Student's t-test
Vložit
- čas přidán 6. 07. 2024
- Excel file: dl.dropboxusercontent.com/u/5...
In this video Paul Andersen explains how to run the student's t-test on a set of data. He starts by explaining conceptually how a t-value can be used to determine the statistical difference between two samples. He then shows you how to use a t-test to test the null hypothesis. He finally gives you a separate data set that can be used to practice running the test.
Do you speak another language? Help me translate my videos:
www.bozemanscience.com/transla...
Music Attribution
Intro
Title: I4dsong_loop_main.wav
Artist: CosmicD
Link to sound: www.freesound.org/people/Cosmi...
Creative Commons Atribution License
Outro
Title: String Theory
Artist: Herman Jolly
sunsetvalley.bandcamp.com/trac...
All of the images are licensed under creative commons and public domain licensing:
1.3.6.7.2. Critical Values of the Student’s-t Distribution. (n.d.). Retrieved April 12, 2016, from www.itl.nist.gov/div898/handbo...
File:Hordeum-barley.jpg - Wikimedia Commons. (n.d.). Retrieved April 11, 2016, from commons.wikimedia.org/wiki/Fi...
Keinänen, S. (2005). English: Guinness for strenght. Retrieved from commons.wikimedia.org/wiki/Fi...
Kirton, L. (2007). English: Footpath through barley field. A well defined and well used footpath through the fields at Nuthall. Retrieved from commons.wikimedia.org/wiki/Fi...
pl.wikipedia, U. W. on. ([object HTMLTableCellElement]). English: William Sealy Gosset, known as “Student”, British statistician. Picture taken in 1908. Retrieved from commons.wikimedia.org/wiki/Fi...
The T-Test. (n.d.). Retrieved April 12, 2016, from www.socialresearchmethods.net/...
Dude somehow you don't make me want to die while learning this stuff haha, I wish my professor was more like you, thanks for the help!
Awesome explaination.
fr, my prof is so boring even though he's done some seriously cool engineering work in the past
"...so, instead of this being known as the Gosset's t-test, it's known as the -"
*cue smooth electronic beat*
Bozeman really turning science into a tv drama, lol.
he pulled an invincible in 2016
Student T Test
...I still have no clue 😭😭 Probably going to re-watch this a dozen times, cause it explains it better than my lecturer this year
Thank you for posting this! You explain statistics far better than most teachers!
i have taken over 10+ stats courses, this was very very well explained
@@NostalgicHearts deez nutz
Completely agree with you. He can explain in one short video what most teachers try to explain in one week's lecture.
0.05 probability means that if we will conduct the experiment 100 times, 95 times we will reject the null hypothesis and 5 times accept....
Why?
Approuved.
I just started my PhD in Psychology and the program is extremely quantitative. It's been years since I've taken a stats test and watching your videos have seriously helped in me remember the basics. I'm so going to keep you in my prayers! Thank you for making it seem less scary.
How did your PhD go?
@@burtons9922 just finishing. I propose my dissertation on the 27th!
@@zainabmohammed1141 Congratulations!! And good luck next week!
@@zainabmohammed1141 How did the dissertation go?
@@yonisabdull I’m a PhD candidate and defend next month Insha’Allah
Crystal clear description of t- Test. I could finally understand this concept. Thank you !!
Was that helpful??
God! You have no idea how much I have struggled to wrap my head around this whole Statistics topics. And again, you have no idea what a boon your videos are turning out for me! Many many thanks, Bozeman! You the boss!
Thank you so much Coach Steve Kerr for this lecture video. Despite your busy schedule as the GSW Head Coach, you were still able to find time and record this important topic in Research.
ha ha! Love this
This was so helpful! I haven't taken stats in years and have had to partially watch so many videos before finding this one. I really appreciate the clarity and step by step approach!
This is the best way of introducing Student T-Test I have ever seen! Every easy to understand! Thanks!
I've been watching and reading lectures for hours, this is by far the best lecture I've seen
Honestly I understood more by listening to you for 10 min, than I have my whole life listening to mundane lectures or reading notes. And the best thing is that listening to u some how doesn't make me wanna die while studying t-test. Truly u r a gem. Thank you so much
This man needs ALL THE AWARDS.
Bruh got me through high school, college, and continues to help me refresh for job relating stuff.
