Chapter 14 Athena en lectures Flashcards

1
Q

One way ANOVA bij welke soort variabelen?

A
  • quantitative response variable
  • categorical explanatory variable (different groups)

(one way want: het is maar een soort explanatory variable)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Wat is de H0 bij een one-way ANOVA

A

H0: u1=u2=u3=u4….=ug

g= number of groups

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

wat is de Ha bij een one-way ANOVA

A

at least two of the population means are unequal (there is at least one difference)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

what does the test analyse

A

whether the differences between the groups could have occured by chance under the null hypothesis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

assumptions of one-way anova

A

quantitative response variable
at least two groups
random
equal standard deviations
normal population distribution for response variable
equal group sizes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

F=

A

between group/within group

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

N=

A

totaal aantal data punten

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

bonferroni method

A

a/k (k= aantal hypothesen)

5 hypothesen = 0,05/5= 0,01. dus dan een CI van 99% gebruiken voor de confidence van 95%.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

which should be larger: variability between or within groups?

A

the variability between groups should be larger than the variability within groups

between > within

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

wata zou de waarde van F moeten zijn

A

GROTER dan 1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

wat is de H0 van F

A

F is ongeveer 1 (want dan zijn er geen verschillen tussen de groepen)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Wat kan F NOOIT zijn?

A

negatieff!!!

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

wat is de centrale vraag over de variabiliteit

A

is de variabiliteit door sampling fluctuations, of door systematic differences?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

nog een keer variability: welke is groter?

A

between!!

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

uitzonderingen op de assumptions van ANOVA

A
  • niet normal: is oke als sample sizes groot zijn
  • equal standard deviations: is oke als groups zelfde grootte zijn, of als largest SD niet groter is dan 2x kleinste SD
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

hoe p value berekenen bij F statistic

A

1 - f.dist(f value; d1; d2; TRUE)

17
Q

wat is er anders bij ANOVA

A

je hebt meer informatie, dus CI is iets smaller bij Fisher method

18
Q

bonferonni

A

a/aantal testen = a*

19
Q

waarom bonferonni?

A

omdat je meerdere testen doet bij ANOVA (want meerdere groepen!!)

20
Q

twee mogelijke oplossingen hiervoor

A
  • bonferonni
  • tukey method
21
Q

type 1 error

A

reject the H0 when it is true

22
Q

type 2 error

A

accept the H0 when it is false

23
Q

voordeel one way analysis tov 3 significance tests per group

A

allows for type 1 error testing

24
Q

hoe bereken je t0,025 voor de fisher method??

A

t.inv(0,975; df) !!

25
Q

hoeveel pair wise comparisons

A

opschrijven! niet dezelfde (1-1) en ook niet 12 en 21.

26
Q

regression variables

A

beiden quantitative

27
Q

chi square test variables

A

beiden categorical

28
Q

independent samples t test

A

response = quantitative, explanatory = categorical

29
Q

ANOVA variables

A

response = quantitative, explanatory = categorical

30
Q

wilcoxon test lijkt op…

A

non parametric two samples t test

31
Q

kruskal wallis test lijkt op…

A

non-parametric ANOVA

32
Q

sign test lijkt op…

A

nonparametric paired t test

33
Q

voordeel en nadeel wilcoxon test

A

You can use the Wilcoxon test in the same situations as the t test for two independent samples. The advantage of the Wilcoxon test is that it has less strict assumptions. The population distribution does not have to be normally distributed. A disadvantage of the Wilcoxon test compared to the t test is that the Wilcoxon test has less power.

34
Q

kruskal wallis =/= chi square!! maar lijkt op….

A

anova (gebruikt alleen de chisquared distribution)