be: exercise > higher caloric intake > weight loss, which doesn't make sense. It is a great video though, just some questionable examples in the end in my opinion.
yep, it could definitely be a moderator if you fix your caloric intake, if your appetite is not strongly affected by your activity. For me, I get hungrier when I work out more... perpetual equilibrium...
Here's a fun pet project I've been working on: udreamed.com/. It is a dream analytics app. Here is the TH-cam channel where we post a new video almost three times per week: th-cam.com/channels/iujxblFduQz8V4xHjMzyzQ.html
Also available on iOS: apps.apple.com/us/app/udreamed/id1054428074
And Android: play.google.com/store/apps/details?id=com.unconsciouscognitioninc.unconsciouscognition&hl=en
Check it out! Thanks!
I've watched about 20 videos trying to get my head around this difference and never quite got it. Finally, I understand. Thank you, James.
Literally the single best explanation of mediating vs. moderating variables. I've struggled through a whole year of my PhD trying to get this. And within 5 minutes, now I get it! THANK YOU THANK YOU
WOW!
James, you will be the reason I will be able to finish writing my Master's thesis proposal today. No video or teacher or lecture slides have been able to help me learn this ever. I wish you many many subscribers, likes and whatever you need to be happy in life. Thank you.
For others, if there are the following variables:
1. motivation to learn about moderators and mediators
2. ease of learning about moderators and mediators
3. efforts in searching for James' YT channel
4. presence of other crappy videos on the same topic
Then,
"motivation to learn about moderators and mediators" is the IV,
"ease of learning about moderators and mediators" is the DV,
"efforts in searching for James' YT channel" is the mediator, and
"presence of other crappy videos on the same topic" is the moderator.
I share your sentiments
I CANNOT BELIEVE I FINALLY GET THIS. I'm sick of wasting my money in classes that leave me more confused. Thank you!!!!!!
The briefest, easiest and efficient way among all explanations. Thank you.
Amazing and very insightful and concise explanation for this confusingly issue, thank you James.
Mr Gaskin, thank you for this explanation. As narrow as the focus of the video may seem, it has influenced my entire life in altering how I understand cause and effect, and will benefit every aspect of every function in which I serve. I cannot thank you enough for this.
I have spent numerous hours, read several books, and watched many videos trying to figure this out. This was by far the most beneficial explanation. Thank you.
I read about the difference 1000 times and always would forget when i needed meaning I never understood, thank you for such a clear example
After years of trying to find a simple way to understand these two concepts I finally get it! Thanks a bunch. If I got it correctly this time, the satisfaction with my learning after watching your video was the mediator that made me sign up to your channel ;)
FINALLY get this. The value of just explaining something in simplistic, understandable terms can’t be overstated. Reading the comments I know I’m not the only one! Thank you! 🙌🏻
This video is by far the simplest and most understanding video. Thank you sooo much James for using these simple and not complex examples.
James, I appreciate this explanation more than you know. While I've had 20 different explanations of this concept and do understand what they mean, this video helps me understand the criteria. Great explanation!
Thank you so much for this. After hours of googling, reading, and watching numerous explanations about the topics, I can now have a break because of your video. Very much appreciated :)
Thank you!! This is by far the best explanation I came across. I understood moderator but not mediator and you made that concept click finally!
This is literally easiest explaination about the difference between them. Thank you so much!!
This is genuinely one of the most helpful videos I have seen in quite a long time. Thank you so much.
You get so many emails on this because so many people are bad at explaining it. This is the best explanation I’ve seen. Bravo.
fantastic, straightforward and clear explanations, James. You're amzing!
I finally understood. This video is way simpler than my lecture slides. Thank you!
This helped me so much, THANKS to you I can finally wrap my hand around this and get cracking . I was all over the place constantly switching variables, so lost. This video clarified everything!
Sooooo glad someone still knows that you either have man parts or woman parts. Thanks for the video as well. I am working a research class for Clinical Mental health and the book was less than helpful in explaining this. Very much appreciated.
This is excellent explanation of Moderator and Mediating variable MR. James Gaskin.
Thank you, I've been sitting here for hours trying to figure this out and this is the only clear explanation I've come across :)
This may also be informative: th-cam.com/video/rxxde64ghL8/w-d-xo.html
@@Gaskination It was, thank you :)
Thank you - seriously, this has helped me more than anything I've been taught or doing currently in my unit.
Studying for my Program Evaluation Final for my MPH degree, thank you for your service !!!!!!
years staying in academia, had to keep visiting these two concepts since always get confused, now finally get it n think won't get confused anymore. thank you James!!!
Thank you for the very simple explanation! The other definitions I found were confusing but yours is simpler and more direct to the point.
Most clear explanation on the topic I've come across, very helpful thankyou!
Now I can finally get a clear idea of the difference between the two. And perfect examples!!! Thank you so much!!!
