YouTube transcript (7l6K0V_x_hw) — Transcript
Full transcript
- 0:00[Music]
- 0:13okay let's look at our water bottle
- 0:15example and I just walked through this
- 0:17um before but we'll walk through it
- 0:18again very quickly so remember a bottle
- 0:20water manufacturer on the product label
- 0:22states that each bottle contains 355 M
- 0:24of water so you work for a government
- 0:26agency that protects consumers by
- 0:28testing product volume volumes so you
- 0:31measure the volume that's actually
- 0:32stated on the label to make sure it's
- 0:33actually is true the case so a sample of
- 0:3650 bottles is tested is there anything
- 0:38we can assume to be true well yes we
- 0:42assume that the 355 M on the bottle is
- 0:46true we assume that to be the case
- 0:48that's our assumption that's our given
- 0:50now which hypothesis pair seems to be
- 0:55appropriate well I would say that the
- 0:57first one does so our null is an
- 1:00assumption and our assumption is that it
- 1:03equals an average of 355 ml and
- 1:08alternative would be the opposite of
- 1:13that so this is one we chose we set it
- 1:16up like this so our null is that the
- 1:19volume is
- 1:20355 milliliters therefore the
- 1:23alternative has to be that it does not
- 1:27equal 355 M it could be more more than
- 1:30that or could be less than
- 1:32that so if the data indicates the
- 1:35bottles are being filled properly then
- 1:38we fail to reject the null hypothesis we
- 1:42fail to reject our
- 1:45assumption so if it comes back that it's
- 1:48actually
- 1:49355 we have to sort of take our
- 1:52assumption as being the case so we
- 1:55cannot reject our null hypothesis we're
- 1:58not saying we have proven the null just
- 2:02that our assumption has held up so again
- 2:06think about it as an assumption that we
- 2:09either reject or fail to reject if the
- 2:13bottles are being filled properly then
- 2:15we would fail to reject that null
- 2:21hypothesis now if the data indicates the
- 2:23bottles are not being filled properly
- 2:26then we would reject the null we would
- 2:30reject our assumption so we kind of like
- 2:33put a big X through that n
- 2:36hypothesis so what we're saying is that
- 2:38our Assumption of 355 milliters is not
- 2:41held up under analysis we have
- 2:45statistical support for the validity of
- 2:48the alternative hypothesis so if we take
- 2:51our sample of 50 bottles and it comes
- 2:52back an average of
- 2:55340
- 2:58millit well probably we would say you
- 3:01know that's not that's not 355 it's not
- 3:03even close enough to 355 to account for
- 3:06natural variation so unfortunately we
- 3:10would have to reject our null that says
- 3:12it does equal 355 and then we would
- 3:15proceed down to our alternative that
- 3:17says it's not equal to
- 3:20355 and that's how the null and
- 3:23alternative
- 3:26work so again it is all in the wording
- 3:29and I alluded to this alluded to this
- 3:32earlier these are some words I avoid I
- 3:36avoid the word truth I avoid the word
- 3:40prove I avoid the word certain like a
- 3:44certain conclusion when it comes to
- 3:47Stats I do not like those words because
- 3:50as I'll say here in a minute stats is
- 3:52never certain stats never proves
- 3:55anything stats never says this is true
- 3:59it's all about
- 4:01probabilities so instead I prefer these
- 4:04words that the data supports something
- 4:07the data leads us to infer something the
- 4:11data seems to indicate something now
- 4:15some people might think I'm splitting
- 4:16hairs here but I really do think the
- 4:18words on the right are more appropriate
- 4:20for stats because statistics is never
- 4:24100% certain but what it does do a good
- 4:28job of is that it stay states its
- 4:31limitations
- 4:32explicitly so we're always talking about
- 4:35probabilities of this being something or
- 4:38that being something or what have you so
- 4:40I avoid the words on the left and
- 4:41instead try to use the words on the
- 4:44right I don't always succeed that's what
- 4:47I try to do to always indicate that
- 4:50there is no certainty there is no
- 4:52absolute truth there is no absolute
- 4:54proof when we're doing statistics
- 5:00okay so example two down on the farm so
