Chi-squared Flashcards

1
Q

Broadly, a ‘goodness of fit’ can be described as?

A

How likely does the frequency match a theoretical expected distribution.

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

How do you get the test-statistic for a goodness of fit?

A

(observed - expected)2 / expected

this is run for every cateogory and then added together to get oen numbe

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

Does a larger x^2 indicate a better or worse fit beteen two groups?

A

A worse fit.

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

What is a goodness of fit comparing?

A

A goodness of fit compares observed counts to a theoretical distribution of expected counts in the population .

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

What does the shape of distribution under the null depend on?

A

degrees of freedom. k-1

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

what does the chi-squared distribution look like?

A

positivley skewed.

Karl Pearson pointed this out.

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

How do you estimate the expected frequencies in a chi-independence test?

A

Add up counts in each row. Add up counts in each column. Divide each row total by column total(s).

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

What is the equation for expected frequencies in a goodness of fit?

A

e = n x expected probability under the null

remember that this equation is run for every participant

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

What is the first thing you need to work out the calculate the expected frequency in a chi-independence?

A

We need the probability for each row. Total observations for that person / n

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

What is the equation for expected frequencies under the null? (chi-independence)

A

expected = total column observations x row probability

this is run for each row / individual

row probability is calculated: row total / n

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

How is df calculated for goodness of fit AND chi-independence?

A

Goodness of fit: rows - 1
Independence: (R -1) (C - 1)

Independence has two df values

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

What does effect size measure?

A

How big is the difference between data and null hypothesis.

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

How does a fisher exact test work?

A

Assumes rows and columns are fixed. Calculates the probability of obtaining a contingency table.

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

What are the conditions needed for a McNemar test, and why would you run it.

A

You would run this test if there was some relationship between your two groups. Outcome needs to be binary and have been measured twice.

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

What does a fisher exact test print?

A

a p-value

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What does a McNemar test print?

A

Chi-squared (x^2)
Df
P value

17
Q

What does a chi-squared print?

A

x^2
df
p-value

18
Q

What does a cramers V effect size rely on?

A

x^2 and sample size.