Chapter 13
Uploaded by
GregorChapter 13
Uploaded by
Gregor13
Analise van Variansie
Analysis of Variance
(ANOVA)
Why is a physician held in much higher esteem than a
statistician?
A physician makes an analysis of a complex illness whereas a
statistician makes you ill with a complex analysis!
What did the new statistics professor do when his lecture on
analysis of variance flopped in front of a large class?
He had to go OVA ANOVA ANOVA it again!!!
Analise van Variansie (ANOVA)
Analysis of Variance (ANOVA)
In hoofstuk 9 het ons gekyk na hipoteses oor een
steekproef, in hoofstuk 10 het ons 2 steekproewe
vergelyk en in hierdie hoofstuk stel ons belang in
hipoteses oor meer as 2 steekproewe.
In chapter 9 we looked at hypotheses about a single
sample, in chapter 10 we compared 2 samples to one
another, and in this chapter we are interested in
hypotheses about more than 2 samples.
Studeer 1ste, 2de, en 3de jaar studente ewe veel per
week?
Do 1st, 2nd, and 3rd year students study an equal amount
of time per week?
Daar is 5 moontlike verpakkings vir jou nuwe produk,
watter een hou persone die meeste van?
There are 5 different packagings you can choose from
for a new product, which one does consumers like more?
Kan kies tussen drie verskillende rekenaar vervaardigers,
watter een is die mees betroubaarste?
You can choose between three different computer
manufacturers, which one is the most reliable?
1
Ons het ‘n numeriese veranderlike wat ons meet, wat
verdeel word in verskillend groepe deur ‘n katergoriese
veranderlike.
We are measuring a numerical variable that is divided into
different categories by a categorical variables.
Studie Tyd Studie Jaar
Study Time Year of Study
Smaakskaal Tipe verpakking
Taste Scale Type of Packaging
Tyd voordat Breek Tipe rekenaar
Time to Failure Type of Computer
Ons wil weet of die verskillende groepe van die kategoriese
veranderlike gemiddeld dieselfde waarde het vir die
numeriese veranderlike.
We want to know if the different groups of the categorical
variable have on average the same value for the
numerical variable.
1ste 2de 3de
6 8 12
4 9 34
1 5 23
23 5 31
4 6 9
μ1 = μ2 = μ3 ????
Een-Rigting ANOVA
One-Way ANOVA
Een kategoriese veranderlike.
One categorical variable.
H0: μ1 = μ2 = … = μk
H1: Ten minste een groep se gemiddelde verskil van die
ander.
At leas one groups’ average differs from the rest.
2
Presies dieselfde stappe in die hipotese proses as in vorige
hoofstukke, die toetsstatistiek is net wiskundig moeiliker
om te bereken en ons gebruik die F-tabel om kritieke
waardes af te lees.
Exactly the same hypothesis test steps as in the previous
chapters, the test statistic is just mathematically more
complicated and we use the F-table to read off cricital
values.
Ons gaan ‘n ANOVA tabel gebruik om die toetsstatistiek te
bereken. Moet slegs tabel kan voltooi as seker waardes
ontbreek. Nie formules gebruik in hoofstuk 11.
We are going to use an ANOVA table to calculate the test
statistic. Only be able to complete table if certain values
are missing. Not going to use the equations in chapter
11.
3
4
Een-Rigting ANOVA Tabel
One-Way ANOVA Table
Oorsaak VG SK GSK
Source DF SS MSS F
Treatment k–1 SSTr
Error n–k SSE
Total n–1 SST
Toetsstatistiek
Test Statistics
Kritieke waarde
Critical Value
EXAMPLE 13-1 (p. 520)
Oorsaak VG SK GSK
Source DF SS MSS F
A ??? 382.79 ??? ???
