This content introduces two new topics relevant to the AP curriculum: systematic random sampling and the distinction between retrospective and prospective observational studies, building upon previously learned sampling techniques and the definition of observational studies versus experiments.
Mind Map
Click to expand
Click to explore the full interactive mind map • Zoom, pan, and navigate
so before we can officially finish
chapter four I have a few minor topics
that I want to talk about with you guys
here in this video these two topics are
things that were added to the AP
curriculum after our book was written so
they're not actually in our textbook and
they don't pop up in the homework
problems I give you from the book
neither of these topics is very hard
there's something we need to talk about
though because they do appear on the AP
test and it's also a good chance to
review some of the big things we've
talked about in this unit our first
topic is going to give us a chance to
review all of the nice sampling
techniques that we learned in chapter 4
and these first three options right here
should be pretty familiar to you guys
because we talked about them in quite a
bit of detail as a quick summary right
here SRS a simple random sample every
group of size n so this group of n
people or this group or this group is
equally likely to be chosen it's your
most basic option that if you think
about making a random sample just like
before you took this class you probably
would have been describing an SRS for a
stratified random sample you need to
think of a variable that matters or that
you think will matter and the results
that you get my classic example that I
use a lot is with freshmen sophomores
juniors and seniors so I'd have each
group separate here freshmen sophomores
juniors and seniors each bubble is
different from one another this bubble
is not the same as this bubble right
here etc so you would need to do a
problem where the year in school that
you are would make a difference in how
you would answer a question um so my
classic example is prom and if freshmen
should be able to go freshmen probably
of different opinions than seniors so
what you do then once you establish your
bubbles is you randomly sample some
people out of each of your separate
bubbles and that becomes your sample so
you need to think of a variable that
matters and separate into those groups
before you do your sampling a cluster
sample on the other on the other hand is
when all of your groups are roughly
similar to each other so this group
right here is about the same as this
group right here is about the same as
this group right here I use in apartment buildings
buildings
analogy and one of them or um well
basically whatever you want to think
about it does just realize that each
bubble is like a mini version of your
population and with the cluster sample
you number off the groups and Group one
is just about the same as group two just
about the same as group three one two of
them numbered you randomly sample the
group's okay I got number two at number
seven and everybody in those bubbles
becomes your sample it's still read them
but the randomness happens in which of
the bubbles you select those three we've
already learned about that was just a
quick refresher the new one that we are
talking about in this lesson is a
systematic random sample and let me
describe what a systematic random sample
is because we will talk about it in the
next page with the definition but just
to give you guys an intuitive idea you
see this in situations like with exit
polls for surveys so like for elections
and stuff like that they do exit polls
so they have a pretty good idea about
who's gonna win before the official
results are even posted if you watch the
news during elections they'll call a
state and be like this state is going
forward this candidate before the
results are made official it's because
the data from the exit polls is just so
strong um
the thing about doing an exit poll
though if you're gonna do a survey you
don't know how many people are gonna
show up that day so if I'm trying to do
a random sample but people going to a
poll I don't know what to take it out I
don't know how many people I need to
choose for my sample because I have no
idea how many people will show up so
what you do is like let's say you decide
I want to talk to every tenth person who
shows up so what I would do is I would
randomly pick a number from one to ten
and like say I get eight person number
eight to that day is my first person
that I go to and then after I hit that
person right there every tenth person
after that I would go to systematic kind
of conjures up an image of there's a
system to it so every tenth person every
fifteenth person or whatever you're
gonna talk to that person the randomness
isn't where you starts so what's your
first person and then from there you go
every and third caithe or whatever you
want to call it person so that's what a
systematic random sample is and that is
the new topic of this video right here
this definition in blue basically
explains what I just said
pick your starting point randomly and
then after that every blank person is
gonna be the people that are in your
sample this works really well too when
you want to do surveys in public like if
you want to just go and find out opinion
of people at the gallery or the mall or
something like that because you don't
have a list of everybody who's gonna be
there and you don't even know how many
people are gonna be there so in that
situation a systematic sample can be
useful so it's very useful in public or
where you don't know or where you don't
have access to a list of the population
and the big drawback to a systematic
random sample is if there's some sort of
a pattern like let's say the people um
every fifth person ends up being an
older person or something like that
that's a weird example I guess but like
if there's something going on where
there's a pattern to the population then
going every tenth person or every 15
person or whatever you might end up
missing parts of the answer being fooled
by the pattern so when you have a
pattern in the population a pattern in
the order of the population you might
get tricked by that when you are doing a
systematic random sample but if your
population is pretty evenly distributed
in the way you have it listed out then a
systematic random sample is a good
option for how you can collect sampling
collect a sample in a good way um that
doesn't fall into some the traps of not
knowing what your population actually
looks like so this second topic of this
video is pretty quick where to talk
about the difference between we'll talk
a little more about observational
studies so the first thing I want to
talk about here is a review of the
difference between an observational
study and an experiment and we know
already that in an observational study
treatments is not imposed
so that is the big thing going on there
is that you're not imposing treatments
when you do an observational study
that's the magical answer you just give
context and you're good to go
in an experiment you are manipulating
variable and imposing treatments on
subjects we have here just a breakdown
of two types of observational studies
there's what's called a retrospective
and a prospective or prospective
observational study in a retrospective
observational study so in both of these
scenarios right here you're not imposing
treatment you know what an observational
study is already in a retrospective
study the data has already occurred the
data you're collecting so the data being
collected has already occurred okay the
word retro makes you think old already
happened retro a to you stuff like that
stuff in the past so if you're doing a
retrospective study its data that's
already happened in the past and you're
kind of going and you're looking at it
and analyzing it so you could do a
retrospective study just by googling
things and finding data piles that are
already there online you could also be
bringing people in for a study and
asking them about stuff that has already
happened to them so if the data of
interest has already been collect
already happens even if you haven't
collected it yet that's a retrospective
so you whereas a prospective study
prospective observational study your
data collection will occur in the future
so data will occur in the future um so
maybe I want to test some sort of effect
on smoking or something that's going on
right now and I want to see what kind of
health effects that will had on you in
the future I'm clearly not going to
force you to smoke or not so it's not an
experiment but I could ask people to
identify if they're smoking or if
they're doing this or doing that and
then follow up with them in five years
ten years next month whatever and check
on their status then so that's just a
little bit more of a detailed look at
observational studies whether your data
happened in the past or whether it's
Click on any text or timestamp to jump to that moment in the video
Share:
Most transcripts ready in under 5 seconds
One-Click Copy125+ LanguagesSearch ContentJump to Timestamps
Paste YouTube URL
Enter any YouTube video link to get the full transcript
Transcript Extraction Form
Most transcripts ready in under 5 seconds
Get Our Chrome Extension
Get transcripts instantly without leaving YouTube. Install our Chrome extension for one-click access to any video's transcript directly on the watch page.
