Endogeneity: An inconvenient truth (a gentle introduction), by John Antonakis

A key assumption of regression analysis (or structural equation modeling) is that the modeled independent variables are not endogenous. Yet, the problems of endogeneity are not well known to researchers working in many social sciences disciplines (e.g., management, applied psychology, sociology, etc.). When the independent variable has not been exogenously manipulated, there is a strong possibility that its relationship to a dependent variable will not be correctly estimated, leading to spurious findings. This podcast gives a brief and vivid overview to endogeneity and why it is engendered. Prof. John Antonakis discusses the problems of endogeneity using non-technical language and intuitive explanations; he shows that the observed relationship that is estimated can be very misleading when the independent variable is endogenous.

[@SebastianBrandes2300]

This is by far the best video I have ever seen on TH-cam. I'm in an Econometrics course in grad school and this video has helped me an incredible amount! Thank you so much!

[@DrPinkShirt]

Love this. This is a great reminder of the topic, and I love the examples. Thanks for making the topic fun.

[@natalialackeski]

I'm an Urban Studies masters student, coming from a social sciences background and was struggling to understand what endogeneity was. This video was everything I needed. Congratulations and thank you so much!

[@DrNealeOConnor]

John - I really appreciate your video and clarity of explanation. Reading your article now - it is very helpful.

[@MegaSaunier]

Mille grazie, mi hai insegnato inglese in poche ore! I may now have great conversation with the ladies, thanks to you!

[@prateekbedi3330]

Excellent explanation of the concept of endogeneity with a practical orientation. Really appreciate the work. Please keep making such useful content on more research related topics. Thank you for this one.

[@Jdigeon]

Excellent presentation, that's how a presentation or class should be taught, with concrete examples (and certainly not with dry notes which can make it boring!). Thanks again, I learned something new!

[@sldsanjose]

Very clear explanation, high production quality, good example and well worth the time invested to watch this.

[@mohammedmahinuralam2796]

Thank you very much, Professor John Antonakis. This is a great video on endogeneity that is useful for beginners as well as experienced users of econometric analysis that aims to find causal or correlational connection. Looking forward to watching more videos on similar issues. In fact, a hands-on exposition of regression analyisis dealing with endogeneity would be a desirable sequel to the video. Warm regards, Professor Antonakis!

[@DanielJBertelsen]

Very very good! More of this, please!

[@lovelyxtyxty8261]

Thanks for the video, it really helped to further understand the paper on "casual claims". My supervisor asked me to read the paper and to summarise it. It was so difficult at the beginning, but thankful this video help to clear some of the confusion. Cheers

[@dr.swapnilsoni]

well explained the curse of endogeneity. Even in Time Series when we have trend data, we de-trend it before modelling, this is also the removal of endogeneity arising due to time.

[@simongriffith3615]

Such a fantastic explanation. Thank you so much!

[@Zonnenbrillenmaker]

Excellent video. One comment on the problem of insufficient modelling in empirical studies which might be caused by a “significant result bias” in publishing rather than insufficient understanding of modelling. To get an article published in economics (my field), it is easier when there is a significant statistical effect in the study. Articles with no significant statistical effects are hard to publish, and this might lead some researchers to bias their results in favor of a significant relationship. This might be the case even when researchers themselves know the estimation technique to be flawed. As researchers are dependent on the number of publications, the “significant result bias” might cause worse research

[@timmykoko1659]

Amazing! coherent, clarity, very useful and helpful explanation. Thank you so much.

[@maxattack777]

Thank you for sharing this! Very helpful - econ student

[@seunosunkoya414]

Thanks for the video.

[@macken004]

this is excellent, thank you

[@blogginbuggin]

Fun & clear examples.

[@eucd1714]

Excellent video!! Thank you for adding that publishing studies with endogeneity is an ethical problem. If more scientist thought like you, researchers could actually make scientific progress instead of publishing papers with spurious relationships.

[@jorgemi3]

Thanks, very clear.

[@augustaaverence7218]

Wow! Thank you very much for such easy explanation!

[@seraye9862]

thank you!

[@CRISTIANO18786]

A great video, however, the loud sound and the disk shattering are in fact statistically correlated. I think what you are trying to say is that one doesn't couse the other. As the famous saying goes correlation doesn't imply causation.

[@Rob-fx2dw]

Correlation is not cause and effect. Interesting article but not new.

[@samcallaghan2731]

not really meant to be 'new' is it? more explaining for those of us lower down the pecking order. The question is not, 'is this new?' but rather, 'is it well explained?'

[@kqbftkr]

It's new to people who didn't know it before, so that's a useful video.

[@youssefkhurmani7600]

terrible explanation