- 5:03according to the United States
- 5:05Department of Agriculture the US da in
- 5:092006 the average Farm size in the state
- 5:12of Texas was 2.3 square
- 5:16kilometers now since the decades long
- 5:19trend has been for Farm sizes to
- 5:22increase due to large aggro businesses
- 5:25buying up land and setting up bigger
- 5:27Farms a business analyst wishes to test
- 5:31if the current 2013 Farm size is larger
- 5:35than it was in
- 5:382006 so let's establish a null and
- 5:41alternative
- 5:43hypothesis now the first question we're
- 5:45going to ask ourselves what is our
- 5:49assumption well in this case we assume
- 5:52that there has been no change in farm
- 5:55size since 2006 this is our null
- 5:59hypothesis now you may be thinking right
- 6:02now well it says in the problem that the
- 6:05decades long trend has been for Farm
- 6:07sizes to
- 6:08increase well so what the only
- 6:11information we have the only given
- 6:14information that we have is that the
- 6:17average Farm size in 20 2006 was 2.3
- 6:21Square km so that's our assumption
- 6:24that's our known that is our given so
- 6:28again that will be our null
- 6:32hypothesis so we assume that there's
- 6:35been no change in farm side since 2006
- 6:38that's our null are we testing any
- 6:41preliminary claim or any
- 6:43conjecture well yes based on our problem
- 6:47we wish to see if the farm size has
- 6:50increased since
- 6:532006 so which hypothesis format should
- 6:56we choose let's look at the first one
- 7:00will our null be an equality and our
- 7:02alternative a
- 7:04non-equality well no I don't think
- 7:06that's really appropriate for this one
- 7:07because we're talking about
- 7:09increasing uh how about the second one
- 7:12we wish to see if the farm size has
- 7:14increased since
- 7:162006 well I think this one is a good
- 7:19choice so I would pick this one well why
- 7:21is that so remember in this case our
- 7:24preliminary claim our conjecture which
- 7:26is the same thing as saying our
- 7:28alternative hypothesis is is that farm
- 7:31size has increased since
- 7:342006 therefore our null would be that it
- 7:38has remained the same or in in some case
- 7:40maybe even decreased so our null is that
- 7:44the farm size has remained the same or
- 7:46decreased and the alternative the
- 7:48conjecture the preliminary claim is that
- 7:51it has increased so that's why
- 7:53alternative is the greater than
- 7:57sign so how can we set this up so
- 8:00remember that we wish to see if the farm
- 8:02size has increased since 2006 so you
- 8:05look down here at our alternative
- 8:07hypothesis that is an orange as well so
- 8:10the increased is an orange and the
- 8:12alternative is an orange and again our
- 8:14null is that the form size has remained
- 8:17the same or possibly even
- 8:19decreased and the reason is that's the
- 8:21opposite of increased so we have to
- 8:24account for all possibilities if our
- 8:27alternative says increased then our null
- 8:30has to be stay the same or
- 8:32decreased so we can write it out like
- 8:35this so our null hypothesis is that farm
- 8:38size is less than or equal to 2.3 Square
- 8:42km because that's what we're given in
- 8:44the problem that's our assumption our
- 8:47alternative is that farm size has
- 8:50increased since
- 8:532006 so we could go out and collect data
- 8:56for many farms in the state of Texas
- 8:59then we could run those statistics we
- 9:01could do that mean now it's going to be
- 9:04one of these two things either the farm
- 9:06size is remain the same or even
- 9:08decreased or Farm size has increased and
- 9:12depending on what we get from our
- 9:14analysis we will know whether or not to
- 9:16reject or fail to reject our null
- 9:21hypothesis so if the data indicates that
- 9:23farm size has increased then we'll have
- 9:26to reject the null we're going to reject
- 9:30our top assumption there so our
- 9:32assumption has not held up under
- 9:35analysis we have statistical support for
- 9:37the validity of the alternative
- 9:44hypothesis okay then our final example
- 9:46Manchester United so during the 2010
- 9:502011 English Premier League season
- 9:54Manchester United home matches had an
- 9:56average attendance of 74,000