This dude is single-handedly responsible for a quarter of millennial contribution to GDP. xD
Mr. Andersen, I have had the privilege to meet you in person and tell you, "Thank you!," but it bears repeating here.
You have no idea how much you have helped me over the years to understand fundamental concepts that are glossed over in lecture leaving my head spinning unable to follow along. I get home, come to CZcams and watch you unravel the whole thing in < 15 minutes.
You are an exceptionally talented teacher, and your work does not go unnoticed. Thank you for your gift of quality instruction to complete strangers on the internet over many years now. I wish that more teachers would watch your style and emulate that rather than trying to rip through slides or go off on pointless rabbit trails during lecture. "Thank you!"
Well he probably got noticed now lol
Bozeman, this is amazing. I learnt and solved a problem in minutes. I had a problem on biostat on my desk and needed solution and this video just helped clear it all. I'm super grateful
Your explanation is simply amazing, in the sense that you answer any question that pops up in my mind at any point of doing the solution! a heartfelt thank you!
I’ve learned more in this 10 mins then I have all semester 😩😩 Thank You so much🙏🏾
omg... lol I was having trouble with this subject & BAM I get a notification of you uploading this video. THANK YOU.
Double BAM
triple BAM(statQuest)
Best 10 min explanation of any statistics concepts so far..I liked the usage of visual aids in explaining and demonstrating the calculations. Thank you!
Giving you a big like man :) really great explanation .... I learned the T-Test at university but haven't used it for several years so wasn't sure how to count it and you brought it back to my mind ..... really great video man :)
Thank you so much! I have to do this for quantitative analysis, and you've made this process so much easier! Again, this video is much appreciated!
Tomorrow is my chemistry exam for which I m watching ur lessons thanks a lot for this understandable video
Thank you for such a clear and interesting explanation. I did t-tests in a statistics course I did, but forgot exactly what it was. This video cleared up the problem brilliantly!
You are excellent at explaining these subjects. I needed a quick brush-up of these concepts for a job interview and this video turned out to be very helpful!
you are indeed a great teacher. I love this video. I got a t critical which is greater than my t statistics and this establish the claim of the null hypothesis to be true, so I fail to reject the null hypothesis which means that there is no statistical significant difference between the samples. also, my probability value is greater than my alpha value, so this also tells me that my null hypothesis should not be rejected.
Thank you so much, very detailed video, and explained perfectly. I had a biology lab report and I haven't taken a statistics course yet, so I had no clue what to do in Biology X since most kids are older than me !
I have watched many videos about this topic, this one is different. Visualization and simple explanation make this difference. Thank you, Mr.Bozeman
This Prof. is No.1 !!!! Fast and simple lecture that cannot be better!
Finally a video that clearly explained t test. thank you!
I actually get it!!! You sir are amazing. I've been reading all this in my textbook and just not getting it. Thank you so much.
Thanks a lot, this is stuff I'm expected to know. But I've never been taught it, nor have I've ever used it. So it's nice to have it explained properly.
Wow - I love the way you crunch the numbers instead of using a methods software. Now I understand the PROCESS. It makes so much more sense now. Thanks!
My dude!! THANK YOU SO MUCH we really really appreciate your channel! God bless you xx
What does it meeeaaan? Thanks for the video man
What does that meeEEEeaaAAn?
We calculate the mean but what does that MEAN? (why so mean?)
a comment I anticipated after watching this video haha
literally
mate I was saying the same thing too myself ahah
WOW. This is the best explanation I have come across so far. Thank you so much. The analogy SNR is fabulous.
You explain in an incredibly easy way to understand! Please keep doing this!
Watched this during my second undergraduate year, come back as a PhD. student to watch it again :)
You are a genius explaining statistics! ;)
Bozeman, I congratulate you.
I am horrible at statistics and it is my achilles heel. I can do wonders with data in R, trully. But when it comes to demonstating significance etc...
This demonstration (and others) really clarified what was going on and why and where.
You know how to teach.
I have been reading biostats for 3 hours and this one video cleared all my queries regarding mean, sd, variance, t test, tests of significance altogether
Excellent example in 10 minutes. Thank you
Oh my god thank you so much! My biology teachers made zero sense when explaining this!
i was long searching for a video on statistics. none of them explained as simple n neat as you sir.. thanku so much. you are an awesome teacher..
The signal & noise analogy helped me LOADS to understand. Great!
