Statistical Significance and p-Values Explained Intuitively

If you’ve ever seen a news story about a scientific study, you’ve probably heard something like “statistically significant results.” More likely than not, what this is referring to is a scientific study that found something like p is less than .05. In this video I’m going to do two things: first, I’m going to explain what it means when someone says that something like “statistically significant result,” then I’ll give you an intuitive explanation of what those p-values are and, finally, if you stick around to the end of this video, I’m going to explain why we use a threshold of .05 and not, say, .04 or .08 or .15 or anything else when determining statistical significance.
Learn more about who I am and why I'm doing this here: • Data Demystified - Who...
Video that has more detail randomized experiments: • Video
Follow me at:
LinkedIn: / jeff-galak-768a193a
Patreon: / datademystified
Equipment Used for Filming:
Nikon D7100: amzn.to/320N1FZ
Softlight: amzn.to/2ZaXz3o
Yeti Microphone: amzn.to/2ZTXznB
iPad for Teleprompter: amzn.to/2ZSUkNh
Camtasia for Video Editing: amzn.to/2ZRPeAV

[@smitasingh8764]

This is the best explanation I got so far. I have watched so many videos but I didn't really get the intuition of what this p-value is. You made it all pretty easy to understand!

[@DataDemystified]

Glad it was helpful!

[@arunavamaulik19]

couldn't agree more!

[@maliamercer140]

So well explained! Thank you for putting out great, simple content that is very intuitive. After watching other videos yours sunk in the best. Nice production quality too. Cheers!

[@danielalexanderyt3232]

Took me 50 minutes to thoroughly go through this video and take notes for my AP psych test. Speaking of it, HOW IS THIS AP PSYCH

[@rockmamabaker8370]

Wow! I love this video! I've been trying to get my head around this for months. Really clear video. Great information. I feel slightly better about the topic- but statistics is hard and it takes a while to stay in my brain! Thank you!

[@Shanlii]

This was the best explanation regarding this topic! Dr. Galak explained it so well in a manner that makes sense and sticks; his intuitive explanation allows me to use this logic and apply it any question on the USMLE step 1 exam regarding this concept. Thank you so much!

[@bens1mo637]

Great explanation! Great, easy-to-understand, and clearly enthusiastic about psychology!

[@TheReef70]

Okay so I understood that the bigger the number the weaker the evidence and truer to the null hypothesis. The pace, clarity and preserving the material to basics is SUPER FREAKING supportive of a growth mindset as an adult learner.

[@DataDemystified]

Happy to help how I can!

[@Mabunt3]

I scoured the internet and my textbook and this is the best explanation I've come across. Thank you!

[@zohaibakhter3049]

The intuition that you provided in the beginning was exactly the piece I was missing to solve this cryptic p value puzzle! Thank you so much, please never stop teaching.

[@DataDemystified]

Thank you for the incredibly kind comment!

[@prabalbhadoria5398]

Amazing !!! Please ,please, keep uploading more such videos. You have no idea how much they help us !!

[@DataDemystified]

Thanks! Have a look at the other videos on the channel as there is already quite a bit of similar content.

[@TheKothay1]

Thank you for using the legal analogy to explain null hypothesis. Had been so confused so far. Also much thanks for such a simple explanation of the concepts of significance and P value. Looking fwd to some more videos.

[@DataDemystified]

I’m glad that this video helped clear things up for you!

[@emmanuelessien9527]

This is the best explanation to statistical significant studies. Even a grade student will understand this. Thanks Dr.

[@jeffreysayed8621]

Thanks for your explanation. You clarified the concept in plain English. I saw in another video how in Statistics significance has nothing to do with big and important. Something can be small or unimportant but still be "statistically significant."

[@martasobralgarcia3996]

This is the best video I ever saw explaining something on statistics! The explanations are great to understand the logic of it.

[@DataDemystified]

So glad it was helpful!

[@cleverhype]

Omg, this is so much easier to understand in the way you explain it. Thank you!

[@bobbiejoohagan2639]

AGREE! This was SO helpful. And so clearly and simply communicated. Thank you!

[@DataDemystified]

Thank you!

[@akash.deblanq]

This is the only video that made sense to me. Watched like 10 videos about p value and null hypothesis. This one was surprisingly the most intuitive. Thank you so much!!! :)

[@DataDemystified]

I'm so glad this helped you understand the concept!

[@bobby7354]

This is such a great video & explanation, thank you so much for bringing p-values & any other kind of data into context, it's so helpful!!