- 10:01961 now a club marketing analyst would
- 10:04like to see if attendance decreased
- 10:08during the most recent season so let's
- 10:11establish a null and alternative
- 10:13hypothesis for this analysis so what is
- 10:18our assumption given to us in this
- 10:20problem we can only assume that the
- 10:23attendance remained the same or possibly
- 10:27even increased cuz remember we're
- 10:29looking at the bottom here he's
- 10:31concerned or she is concerned if the
- 10:33attendance decreased but for now we can
- 10:35only assume the attendance remained the
- 10:39same are we testing some preliminary
- 10:42claim well yes the marketing analyst
- 10:45wishes to see if attendance has
- 10:47decreased since 2010 2011 so which
- 10:52hypothesis format should we
- 10:54choose well I would pick this one and
- 10:57why is that well keep in mind that our
- 11:00assumption has inequality so they're in
- 11:02the top and our alternative hypothesis
- 11:05the conjecture we want to test is if
- 11:08it's decreased so if you look down here
- 11:10in the alternative we have the less than
- 11:15sign so we can set up like this remember
- 11:18we're saying if it's decreased that's
- 11:20our research sort of hypothesis so that
- 11:22would be our alternative and our
- 11:24assumption is that it has remained the
- 11:26same or even possibly increased now in
- 11:28this case I'm not sure it could increase
- 11:30any further because I'm sure those are
- 11:31all sellouts but I guess it is possible
- 11:35but we're concerned about decrease so
- 11:37that's our alternative so we would set
- 11:39up a hypothesis like this so our null is
- 11:43that the attendance has increased or
- 11:45remain the same and alternative is the
- 11:48opposite of that that it's decreased and
- 11:51again that is the claim or the
- 11:54preliminary conjecture that this analyst
- 11:57wants to test
- 12:00so again if the data indicates that
- 12:02attendance has decreased then we would
- 12:04reject our null we would reject our
- 12:08assumption that attendance has stayed
- 12:10the same or even increased we would
- 12:12therefore have to go on to the
- 12:14alternative so we would say our
- 12:16assumption has not held up under
- 12:18analysis we have statistical support for
- 12:21the validity of the alternative
- 12:23hypothesis which is that attendance has
- 12:26in fact decreased since
- 12:302010
- 12:342011 okay just a couple reminders and
- 12:36then we are done so remember when trying
- 12:39to formulate a statistical hypothesis I
- 12:40want you to ask yourself the following
- 12:42question am I testing assumption or the
- 12:44status quo that I already know or it
- 12:47exists or am I testing a claim an
- 12:50assertion or some conjecture beyond what
- 12:53I already know or can
- 12:55know and they always follow this pattern
- 12:59the equality sign is always in the null
- 13:02hypothesis and then the opposite of that
- 13:04is always in the alternative hypothesis
- 13:08so always think about assumptions always
- 13:10think which will always be in the null
- 13:12think about unknown conjectures or
- 13:15unknown claims or assertions or research
- 13:18hypotheses those are going to be in the
- 13:20alternative hypothesis remember these
- 13:23are always in opposition to each other
- 13:25they can not both be true
- 13:30so a few reminders all statistical
- 13:32conclusions are made in reference to the
- 13:34null hypothesis as researchers are
- 13:36inless we either reject the null
- 13:38hypothesis or fail to reject the null
- 13:42hypothesis we do not accept the null so
- 13:45this due to the fact that non hypothesis
- 13:47is assumed to be true so we either
- 13:49reject or fail to reject our assumption
- 13:53if we reject the null hypothesis then we
- 13:56conclude the data supports the alter
- 13:59alternative hypothesis so if we do the
- 14:01data for Manchester United and we come
- 14:04up with a average attendance of
- 14:0968,000 then we would probably reject the
- 14:12null that says it's
- 14:1574,700 or more we would reject that move
- 14:19on to our alternative which says it's
- 14:21less than that if we had an average
- 14:24attendance of say
- 14:2668,000 however if we fail to reject the
- 14:29null hypothesis it does not mean we have