I wish you were my lecturer. My teacher is simply wasting hours and hours of my time using complicated research terms when you did it in 30 seconds.
Finally I could find a perfect explanation! Thank you for sharing this video. It is extremely important for me to understand the difference as I will be dealing with moderation analysis soon. Thank you, James! :)
I had to read an article about the differnce and it made no sense because it didn't include context. Thanks for the examples, really all I needed!
Very clear explanation and lessens my headache in understanding their difference 😂 Kidding aside, thank you, very helpful as I am doing my dissertation paper now.
That is an excellent video and explains everything I want on this topic! The best explanation, thanks James!
I was really struggling with this and after watching your video its very clear now
Awesome explanation! Simple, clear, straight to the point. Thank you!
Extremely clear and to the point. The illustration is very easy to understand as well. Thank you so much!
Gonna have a quiz on this soon. This video does clarify the difference. Thanks a lot!!!!
Many videos I have watched but I didn't understand...but for your video I already understood
Thank you so so much. Got through bachelor's without understanding but I finally get it!!
Thank you for this channel, James Gaskin. You are my guru.
This one is an oasis in the midst of a vast confusing desert of information. Thanks a ton!
The best video I watched explained the differences. Thanks @James
This video really help to understand the difference between mediator and moderator.
3:20 punchline to understand it :) very clear and thank you 👏
Unbelievably crystal clear
Simply presented, and well understood. Thanks a lot
thanks james. i love this explanation. simple and robust. your rock.
Simple explanation of a complex issue. Thank you, you helped a lot!
Thank you so much for explaining the difference. You made it crystal clear.
Good explanation and very brief. Thanks James 👍
Literally the best explanation. Thank you!
Very good explanation and the examples are right on point. Thank you so much!
Very well explained, James.
James❤ Well explained I got my answers 👏🙌👌
Simple and clear explanation! You may go to heaven!
very simple, straightforward and easy to understand. Thank you
Maaan. Thank u. Finally a good answer for this issue
Super clear now...thanks so much Sensei James
I had a problem actually keeping the words themselves straight.
I think because for me they kind of bring up a third thing in relation to two things, especially in reference to people.
(A moderator in a meeting, A mediator in a debate)
So what I'm going to associate it with is "moderation" as in the saying "everything in moderation."
What is this everything? It's a situation that tips the scale in a certain direction. The context. The condition. When and when not? When one drinks too much. When one exercises too much. or Whom. A person who is too restrictive in their diet.
Thank you for the easy-to-understand explanation. :)
Hi, James! Thanks a lot for more this useful video for structural equation modeling classes. Take care!
Omg thanks you Dave, this clarifies the difference between moderator and mediator!
Thank you the well explained video! Finally got the concept !!
The best description that I have found. Thanks
Your videos helped me so much when I was a PhD student. I am now a lecturer and I still come here to refresh on stats. Thanks so much for these videos it’s so appreciated.
Thank you!!! Super clear explanation I LOVE IT 😍
Thx for your clear explanation
This video solved my puzzle
3:17 So you are saying that I can not change my gender by exercising, damn it!!!
hahahhahah
Very clear and straight to the point! :) Thanks for sharing!
Dear Professor Gaskin, I have been learning about your videos since a few years ago, and they are all beneficial! Lately, I came across a mediation model (simple mediation model, model 4). My variables are as follows: IV (emotional intelligence), MV( resilience), and DV (perceived stress). In my study, emotional intelligence is believed to positively predict resilience (MV), and resilience (MV) is believed to negatively predict perceived stress (DV). I have an IV-DV hypothesis in which emotional Intelligence is believed to negatively predict perceived stress (DV). However, I got confused after running PROCESS macro for my test. Here are my results: I found a significant indirect effect (presence of a mediation), a negative sign, a nonsignificant direct effect (positive sign), and a significant Total effect (negative sign). I don't know why the direct effect has the opposite sign to the indirect and total effect, even if it is nonsignificant. Is this common in mediation and what will be the implications if I have an opposite direct effect sign (nonsignificant)? I would greatly appreciate your reply.
To summarise my results: significant Indirect effect, negative sign (B= -.17), Significant total effect, negative sign (B= -.12), and nonsignificant direct effect, positive sign (B=.05)., other path coefficients; significant path a - emotional intelligence (IV) -> resilience (MV) (B=.47), significant path b - resilience (MV) -> perceived stress (DV), (B= -.32)
Since the direct effect is not significant, its sign does not matter. You only have a significant negative indirect effect.
This was so straight forward and really cleared things up for me! Thanks!
It is really really illuminative. Thank you so much
Very nice way of teaching
For the formative variable p-value, you mentioned taking the liberal approach with a p-value of .10 when assessing bootstrap weights. Question: Do you use the same liberal p-value of .10 when explaining moderation or mediation and using the indirect effects or total effects assessment tables in SmartPLS V3.3?