[@DataDemystified]

Glad it was helpful!

[@reuvenmeirherrera9580]

wow. very good at simplifying and explaining this VERY back and forth/ hard to grasp ideas ty

[@wotjake]

THANK YOU! This made so much more sense than other explanations. You Rock!

[@DataDemystified]

Glad you found it helpful!

[@shivapundir7105]

OMG! This was like the best, most intuitive candid explanation ever. After struggling with so many videos, finally this one made sense. Thank you very much Galak sir.

[@DataDemystified]

Really glad you found this useful!

[@aliasgari5152]

After watching 3 videos on p-value got to this one and I'm happy now. Keep doing.

[@DataDemystified]

So glad that this was helpful!

[@jaypalmer2463]

this was the best explanation I have ever come across - thank you!!!!

[@glowingmemories]

I've watched a few videos on P values/statistical significance and this is the one that made it click. Thanks!

[@DataDemystified]

Glad I could help you make the concept click!

[@filipzizak8676]

Thank you, well explained! Been having a tough time understanding this

[@DataDemystified]

Happy to help!

[@robertue1]

Thank you for sharing this video, you made it really easy to understand and digest. Keep up the good work.

[@DataDemystified]

Glad it was helpful!

[@HKC6924]

God sent.. very good explanation.. after a lot of head scratching, I now finally understood the whole idea behind the p values and rejecting the null hypothesis. Thank you 👍

[@DataDemystified]

Glad I could help!

[@AGWalp]

Ph.D. candidate (second doctorate): Outstanding, sir! The easiest (best) description that I have ever read or watched. I feel more confident in my comments about p and H0. Thank you for not getting into the "weeds!" That is where most of us get lost, and the jargon only makes it worse.

[@thekalyangk]

What an AMAZING explanation !! Super clear. Thank you !!

[@DataDemystified]

Very glad you found it useful!

[@triptajaswal8276]

Thankyou 😭 it is bestest video for understanding p value ... God bless you

[@DataDemystified]

Glad it was helpful!

[@al__karita]

Thank you for sharing your knowledge. Keep up the good work.

[@DataDemystified]

Thank you!

[@xxerggher2667]

After watching so many videos on p-values, this is the clearest explanation by far. Liked and subscribed!

[@DataDemystified]

I'm so glad you found it useful! If there are other topics you think could use some intuitive clarification, please let me know!

[@xxerggher2667]

@@DataDemystified Thanks for the reply. Maybe some people would appreciate a clear explanation of p-hacking.

[@festnotfirst]

This is a very easy to understand explanation, thank you.

[@SethWieder]

This was super helpful in my early understanding of theoretical statistics and what the null hypothesis means. Thank you!

[@DataDemystified]

Happy to help

[@robinvandepoele2117]

Hi! Just wanted to thank you for the video and to let you know that it’s so good, it and the video about confidence intervals are part of the obligatory curriculum in med school here in Belgium! Great work and keep it up

[@DataDemystified]

Thank you! Really glad it helps!

[@vupham7361]

OH MY GOD! This video really helps me understand the meaning of p-values not just some statistical jargon.

[@jbiswajit]

Thanks Professor. This was super easy to understand. Please keep posting videos.

[@DataDemystified]

Glad you found it useful!

[@Daizyonceagain]

This is the first spss lecture I enjoyed watching

[@DataDemystified]

Thanks! Glad you found it useful!

[@unnatigupta940]

OMG Thank you so much i was having hard time understanding the concept of P-value, but you made it very simple, even the people from non data scientist field can understand it.

[@DataDemystified]

Very happy to help!

[@mrazafarooq]

That is the best explanation so far for statically significant. Please introduce a course for statistics used in science to arrange the data and understand what statistical measures can be taken for the data?

[@douglassilver3606]

This is great information, thank you!

[@apoorvadk9635]

Thank you! Kudos for such a simplistic way of presentation!

[@DataDemystified]

Thank you!

[@whyisgamora4191]

OH MY GOD THANK YOU. I've watched soo many videos. My professor never explained that rejecting Null was something desirable in a study & always used abstract definitions. I finally understand it now. Subbed

[@DataDemystified]

Happy to help!

[@jakedrums1469]

He just explained it 300 times better than my teacher

[@minhhiennguyen6798]

really well explain video ! Hope to see more content from you

[@asma7918]

Finally I understood p value.. Best explanation ..bless you..

[@nutrimum2904]

THANK YOU!!!! What a confusion you've dissolved. Wonderfully explained! Thank you :D

[@DataDemystified]

Happy to help!

