Equality of Proportions (chi squared test) Flashcards

1
Q

What tests would you used to compare categorical/non-parametric data?

A

Chi squared tests

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

What test would you used to see whether the population proportion is equal to a certain value?

A

One sample χ2 test

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

What assumptions are needed for the one sample chi-square test?

A

▪️Observations are randomly and independently drawn
▪️Number of cells (categories) with expected frequencies less than 5 are less than 20%
▪️The minimum expected frequency is at the very least 1

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

What is the df for a one sample chi-square test?

A

categories-1

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

What test would you used to see whether the proportions in the population of one variable change based on another variable (are the proportions different) ?

A

Person’s chi square test

(Independent samples)

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

What is the null hypothesis of a person’s chi square test?

A

There is no association between the two variable (proportions are the same)

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

What assumptions are needed for Pearson’s chi square test?

A

▪️Observations are randomly and independently drawn
▪️Number of cells with expected frequencies less than 5 are less than 20%
▪️Minimum expected frequency is at the very least 1
▪️Observations are not paired

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

When do you check assumptions for chi-square tests?

A

AFTER the test

Look at bottom of the table

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

How do you know whether to compare proportions by row or column for a Pearson’s chi square test?

A

You cannot statistically compare proportions that add up to 100%

If they add by column, compare them by row!

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

We cannot compare proportions that….

A

Add up to 100%!!

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

Why can’t you compare proportions that add up to 100?

A

▪️Only the descriptive indices of one of the variables
▪️Doesn’t consider the relationship between variables

E.g. 15 of the 30 who died were vaccinated. Reporting it as 50% is just descriptive of those who died, it doesn’t consider the rest of the population who didnt

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

What is the df for a Pearson’s chi square test?

A

(c1-1)(c2-1)

For a 2x2 grid this is 1

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

What test do we use to see whether the proportion of a variable changes based on another matched variable (is proportion in one condition different to that in the same/paired group in another condition?

A

McNemar χ2 test

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

What is the null hypothesis of a McNemar chi square test?

A

There is no association between two (paired) variables

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

What assumptions are needed for the McNemar chi square test?

A

▪️Observations are randomly and independently drawn
▪️At least 25 observations in the discordant cells
▪️Data are paired

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

What are the discordant cells?

A

Cells that have changed from yes to no or vice versa

17
Q

What are the concordant cells?

A

Cells that haven’t changed (still yes or still no)

18
Q

Which cells do you use for the McNemar chi square test?

A

Only the discordant cells

19
Q

What symbol is used for means?

A

μ

20
Q

What symbol is used for proportions?

A

π