WEBVTT

00:00:00.000 --> 00:00:03.486 align:middle line:90%
[MUSIC PLAYING]

00:00:03.486 --> 00:00:16.460 align:middle line:90%


00:00:16.460 --> 00:00:18.230 align:middle line:84%
Hello, and welcome
to a quick tutorial

00:00:18.230 --> 00:00:21.470 align:middle line:84%
on how to calculate a multiple
linear regression using

00:00:21.470 --> 00:00:24.140 align:middle line:90%
Excel and Statistician.

00:00:24.140 --> 00:00:26.510 align:middle line:84%
For our purposes, we're going
to go ahead and calculate

00:00:26.510 --> 00:00:29.030 align:middle line:84%
the same multiple linear
regression discussed

00:00:29.030 --> 00:00:30.980 align:middle line:90%
in the chapter on regression.

00:00:30.980 --> 00:00:33.920 align:middle line:84%
In this case, we wanted to
see if there was a combined

00:00:33.920 --> 00:00:37.340 align:middle line:84%
relationship between
communication apprehension,

00:00:37.340 --> 00:00:39.560 align:middle line:84%
willingness to
communicate, assertiveness,

00:00:39.560 --> 00:00:42.860 align:middle line:84%
and responsiveness with
an individual's belief

00:00:42.860 --> 00:00:44.570 align:middle line:84%
that everyone should
be required to take

00:00:44.570 --> 00:00:46.730 align:middle line:90%
public speaking in college.

00:00:46.730 --> 00:00:49.910 align:middle line:84%
So let's go ahead and
calculate this multiple linear

00:00:49.910 --> 00:00:50.840 align:middle line:90%
regression.

00:00:50.840 --> 00:00:52.250 align:middle line:84%
So to start with, I want
to go ahead and scroll over

00:00:52.250 --> 00:00:54.760 align:middle line:84%
here so that there's going
to be some empty space to get

00:00:54.760 --> 00:00:56.460 align:middle line:90%
us started for the results.

00:00:56.460 --> 00:00:59.750 align:middle line:84%
So you'll notice I am in the
Textbook Data Set Shortened No

00:00:59.750 --> 00:01:01.460 align:middle line:84%
Missing Data because
that's what we're

00:01:01.460 --> 00:01:03.800 align:middle line:84%
going to have to be
using to be able to run

00:01:03.800 --> 00:01:05.610 align:middle line:90%
a multiple linear regression.

00:01:05.610 --> 00:01:07.880 align:middle line:84%
So I want to go up here
to Regression Analysis.

00:01:07.880 --> 00:01:09.430 align:middle line:90%
I'm going to click on it.

00:01:09.430 --> 00:01:10.970 align:middle line:84%
And the first thing
I want to do is

00:01:10.970 --> 00:01:13.490 align:middle line:84%
select my Dependent
Variable, which

00:01:13.490 --> 00:01:15.970 align:middle line:84%
is going to be the belief
about public speaking.

00:01:15.970 --> 00:01:17.840 align:middle line:84%
So I'm going to go
ahead and click on that.

00:01:17.840 --> 00:01:21.530 align:middle line:84%
From there, I want to click
on Communication Apprehension,

00:01:21.530 --> 00:01:24.060 align:middle line:84%
Assertiveness, Responsiveness,
and then lastly,

00:01:24.060 --> 00:01:27.270 align:middle line:84%
down here, Big Willingness
to Communicate.

00:01:27.270 --> 00:01:30.200 align:middle line:84%
And so that is all that
you have to click on

00:01:30.200 --> 00:01:32.970 align:middle line:84%
to be able to run your
multiple linear regression.

00:01:32.970 --> 00:01:35.240 align:middle line:84%
So then I'm going to
click on Output Results,

00:01:35.240 --> 00:01:36.660 align:middle line:84%
and it already
has that selected,

00:01:36.660 --> 00:01:38.180 align:middle line:84%
so I'm going to
click OK, and then

00:01:38.180 --> 00:01:40.620 align:middle line:84%
it's going to go ahead and
paste the results for us.

00:01:40.620 --> 00:01:43.689 align:middle line:90%
So let's look at this real fast.

00:01:43.689 --> 00:01:45.230 align:middle line:84%
So the first thing
you're going to do

00:01:45.230 --> 00:01:47.390 align:middle line:84%
is, I'm going to kind of
bypass this information up

00:01:47.390 --> 00:01:51.325 align:middle line:84%
here because we're going to
come back to that in a second.

00:01:51.325 --> 00:01:52.700 align:middle line:84%
So we're going to
start down here

00:01:52.700 --> 00:01:57.335 align:middle line:90%
with R Squared, which is 0.1083.

00:01:57.335 --> 00:01:59.370 align:middle line:84%
You also have the
adjusted R Squared,

00:01:59.370 --> 00:02:01.370 align:middle line:84%
but we're just going to
be focused predominantly

00:02:01.370 --> 00:02:02.840 align:middle line:90%
on R Squared.

00:02:02.840 --> 00:02:07.460 align:middle line:84%
We also have our F Statistic
down here, which is 16.53.