[@tanushreebhuimali2289]

Thank you making me understand the concept!

[@jamessujitmalo3910]

Great job. Thank you. Got a lot of valuable insights from the video.

[@vikashprasad557]

Thank you, it was very helpful. Just one feedback, if you control your pitch it will be very helpful

[@venkateshgopalarathnam1933]

Brilliant explanation of intuition behind p-value!

[@ankitakamath3120]

Thank you, your explanation was very effective. I finally understood the essence of p-value from your example. Subscribed! :)

[@user-ez2zd4ug9m]

Thank you very much for your explanation! I finally understood!! Great Job

[@aleksmusic94]

That was very helpful and clearly put. Thank you so much!

[@DataDemystified]

I’m really glad you found the video informative!

[@kateharrow8720]

This is so well explained!!! Thank you!

[@richardandoh7986]

This is well enunciated.

[@stephenaloia6695]

Very straightforward - thank you!

[@DataDemystified]

Glad you found the video useful!

[@huyarz61]

Your explanation is really helpful! I spent about an hour to watch others videos and try to understand what level of significance means, but most of them merely just mention it in a sentence without explanation. I was a bit frustrated until I watched your video !

[@xDAZZE]

You are so amazing. This made it super clear. Thank you!

[@DataDemystified]

Thank you! Do let me know if there are other topics I can help with!

[@SuhasMathavuVasanthsenaGowda]

Amazingly explained!!!👌

[@wannabeevangelist8481]

I have taken moment to like the video and subscribe to your channel, thank you berry berry much, it feels good to finally understand

[@DataDemystified]

Thank you and happy to help!

[@marcbeattie4484]

Such a great definition. Thank you!

[@DataDemystified]

Glad you found this useful!

[@rootenshi]

After hours of looking around and searching this topic including in my university book, KZbin videos, and scientific websites.. I finally understood the Statistical Significance from your explanation! So intuitive and clear. Thank a lot :) Also... Definitely subscribing to this great channel :D

[@DataDemystified]

I'm so glad you found this useful! If there are other topics in statistics that you find challenging to internalize, let me know. I'm always looking to make content that will actually help people!

[@rootenshi]

@@DataDemystified I really appreciate it. You will definitely be the first place I go to for demystifying data :)

[@JaneChirwa-db8ij]

terrific explanation. Thank you.

[@timhowell6929]

Very nice explanation, thank you!

[@aragzan9]

Simply amazing explanation

[@DataDemystified]

Thank you!

[@yvetteelizalde7396]

Great work! Going to share with all my CFA study friends!

[@DataDemystified]

Glad you found it useful! Just out of curiosity, is CFA: certified financial analyst? I ask because at CMU we have the College of Fine Arts (CFA) which is very different!

[@miriambendayan6530]

Thank you !!! Very well explained

[@edangenetfeleke1053]

Really amazing with good and simple example keep it up!!!!!!!!!!

[@DataDemystified]

Glad you found the video helpful!

[@arunavamaulik19]

I find your videos extremely intuitive and would love to see you make a whole series on probability and statistics. There are quite a few on the internet but not many are intuitive.

[@DataDemystified]

Thank you! Are there any specific topics in probability and statistics you find confusing? It's always nice to know where to focus my efforts.

[@arunavamaulik19]

@@DataDemystified I would love to see a series on Bayesian inference! To be honest it could be any topic but a complete mini series would be helpful. Eg: I loved your video 'what's a distribution?', you could follow that up with a few more videos on similar topics. In any case, I hope you continue making these videos! pure gold

[@jeffgalak9402]

@@arunavamaulik19 Thank you for the suggestions!

[@hsrhmij8850]

Finally, You give me a strong evidence to understand Statistically Significant! Thanks and Keep it up!!❤❤❤

[@DataDemystified]

Glad you found it useful!

[@hsrhmij8850]

@@DataDemystified hope we'll learn more and more useful thing like this!!!❤

[@emilyrosenzweig9946]

Love the ball and jury analogies!

[@DataDemystified]

Thanks!

[@Sluo1947]

Awesome explaination. I never understood why want to find evidence against H0, now it's clear.

[@DataDemystified]

Thanks! I'm so glad this helped you understand the concept better!

[@philad2079]

Great video. It makes a lot of sense.

[@ayanamcgrew7736]

such a good video, thank you for you help!!!

[@franciscogalan8017]

Thanks so much. Very useful video!