- 14:32proven the null is true again important
- 14:36distinction failure to reject the null
- 14:39does not mean we have proven the null is
- 14:41true we failed to reject our assumption
- 14:45and why is that because remember from
- 14:46the outset we assumed it was true from
- 14:49the beginning failure to reject the null
- 14:51does not equate to proof about its truth
- 14:55we're only talking about the null as an
- 14:57assumption to either be held up or
- 15:00knocked
- 15:04down okay so that wraps up our video on
- 15:07the null and alternative hypothesis so
- 15:11again in the future I plan on doing some
- 15:12more videos about the null and
- 15:13alternative and we're going to talk
- 15:15about type one and type two errors more
- 15:18specifically in our next video so the
- 15:20null and alternative will come up in
- 15:22future videos this was again just a
- 15:24basic introduction to how they are
- 15:26formulated how we interpret them how we
- 15:28talk about them and write about them and
- 15:30how we formulate them in mathematical
- 15:33equalities or greater than less than and
- 15:36things of that nature and again I wanted
- 15:38to use some real word examples so you
- 15:39can see how you pull information out of
- 15:42a problem to set them up now I'm not
- 15:45sure if your stats Professor or teacher
- 15:48will be as picky as I am about the
- 15:50language as far as when you talk about
- 15:53proving something or something is true
- 15:56again I prefer words like support or
- 16:00indicate um which a little bit softer
- 16:03words than deterministic words like
- 16:05truth and certainty and proof so again
- 16:08just keep in mind that we live our lives
- 16:11in one great big null hypothesis every
- 16:14second of every day we make no
- 16:17hypotheses about the world around us and
- 16:19about ourselves and we assume that those
- 16:23are indeed true so I gave several
- 16:26examples at the front I won't go into
- 16:27them but remember we always assume the
- 16:29world Works a certain way so we proceed
- 16:33through life with those assumptions
- 16:34those are our null hypothesis and of
- 16:37course when one of them is broken then
- 16:39we have to proceed to the alternative
- 16:42that something else might be the case so
- 16:44if I drop a rock in my backyard and it
- 16:46floats in the air and goes up to space
- 16:49then I will know that my null Assumption
- 16:50of gravity has a problem with it now I
- 16:54don't think that's going to happen but
- 16:55that's sort of an assumption we make
- 16:57right okay so that wraps up this video
- 17:00so just for a few reminders if you're
- 17:02watching this video because you're
- 17:03struggling in a class stay positive and
- 17:05keep your head up you're smart and
- 17:07talented I know that many of the people
- 17:09around you know that so you should think
- 17:10that as well if you like the video
- 17:13please give it a thumbs up share it with
- 17:15classmates or colleagues or put it on a
- 17:17playlist that does encourage me to keep
- 17:19making them please feel free to follow
- 17:21me here on YouTube on Twitter or on
- 17:24LinkedIn that way when I upload a video
- 17:27you know about it and I always like
- 17:29connecting with people who watch my
- 17:30videos online and finally just keep in
- 17:33mind that the fact that you're on here
- 17:35trying to learn trying to improve
- 17:37yourself as a student or as a business
- 17:39person that's what really matters I
- 17:41firmly believe that if you have the
- 17:43right learning process in place the
- 17:46results will take care of themselves so
- 17:49thank you very much for watching I wish
- 17:51you the best of luck and your studies
- 17:53and in your work and I look forward to
- 17:55seeing you again next time
- 18:00[Music]
About this transcript
This page contains the full transcript of YouTube transcript (7l6K0V_x_hw) , generated from the public captions YouTube serves with the video. The transcript has 2,637 words across 391 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.
What you can do with it
Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.
Free YouTube transcript tool
YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.