This was incredibly helpful! Thank you so much for this amazing lesson!
Thanks Mr. Bozeman! You really helped me out! I have actually been studying this and I could never understand it.
great video! i spent 2 weeks trying to understand the results of my data, and i finally found here. big help. tnx
Man, you are an EXCELLENT teacher! Wonderful clarity.
I wish I somebody showed me this damn video instead going through a readers hundred times and not getting it. Thank you! 🤜
right?!
BOZEMAN BACK AT IT AGAIN WITH THE SCIENCE ;)
+Ks Parikh Been traveling for the last few weeks. Feels nice to be home uploading videos again.
+Bozeman Science
people like us always love the science, statistical analysis and other education videos
Thanks Bozeman
هه
اوحه
I just learned a semesters worth of work for my research in psych class from your account. I have used this account all throughout college. I owe you my tuition.
Finally! Someone who speaks english! Perfect video showing the real world applications and methodology to come to a conclusion, thank you so much for the entertaining lecture!
the link to the excel file does not work. it says 'File not found
Sorry, that file doesn’t live here anymore. It might have been moved or made private.'
you saved my ass, for my exam is tomorrow
This man deserves genuine respect from students. I wish he was my professor.
Thanks so much for taking the time to create such an informative and engaging video. Your efforts are greatly appreciated.
This video's explanation is excellent! Though I am Japanese, I could not understand T-test, P-value and etc by reading Japanese textbooks of statistics. But now I have understood of them by watching this video. I think, Japanese Professors of all university and college are really bad at explaining in the lectures and
of books written by themselves. Thank you a lot!
funny my Professor is Japanese
this is freakazoid clear. thank you. dropbox link is dead tho, so can't run the example.
I like that t-table link you gave because it gives df all the way up to 100 without skipping any values in between
Man, excellent video/explanation! I've read a lot of articles about this test, but only understood it clearly with your video. Thanks!
I got:
t=0.206
t-test(p-value)=0.83814447
Therefore, I fail to reject the null hypothesis, meaning there is no statistical significant difference between the two samples.
I had a question though...on the excel spreadsheet do I type in: =ttest(plant A, plant B, 2, 2) or (plant A, plant B, 2, *3* )? The last number is for the "type" and excel reads to put in "3" instead of "2" if the variances are not equal for the two-tailed test. The variances I got seem pretty far apart: *Plant A variance= 138.69 and Plant B variance= 226.29*. Your variances were closer to 1, so I see why you put in "2" for the type. When I put in "3" my p-value turns out to be slightly different= 0.83820433 than 0.83814447. Which is more accurate?
hmm.... Same question here....
I got he same results as you did. I had to type in the 2 in place of the 3
Actually, a variance ratio of 2 isn’t all that uncommon...the t-test is rather robust to unequal variance as I recall, and you probably don’t need to worry until your variance ratio is over 3, in my experience. If you’re in doubt, you can always use the unequal variance option regardless of your variances, since the statistical power of the two methods is pretty much the same. But since the unequal variance option reduces the degrees of freedom compared to the equal variance option, the corresponding critical value for the unequal variance option will be slightly higher., and thus the p-value will be affected as well.
I want to see if I am understanding this correctly. We have a t-value of 0.211, p value of 0.838, and df value of 40. Even though the critical value for p=0.05 and df=40 is 2.02, do we say that we don't reject the null hypothesis because it probably won't go below 0.211? Or are we comparing the t-value and p-value directly, which doesn't seem to make much sense to me?
no
we are looking at the value of the t stat and the t critical
if t stat is less than t critical, you don't reject null hypothesis and vice versa. likewise, you can use p-value and alpha value to make judgment as well. if your pvalue is greater than your alpha value, you must not reject your null hypothesis.
the p value and t value both go inverse to each other
thank you sooooo much. these 3 lines by bolaji enabled me to get rid of all doubts.
Thanks so much, I just undertook my first t-tests for years using your guide - very easy to understand method/syntax etc
These kind of videos make me feel like we should drop the unis and get the true knowledge through an CZcams :) thanks for the great video !
I don't know how but I understand this now.
