This content explains the principles of experimental design, specifically focusing on how to conduct an experiment to test the effect of physical exercise on pulse rates and how to interpret the results, emphasizing the concept of statistical significance.
Mind Map
Click to expand
Click to explore the full interactive mind map • Zoom, pan, and navigate
okay so the bulk of this lesson has come
from our live thing when we actually ran
our experiments whether that was in
class or I'm recording this right now in
2020 so whether that ends up being
online for the coming school year but
we've already done some sort of
experiments based on our pulse rates and
I am recording this video before we've
actually done the experiments so I don't
know exactly how things played out right
here so I'm gonna talk in kind of a
generic sense about one type of
experiment I've done before but it
should still be helpful to hear how this
is organized even if we did something
slightly differently if you're like this
isn't helpful you're talking about a
different thing that we actually did in
class skip to the ends where I talk
about statistical statistical
significance because that's the most
important part of this lesson so
generally in class I don't do the
caffeine problem it's um kind of a
hassle to get it all set up and get the
caffeine poured and the grocery store to
get the caffeine and kids sometimes
aren't comfortable drinking soda in
class so it's just I found that it's
better to do other things to measure
pulse rate instead so one thing that
I've done which is what I'm gonna run
with in this example is I can look at
whether physical exercise increases
pulse rates would increase his pulse
rate which seems obvious that it would
but it's still kind of nice to do
something and to see results that
actually are significant so let's say
for my problem right here what I'm gonna
run with again we may have done
something differently in class I don't
know yet I can't see into the future but
what I'll generally do is something with
exercise so I'm gonna do three groups
here to be real fancy with it we're
gonna have no exercise at all we're
gonna have a group that does 60 seconds
of jumping jacks and we're gonna have a
group that does 60 seconds of push-ups
and then we're gonna see between those
groups if there is a difference in the
pulse rates before and afterwards so
let's talk about I'm not gonna have you
necessarily write these but think about
how we can make sure each of these four
principles of experimental design is
addressed within the context of this
study so think about that for a second
then I'll talk my way through it right
now comparison I've got three separate
groups and I'm going to compare my pulse
rates between the three groups that's
great I've done that and I need to do
that because if I didn't I wouldn't know
if the change I saw was because of the
exercise or because of something else
going on like my lesson etc etc so I've
talked about before on having different
groups that have all gone through the
same stuff is important I can't just
have everybody do the same exercise the
next thing I would need to do is have
random assignments so in my class
hypothetical class of 18 kids right here
I would randomly assign six kids to each
that's an arrow six kids to each of
those three groups and I would make sure
I did it randomly so like all the kids
who are in the same sports don't end up
in the same group and they're more fit
than everybody else and stuff like that
so by randomly assigning even amounts to
each of my groups I'm ensuring that my
groups aren't inherently unbalanced
which prevents confounding from Krank
control so with control I want to keep
everything else the same
so one way of doing that the other two
groups with jumping jacks and push-ups
are like getting up out of their chairs
and doing things I should probably have
my no exercise kids get up and stand up
anyway so it's not like they're still
sitting in a chair chillin they're still
doing something umm if I just had them
sitting there that I wouldn't know maybe
just by simply getting up there is a
difference being cause I'm recording
this video summer of 2020 so I still
don't know yet in 2020 if my kids are
going to be with me in class when we do
this lesson or if we're gonna do this
virtually so is experiments if we're
doing it from our own homes virtually
would be terrible in terms of control
because there are so many things that
are different in your house which is
your house versus your house so if we
are doing this virtually my experiments
had no control at all even if we're in
class I want to make sure like I said my
one group gets to like still stand up so
that's going on on I shouldn't have one
group in like hotter
part of the room or out the hallway
where the temperature might be different
etc etc etc make everything the same
that I can
and finally replication six people per
group is not great not making any
scientific journals with this amount of
data right here but I don't really have
control over that and my problem I can't
influence the number of people I have so
my replication is not ideal going in all
right so those are kind of how I am
addressing we're not addressing my
experimental design and I could make a
little diagram again but I've done that
before I don't really want to have you
guys write one of these whole deals the
idea is again I would take my 18 people
and I would randomly assign six of them
to each treatment so random assignment
you have six here six here 6 here I said
I didn't want to do it but it looks like
I'm kind of doing it anyway these people
can get treatment one which is no
exercise these people can get treated
two which is the jumping jacks and then
these people who get treatment 3 which
is push-ups it doesn't matter which I
call each of these because people were
randomly assigned so as long as I don't
like look at the people on the groups
and then decide what they get it's ok
these are in order but after I actually
do the exercise then what I'm gonna do
with each of the groups is I am going to
measure the pulses and then I'm going to
compare my treatment groups see if
there's a difference between the push-up
group and the jumping-jack group and the
no treatment group and that's basically
how my experiments gonna run so um
generally when I do these in class it
works a little better with exercise than