[@v9318]

Thank you, I was so stuck on this!

[@DataDemystified]

Glad I could help!

[@user-zs8xv1sh6t]

Thank you for this great example!

[@joannameringoff6870]

Loving this - a little slower talking would be wonderful but can always go back a few seconds and re-listen.

[@DataDemystified]

Thank you! Yes, I'm working on finding just the right speaking speed. Feedback taken!

[@zinnijha7556]

Thanks for simplifying for us. I wondered why .05 ...:)

[@DataDemystified]

You're welcome!

[@towobolafunmilayo3224]

Thank you, you real made it better to understand

[@yellowsniper1995]

Amazing video! Just subscribed! 👍🏽

[@jsklair]

Great video as always. Can you do similar primer videos on those alternative methods you mentioned ("Confidence intervals", "Bayesian estimates", and "Effect sizes")?

[@DataDemystified]

Absolutely. I already have one on Confidence Intervals (see here: kzbin.info/www/bejne/i4Wro6ltjqatnq8). I also have one that intuitively explains how to think about Bayes' theorem without really calling it that (see here: kzbin.info/www/bejne/b56yomyaYsqLrqs). I have one about Effect Sizes in the queue!

[@me951129]

Unfortunately giving more than one thumbs up isn‘t possible. This was brilliant!!

[@ShazWag]

Good explanation. Thank you 👍🏼

[@DataDemystified]

Thank you! Glad you found it useful!

[@shantimodel1]

Excellent explanation

[@tonynikolaos3527]

Clear explanation, Indeed! Thank you! An area where I feel a “black box magic” happening, that would hope you can explain clearly is: in random permutation tests, data is shuffled randomly from both groups, some (less then factorial for practical purposes, yet sufficient ) number of times, and the subtracted means result makes a normal disturb plot. Now, we know: garbage in-garbage out. The randomization and permutations is possible “garbage”, over exposed and shafted with the original data from the 2 groups. Now, what I feel is magic, I’m supposed to believe, and come to conclusion for validation or rejection of the Ho, based on that original data, from Groups 1 and 2, when mixed with “garbage” to actually retain some of its qualities(or dilute them if it doesn’t) , through the random shuffled permutations, and if it has retained sufficiently( or diluted sufficiently) show me a P value that will disclose strong or weak tendency toward rejecting or accepting the Ho. So , basically, we use the random permutations, as if, analogizing, as if we could use some substance, and say it has those features, and then we mix it with “all kinds of “garbage tests” and eventually are able to say: see, it kept its features, through all those substantive permutations, must be strong, or, you see, it got diluted so much that lost those features, must be not enough strong, what you saw as feature was random feature case, so we keep the Ho. Something like that? Thanks a million- if you can touch upon this theme!

[@yubuwang8823]

This video is amazing!!!!!! I'm a Chinese student, I do a lot of research about this topic, no one explains this as clear as you are! I will pay to see your lecture!

[@DataDemystified]

Thank you! No need to pay :). Just tell your friends if they’re also trying to learn!

[@DHDH_DH]

I am also from China and am currently taking an introductory statistics course in the U.S. No one on KZbin gives a better explanation than you on the significance level and P-value.

[@jeffgalak9402]

@@DHDH_DH thank you!

[@Do_Worrk]

This is an amazing explaination. Subbed! Can you go into Confidnece Intervals, Bayesian Estimates, and Effect Sizes?

[@platinum11110]

I discovered the channel today, went to see what other videos there were but saw that it's been a year since his last video 😪

[@user-ws5sl5jg8t]

Thanks, that's definitely a great video!

[@srinivasbringu67]

Beautifully Said!

[@Data4DecisionMaking]

Thanks for the great video!

[@gauravmohan9271]

Stunning….work brother

[@DataDemystified]

Thank you!

[@debla123]

Really liked the way you explained

[@emmanueldarkwa8311]

Excellent teachings. Please I need videos on sample size calculations and z score calculation

[@iamafollowerofChrist]

YOU SAVED ME! thank uuuuu!!

[@nadiashareef4500]

veeeeery helpful thank u sooo much bro

[@DataDemystified]

Glad to be of help!

[@nickiemmy9177]

Thank you so much for this!!

[@DataDemystified]

Happy to help!

[@hassanrevel]

Thanks bro. You are great.

[@DataDemystified]

Thanks!

[@sahareeshjabeen]

excellent explanation really helped me!

[@DataDemystified]

Glad to hear it!

[@dr.strangelove8846]

Well done ! And thank you