I do. But this is personal preference. I think that anything other than a confirmatory direct effect should be assessed with some tolerance.
excelent explanation of difference . thanks
Thanks this has been so helpful. It was simply explained clearly.
This was extremely helpful. Thank you!
Can a variable be both a mediator and a moderator. For example if Gender has an effect on Income, it can both be mediated and moderated by employment-status?
Employment status can be a mediator and a moderator because gender can have an effect on employment status (assuming there is gender discrimination), and employment status can change the effect of some variable on some other variable (not sure about gender to income relationship moderated by emploment status, since income should be zero for male or female if unemployed...)
@@Gaskination Ah yes I see, thanks. Maybe Sector (private/public) would have been a better example...
Wonderfully Explained!!
Could there be a grey area? Caloric-intake could be a moderator: Exercise leads to weight loss, but how much depends on (conditional) caloric intake . Just as the amount of distractions moderates curiosity > learning .
I agree that caloric intake could actually be either since exercise may also lead to eating more (because you are hungrier). It is not always clear cut whether it should be a mediator or a moderator, but the rules in the video provide guidance when a variable might clearly violate one of the rules. I guess the strictest rule is that a mediator must be a causal outcome and causal antecedent of the IV and DV. If it is not, it cannot be a mediator. However, a moderator might have a mediating and an amplifying effect...
@@Gaskination In this case it might be the context which is relevant to decide on whether the variable is a moderator or mediator.
@@printopdracht6462 correct, as long as it doesn't violate the mediator rule.
@@Gaskination I'm dealing with this exact situation in my research at the moment. I'd like to investigate the effects of a variable as both a mediator and moderator (because it really is both) but I don't believe it can be done in a single SEM model? Is that right?
@@esztellavezer767 Correct. You must test these two hypotheses separately.
In this day and age gender could probably be a mediator variable (not saying I agree nor disagree with that!). Clear video, thanks James.
This guy saved my life. Thanks.
Thanks James! This was very helpful.
Waaauw! Thank you so much for clearing this out! 👍👍👍🥳
Hi James-Thanks for this video. Question-could caloric intake be a MODERATOR if it's not exercise that's causing the change in intake? Like, what if how much you eat is independent of exercise and for people who choose high calorie foods, exercise doesn't really impact weight loss, but for people who choose low calorie foods, it does. (I'm not saying this is actually true, just an example to see if I understand the concept...)
Yes, that is correct.
Fantastic video! Thanks for the explanation
good explanation, thank you James
Thank you for your illustration, simple and effective.
thanks for simply the concept and made it fun
Thanks for the video. I have a question: Can a variable be a mediator between A and C; and at the same time be a moderator of B and C? Does such relationships exist?
I would recommend testing these hypotheses separately.
So informative, as usual: thank you very much James :)
Hi James, When we have constructs with varying items say Construct Attitude has 4 items and Construct Behaviour has 6 items, Should we average out the total score of each construct before we carry out any statistical test like t-test, anova and so on? Please clarify
For t-test and anova, these will look at the average anyway, so you will have to combine the items in some way. An average is fine. As for whether there is an issue with having four and six items, no, it is fine.
@@Gaskination Thanks a lot
thanks for making it simple. So, from your explanation, its evident that a variable can't be both mediator and moderator in a given causal relationship between other two variables?
More over, are they distinguished at only one level/reason of causality, not any other dimensions on which they can be distinguished?
HIRANMAYAKAPARTHI R writing MCAT?
@@wahajuddinmohammad6995 I don't even know about it, so NO.
@@ramaduguhiranmayakaparthi6003 There are some rare cases where a mediator might be considered as a moderator (in a separate model), but not at the same time. The variable either:
1. is the reason for the effect X has on Y (i.e., it's a mediator)
2. or has an augmenting (strengthening, weakening, reversing, nullifying) effect on the relationship between X and Y (i.e., it's a moderator)
Thank you James.
This is insightful.
Amazing explanation!
Hi James thanks for the great video! However, I got a question regarding the effect of exercising on caloric intake. In the video, you mentioned it is a mediator, since the more you exercise the higher your caloric intake will be. But I don't think that is correct per se, since one theoretically can take a fixed 1500 kcal per day, regardless of exercising or not (for those who try to lose weight). Hence, I would think that caloric is a moderator variable. I was wondering what your opinion is about this?
ctrl+f'd to look for a comment like this, I thought the same thing. Also, if you look at it in the way he explained, it would be: exercise > higher caloric intake > weight loss, which doesn't make sense. It is a great video though, just some questionable examples in the end in my opinion.
Scrolled down a bit and he answered this like two years ago, he says it could also be a moderator
yep, it could definitely be a moderator if you fix your caloric intake, if your appetite is not strongly affected by your activity. For me, I get hungrier when I work out more... perpetual equilibrium...
@@Gaskination Thank you James!
Clarified it! Thank you so much!