YouTube Transcript:
Measure 20 Scatter Plot
Skip watching entire videos - get the full transcript, search for keywords, and copy with one click.
Share:
Video Transcript
next would be scatter plot so of course
scatter plot is another graphical a very
powerful graphical analysis tool so this
is a scatter plot is a diagram to
present the relationship between two
variables of a data set most of the
cases it's used for quantitative data
sets so for example if you have a
quantitative y paired with the
quantitative X so that's going to be
using a scatter plot so a scatter plot
consists of a set of data points so on a
scatter plot a single observation is
presented by a data point with its
horizontal position equal to the value
of one variable that could be found on
the other position which is the vertical one
so a scatter plot helps uh us to
understand whether the two variables are
related to each other or not
how is the strength of their
relationship what is the shape of the
relationship what is the direction of
their relationship and when outliers are
let's try to use a example
oops sorry
so let's try to create another worksheet here
so we're talking about in this case
we're talking about the mileage or the
miles per gallon
so pretty much this is the fuel yield if
you own a car you you are I I would
guess in this time of uh the year you
would be very conscious about your uh
fuel yield
okay because uh the the cost of the fuel
right is too high right now I think
diesel is about at about 82 per per liter
liter
so this is my um miles per per gallon
and this is the weight of the car so
using we can use
histogram okay I'm sorry we can use we
can use scatter plot right to
check whether this particular two
variables that are quantitative are
associated with each other or correlated
to each other so correlated is a
correlation rather is a term that we use
to uh to potentially Express that there
is a or to express but if there's a
potential Association or relationship
between the two variables that are being
paired okay
so what what we can do is we go to stat
sorry graph
if we go to graph
click the scatter plot here
and then there are there are so many
options so you might want to use this
third one here because this gives you
the idea of where the line or what we
call the fitted line or the line that we
draw in the clustering of data points
that we have in a scatter plot
is located no you don't have to imagine
if you use this you'll imagine where
should I put this line but if you use
this it's uh the software will give you
and then you put the let's say for
example that that we assume that uh
uh the miles per gallon is affected by
the weight of the car so remember Y is a
function of X so when you're doing root
cause analysis or hypothesis testing
this is a very important concept that
you always have to remember and then you
put uh you click OK and we will have
this view here
so from
I think I don't know if how many data
points this is
but this is a lot I mean this is
about 398 paired data points so from
that view we see that uh the line is
actually leaning to the left okay so the
slope is the slope of the line is the
one that we use to check the possible
correlation so if you see something like
this it denotes a negative
correlation so why negative correlation
again correlation is a term that we use
to ex Express potentially the possible
Association or correlation or
relationship between two paired
variables so what we do uh by hand is we
plot the data points and then we plot
this red line here it's called the
fitted line so the fitted line from the
treated line we try to draw the slope
and depending on the slope we will
assess whether there's a possible
positive correlation
uh possible negative correlation or no
correlation at all so if you see this
kind of slope uh the line is leaning to
the left side it means that there's a
possible negative correlation what does
it mean so as the weight of the car or
the vehicle increases so from zero let's
say to this number
so okay what happens to the miles per gallon
gallon
so as this one increases
the miles per gallon what
geek creases so that is why it's a
and with this you can actually
uh identify with this concept you can
identify important factors that are
affecting your certain kpi or metric
okay so this is highly used to establish
a an association or relationship or
correlation but this is just the visual
or graphical part of it there is a thing
called correlation and regression
analysis that we will be covering on the
analyze space but for now
um as part of also as part of the seven
basic UC tools we are trying to explore
this as part of the graphical analysis
tools so that's for negative correlation
so if you see um something something uh
that is opposite of this let's say if
the line of I mean the clustered of data
points are just like this foreign
it's something like that so the slope is
here then this signifies a possible positive
positive
correlation we're in as one increases
the other also
increases okay
and there's also a case wherein there
are no lines that could be drawn since
the data points are two dispersed so
that denotes that your
um data point
only shows no correlation at all
so that's pretty much how we use the
concept of scatter plot when we are
dealing with graphical analysis okay
okay
so for example if you want to check if
output is dependent on the number of
attend a number of people who are
present in your workplace or the
attendance rate so you can do this so if
let's say for example if the inventor
level of your
medicine at your Pharmacy is let's say um
um
dependent on the number of
inpatients or outpatients
so something like that then you can use
this and for as long as there is a
quantitative the quantitative nature of
both Y and X are are there so you can
use this so I I would suggest that uh
forever because this is paired data
points okay so as much as possible say
for example if you talk about the the
output for this day one the data should
be paired with uh of course the
attendance rate for this day so not
pulling only the data let's say for
example if a different data set from
from another team that contains uh the
output data and then you pull in another
data that is from another team that
contains the attendance data still there
are both qualitative but when you try to
pair them there are not actually
on they don't have the same reference
maybe uh let's say for example this data
could be this First Data could be uh for
for the output of let's say Monday but
the data that you were able to pair from
that particular random data set that you
were you tried to pull that from that
group uh attend who provides attendance
report it's from a Tuesday so there's
gonna be some disconnect so as much as
possible we try to get really the paired
data point
and this this is just my uh you know
idea and best practice since uh what
what I'm expecting is there's gonna be
some sort of you know
um false context
because any data that you can pair would
would pretty much have like some sort of
relationship or correlation but it's
very important that we stick to the
context so and then context isn't just
about when you're trying to interpret it
but also when you're trying to get the
data that you will use for certain tools
and and you know gaps either graphical
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.
Works with YouTube, Coursera, Udemy and more educational platforms
Get Instant Transcripts: Just Edit the Domain in Your Address Bar!
YouTube
←
→
↻
https://www.youtube.com/watch?v=UF8uR6Z6KLc
YoutubeToText
←
→
↻
https://youtubetotext.net/watch?v=UF8uR6Z6KLc