No doubt, the teaching, the content, delivery and everything else was incredible. This is one of the best videos I ever found with complete knowledge. I have just one doubt at 8:35 you mentioned that 0.95 is the probability of rejecting the null hypothesis. But I was unable to grasp it as it always seemed that it is the probability of the acceptance region. Is it different from Z- test? And it will be great if you could make a video on Z test. :)
You are RIGHT! It is the probability of accepting the Null hypothesis. That is how we frame the hypothesis by giving more weightage to Null hypothesis.
i love people who make the time to make these videos so much
Thank you for making this video. I'm studying stats as part of a university epidemiology course and this makes the concept sound less complicated.
Meaning of 95% confidence level, the meaning that I wanted
I think its 95% of the time we fail to reject and 5% we would reject since it's based off the null at 5:48
I had the same doubt while watching the video. Did you find the answer to your query?
I believe the explanation in the video is correct. If the t-test gives you a value that is less than the p-value, then you reject the null hypothesis(the difference between the samples is statistically significant). In other words, you compare the p-value with the critical value (0.05 or 0.025), but compare the t-value(the result after putting the numbers in the formula) with the corresponding cell from the distribution table.
If the t-value statistic is 'greater' than the corresponding value in the table -> you reject the null hypothesis.
If the p-value is 'lesser' than the critical value (0.05 or 0.025) -> you reject the null hypothesis.
@@samsung6980 THANK YOU. I HATE STATS SO MUCH
This video was amazingly helpful for my studies this week. Thank you so much for taking the time to make it!
I'm working on a lab report requiring statistical analysis for my data, your 10 minute video did two hours of work for me!
Salman Syed I’m working on a lab report too lol
"use a spreadsheet" oh I wish 😂. Gotta do this in my exam 😂. Thank you.
You have done your test already but if you were allowed to use a calculator than it could have done it for you.
I would like to know about ANOVA
me too
thank you for making this video. I finally understand the t-value and the t-test. It's so nice that you even demonstrate how to compute the data in excel spreadsheet. Thank you so much.
Thank you so much for actually explaining the equation. First time I've seen this. It's so difficult for me to use a formula without understanding why it looks the way it does!
GOD BLESS YOU
I found this video too late. I have just passed my STAT class. This video would have helped.
This was a lot better than watching my lecturers 1 and a half-hour lecture on t-tests, thank you (:
This is an excellent, very well explained video simplifying and making accessible an incredibly complicated data test. Thank you very much for creating this.
What does that meeeean?
I injected mariweed and I dieded
Which one of your joints had a higher average high?
Fantastic thank you
3:00 t-value formula compares difference in means to difference in variance of the samples. Like a signal to noise ratio.
4:55 once we have a t-value, we can read from the table and see what p-value that t-value corresponds to. That is the t-test, seeing if our t-value is high enough to give us a low p-value (and if so, then the difference in means is statistically significant and we can reject the null hypothesis)
AMAZING VIDEO!!!! Thanks a lot. I have done a time series, and statistical analysis course, and no one explained the concepts as you did! Keep going to help others. Thanks again!!
It is a very great video, i learn t test in high school, pre-uni, year 1, master degree...........now i graduate for 4 years and finally understand this after watching this video, i love youtube!
This is a very clear and useful video. One of the best I've ever seen. Thanks for posting this.
Super clear, I would recommend this to any student in high school that is having a tough time understanding t-tests-
Much appreciated kind sir, your explanations are better than all I've seen so far.
THANK YOU! This was exactly what I needed, the concept behind the t-test to the spreadsheet calculation.
The best explanation of t test I have ever heard. Thank you very much for this amazing video
Other explanation videos skipped the most tricky parts but this video was very helpful!
So clearly explained! Awesome thanks, so many lecturers have been through this with me (and know it well), but are terrible at explaining it this clearly!!
by far the best t test video on youtube.
wow... I have seen many videos, but your explanation made it clear for me to do a two-tailed t-test. Thank you!
Absolutely excellent amazing & clear video. Gave amazing clarity on the concept and use and significance of t test. Im awaiting more such videos regarding statistics and research be uploaded soon. Thank you so much sir. 😊
This was great-- Nice work on this video. I've been a teacher for a long time and you really hit all the bases-- Thanks for this!
Thanks for the excel tips!!! Just saved me hours of hand jamming!
Thanks for the refresher. This really fortified my understanding of t tests and learned about ambiguity in direction of distribution
Excellent video, Precise, Crisp and quick !
Thank you for this excellent and lucid explanation. Best video on statistical applications found so far !