00:02:07.460 --> 00:02:08.970 align:middle line:84%
We have our
probability, and you can

00:02:08.970 --> 00:02:10.720 align:middle line:84%
see that it's done in
scientific notation,

00:02:10.720 --> 00:02:14.180 align:middle line:84%
so we know that the probability
value or the p-value

00:02:14.180 --> 00:02:17.900 align:middle line:84%
is going to be p
is less than 0.001.

00:02:17.900 --> 00:02:19.610 align:middle line:84%
And so that is the
basic information

00:02:19.610 --> 00:02:21.447 align:middle line:90%
that we can pull from that.

00:02:21.447 --> 00:02:23.780 align:middle line:84%
Of course, if you want to
just get your regular R value,

00:02:23.780 --> 00:02:27.320 align:middle line:84%
all you have to do is take the
square root of the R Squared

00:02:27.320 --> 00:02:30.470 align:middle line:84%
function and it'll give
you your actual R value.

00:02:30.470 --> 00:02:30.990 align:middle line:90%
OK.

00:02:30.990 --> 00:02:33.500 align:middle line:84%
So once we know that it is
statistically significant,

00:02:33.500 --> 00:02:36.830 align:middle line:84%
because we had this lovely
probability value down here,

00:02:36.830 --> 00:02:40.100 align:middle line:84%
then we can come up here and
see which ones are possibly

00:02:40.100 --> 00:02:42.150 align:middle line:90%
statistically significant.

00:02:42.150 --> 00:02:42.682 align:middle line:90%
OK.

00:02:42.682 --> 00:02:44.390 align:middle line:84%
So the first thing
you're going to notice

00:02:44.390 --> 00:02:47.900 align:middle line:84%
is this variable right
here, which is 2.33,

00:02:47.900 --> 00:02:49.890 align:middle line:90%
and it goes to the negative 12.

00:02:49.890 --> 00:02:52.070 align:middle line:84%
Now that, again, if you
look in scientific notation,

00:02:52.070 --> 00:02:56.070 align:middle line:84%
is going to be p
is less than 0.001.

00:02:56.070 --> 00:02:58.820 align:middle line:84%
So we know that this one is
going to be statistically

00:02:58.820 --> 00:02:59.600 align:middle line:90%
significant.

00:02:59.600 --> 00:03:03.740 align:middle line:84%
And right next to it is
going to be the t-ratio.

00:03:03.740 --> 00:03:05.450 align:middle line:84%
And then next to
that is going to be

00:03:05.450 --> 00:03:09.650 align:middle line:84%
the coefficient over here,
which is a negative 0.177

00:03:09.650 --> 00:03:11.060 align:middle line:90%
or negative 0.18.

00:03:11.060 --> 00:03:13.220 align:middle line:84%
Now for our purposes,
that negative

00:03:13.220 --> 00:03:15.902 align:middle line:84%
there can actually be considered
kind of like a correlation.

00:03:15.902 --> 00:03:17.360 align:middle line:84%
So in this case,
we know that there

00:03:17.360 --> 00:03:19.940 align:middle line:84%
is a negative relationship
between communication

00:03:19.940 --> 00:03:22.400 align:middle line:84%
apprehension and the belief
that everyone should take

00:03:22.400 --> 00:03:24.050 align:middle line:90%
public speaking in college.

00:03:24.050 --> 00:03:27.560 align:middle line:84%
So again, this one right here,
big communication apprehension,

00:03:27.560 --> 00:03:31.040 align:middle line:84%
is accounting for unique
variance in this model.

00:03:31.040 --> 00:03:33.020 align:middle line:84%
Now don't forget--
the overall model

00:03:33.020 --> 00:03:35.210 align:middle line:90%
is statistically significant.

00:03:35.210 --> 00:03:38.390 align:middle line:84%
It just happens to be that
Big CA, or communication

00:03:38.390 --> 00:03:41.690 align:middle line:84%
apprehension total, is
accounting for unique variance.

00:03:41.690 --> 00:03:43.430 align:middle line:84%
So let's look at
the other variables.

00:03:43.430 --> 00:03:44.990 align:middle line:84%
Well the p-value
for assertiveness

00:03:44.990 --> 00:03:48.620 align:middle line:84%
is 0.28, so that is
greater than 0.05.

00:03:48.620 --> 00:03:51.920 align:middle line:84%
The p-value for
responsiveness is 0.51--

00:03:51.920 --> 00:03:54.200 align:middle line:90%
again, greater than 0.05.

00:03:54.200 --> 00:03:58.130 align:middle line:84%
And lastly, the p-value for Big
Willingness to Communicate is

00:03:58.130 --> 00:03:59.780 align:middle line:90%
0.58--

00:03:59.780 --> 00:04:03.350 align:middle line:90%
again, greater than 0.05.

00:04:03.350 --> 00:04:06.620 align:middle line:84%
So in this overall model, one
of the things that we know

00:04:06.620 --> 00:04:09.350 align:middle line:84%
is that only
communication apprehension

00:04:09.350 --> 00:04:14.210 align:middle line:84%
accounted for unique variance
in this overall model.

00:04:14.210 --> 00:04:17.060 align:middle line:84%
And so that is how you
can run and interpret

00:04:17.060 --> 00:04:21.979 align:middle line:84%
a regression analysis using
Excel and Statistician.

00:04:21.979 --> 00:04:25.329 align:middle line:90%
[MUSIC PLAYING]

00:04:25.329 --> 00:04:36.200 align:middle line:90%