WEBVTT

00:00:00.000 --> 00:00:00.630 align:middle line:90%
[MUSIC PLAYING]

00:00:00.630 --> 00:00:03.990 align:middle line:84%
Analysis of variance,
or ANOVA, is a tool

00:00:03.990 --> 00:00:07.050 align:middle line:84%
used in statistics that works
by splitting variability

00:00:07.050 --> 00:00:10.380 align:middle line:84%
within a data set
into two parts--

00:00:10.380 --> 00:00:13.740 align:middle line:84%
systematic factors
and random factors.

00:00:13.740 --> 00:00:16.680 align:middle line:84%
Systematic factors have
a statistical influence

00:00:16.680 --> 00:00:20.280 align:middle line:84%
on the data set, while
random factors do not.

00:00:20.280 --> 00:00:22.920 align:middle line:84%
The ANOVA test is used
in regression studies

00:00:22.920 --> 00:00:25.770 align:middle line:84%
to determine the influence
independent variables have

00:00:25.770 --> 00:00:27.870 align:middle line:90%
on the dependent variable.

00:00:27.870 --> 00:00:30.180 align:middle line:84%
You'll also hear ANOVA
called the Fisher

00:00:30.180 --> 00:00:31.770 align:middle line:90%
analysis of variance.

00:00:31.770 --> 00:00:35.550 align:middle line:84%
It's the extension of the
earlier t and z tests.

00:00:35.550 --> 00:00:37.740 align:middle line:84%
To calculate the
ANOVA coefficient,

00:00:37.740 --> 00:00:40.980 align:middle line:84%
use the formula F equals
the mean sum of squares

00:00:40.980 --> 00:00:44.190 align:middle line:84%
due to treatment over the mean
sum of squares due to error.

00:00:44.190 --> 00:00:46.980 align:middle line:84%
The result, also
called the F ratio,

00:00:46.980 --> 00:00:49.770 align:middle line:84%
allows for the comparison
of multiple groups of data

00:00:49.770 --> 00:00:53.730 align:middle line:84%
to identify variance
between or within samples.

00:00:53.730 --> 00:00:56.160 align:middle line:84%
This is what's called
a null hypothesis.

00:00:56.160 --> 00:00:59.210 align:middle line:90%
[MUSIC PLAYING]

00:00:59.210 --> 00:01:01.090 align:middle line:90%