it used to with caffeine with caffeine I
would rarely see a difference between
the groups probably cuz I wasn't giving
enough caffeine and I wasn't doing a
good enough job controlling it's
actually very hard to execute a good
experiments where other things don't
come into play there's always things
that pop up that well that kind of mess
with my results so it takes a lot of
deliberate thoughts or and I put into
this to get good results from experiments
experiments
but let's do a hypothetical here so
again I don't know what happens when you
actually did the experiment with me but
just like before I'm gonna have a graph
or the change in pulse rates and I'm
gonna have 0 here at the middle so I
have my dot plots kind of stacked on top
of each other so you can see what's
going on we're gonna have our no
exercise we're gonna have our jumping
jacks and we're gonna have our push ups
let's say my no group hypothetically
looked like this so those are my six
people right there
and then let's say my jumping jacks
group hypothetically it looks like this
and then my push up group looks like
this just making stuff up right here in
this situation if I were to look at
these results what you want to look for
is little over that if you have lots of
overlap between your groups it's not
really convincing that there's a
difference between this and this if you
see lots of separation between the
groups that does suggest that maybe
there's a difference so if I were to
look at no exercise versus jumping jacks
fair amount of separation right there
not too much overlap same with no
exercise versus push-ups so in my
hypothetical right here I would say no
vs. either jumping jacks or push-ups
there is little overlap and little
overlap means there probably could be a
difference meanwhile if you look at
jumping jacks verse is the push-ups
jumping jacks versus push-ups have a lot
of overlap in them yes these people
right here are higher than anybody in
the push-up group but there's an awful
lot of overlap between these so this has
lots of overlap
and in general when there's lots of
overlap we're not convinced of a
difference so that is what you want to
look for in a nutshell when you're doing
these is there lots of overlap or is
there nots and this is so much more
complicated like they're not complicated
but complex then I'm making it up to me
right now we will talk a lot a lot a lot
this year about deciding whether results
are statistically significant this is
like our introductory chapter just kind
of dip our toes in the water aspect of
this look to see if there's separation
between your roots or not we will
revisit this many times over the course
of our time together in AP stats so
based on the results of the experiment
is there evidence that the one of the
groups has increased pulse rates and
then I'm gonna say again I'm just going
to write down a reminder to look for
overlap we want to see if there is
overlap or not and then assuming there
is a difference between the two groups
or the only three groups here what are
the possible explanations this is
getting at some of that more complicated
stuff that we will address a lot more as
class progresses here but anytime you
see a difference there are two
possibilities option one is there
actually is a difference caused by the
treatments so it's possible that I saw a
difference between jumping jacks and no
exercise because the jumping jacks
actually did cause a difference the
second option is sometimes a little
harder for kids to understand there is
no difference so there's no real
difference and the results occurred by
chance my face is in the way there sorry
so James um technically the chance is
due to rain
assignments you know by chance due to
random assignments so let me explain
what that means and I don't want to do a
super ton with this right now because we
will hit this again later in future
chapters a few true lessons but let's
say I had my no group that was like this
and I had my jumping jacks group it was
like this there's a lot of separation
there so this would lead me to believe
wow all these people did better than all
these people that happened probably
because the jumping jacks made a difference
difference
it's especially appearance if you have
only a little bit of data so let's say I
took the same problem but I had four
people in my study two per group so I
had these two kids right here
four actually let me close to zero so I
got my two people right here for this
group and I got my two people right here
for this group okay when you have too
little data the chances are like okay
maybe this person right here their pulse
rate was gonna go up either way no
matter what group they got putting in
they just happen to get put in this
group by random chance if they have got
put in the no root they still would have
increased the same amount because that
was just destined for happen anyway so
when you have too little data you can
make the arguments that the results
you're seeing are because of where
people were randomly assigned and not
because of the treatment itself that is
largely fixed by having enough data and
seeing enough separation and what's
going on but that's the basic idea going
on right there all right so the last
thing we need to talk about how can we
determine if the evidence is convincing
the evidence is considered convincing if
we can replicate it if we can do it
again and again again so we can do this
through replication or in other words
repeated trials so that is what it takes
to be able to decide that results are
convincing you do it again and see if
the same thing happens you can actually
conduct the experiments again or you can
use something called a simulation
you saw on the first day of school and
we will practice this a super time
between now and then
now results less than here results are
considered statistically significant if
they are unlikely to occur due to chance
so results are statistically significant
important vocab concept we will talk
about a lot in this class their
significance if they are unlikely to
occur by chance so if I were to deal out
my groups again and not reproduce my
results right here it may be that the
results I saw the first time were just
you the chance of how things were the
sign so you have to be able to repeat
the process and do it again to see if
it's due to chance or if it's actually
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.
