CK020 - Statistical Tests for Logistic Regression

Question 1

What does the ‘One-sample z-test’ test?

Answer 1

Tests if the mean of a single sample is equal to a known value or population mean.

Question 2

When do you use the ‘One-sample z-test’ test?

Answer 2

Use when the sample size is large (n>30), and the population variance is known.

Question 3

What does the ‘Two-sample z-test’ test?

Answer 3

Tests if the means of two independent groups are equal.

Question 4

When do you use the ‘Two-sample z-test’ test?

Answer 4

Use when comparing two independent groups with large sample sizes and known population variances.

Question 5

What does the ‘Chi-square test’ test?

Answer 5

Tests if there is an association between two categorical variables.

Question 6

When do you use the ‘Chi-square test’ test?

Answer 6

Use for contingency tables to assess relationships between categorical variables. Requires sufficient sample size.

Question 7

What does the ‘Fisher’s exact test’ test?

Answer 7

Tests if there is an association between two categorical variables.

Question 8

When do you use the ‘Fisher’s exact test’ test?

Answer 8

Use for contingency tables when sample sizes are small and Chi-square assumptions are not met.

Question 9

What are the assumptions of the ‘One-sample z-test’ ?

Answer 9

• Observations are independent of eachother
• Sample size is large enough to use normal approximation

Question 10

What are the assumptions of the ‘Two-sample z-test’ ?

Answer 10

• Observations are independent of eachother
• Sample size is large enough to use normal approximation

Question 11

What are the assumptions of the ‘Chi-square test’ ?

Answer 11

• Study groups must be independent
• There are 2 categorical variables
• Levels of the variables are mutually exclusive

Question 12

What are the assumptions of the ‘Fisher’s exact test’ ?

Answer 12

• Study groups must be independent
• Variables should be dichotomous
• Both row and column marginal totals are fixed in advance

Question 13

What does the ‘McNemar test’ test?

Answer 13

Whether the proportion of individuals who change from one category to another differs significantly from the proportion of those who change in the opposite direction.

Question 14

When do you use the ‘McNemar test’ test?

Answer 14

If you assess the patients both before and after treatment, the McNemar test can be used to compare the number of patients who:

Went from “not improved” to “improved” (0 → 1).
Went from “improved” to “not improved” (1 → 0).
If these two proportions differ significantly, it suggests that the treatment had a significant effect on the outcome.

Question 15

What are the assumptions of the ‘McNemar test’ ?

Answer 15

• Variables are categorical
• Data is matched
• Pairs are independent of eachother
• Levels of the variables are mutually exclusive