Choosing a statistical test.

Question 1

What is a quantitative variable?

Answer 1

Numbers which mean something in the real world.

Continuous // Discrete variables.

Question 2

What is a categorical variable?

Answer 2

Numbers which assign participants to groups.

Nominal // Ordinal variables.

Question 3

What is the difference between paired and unpaired data?

Answer 3

Paired means there is some natural link between 2 observations whereas with independent this means the probability of A occurring DOES NOT influence the probability of B occurring.

Question 4

What is a t - test used for?

Answer 4

Used to test the difference between the means.

Question 5

What is the purpose of ( independent ) two sample t tests?

Answer 5

Compare the means between 2 independent groups.

Question 6

What is the purpose of ( paired ) two sample t tests?

Answer 6

Compare the means of the same measurement made at 2 different time points in one group.

Question 7

What is the purpose of one sample t - test?

Answer 7

Compares the mean of sample with estimated population mean.

Question 8

What is the H0 for a t - test?

Answer 8

There is NO difference in the continuous variable between groups (independent) or time points (paired).

Question 9

What is the H1 for a t - test?

Answer 9

There is A difference in the continuous variable between groups (independent) or time points (paired).

Question 10

What happens if the p - value is <0.05?

Answer 10

You reject the null hypothesis : there IS a difference in the continuous variable between groups.

Question 11

What are the assumptions for a t - test?

Answer 11

Data is RANDOMLY sampled from the population of interest.
Variable = One categorical variable which has 2 group (independent) or timepoints (paired) AND one continuous dependent variable.

Question 12

What is a chi - squared test?

Answer 12

Calculates the chi - square statistic by comparing the observed value to the expected value.

Question 13

How do you calculate the degrees of freedom?

Answer 13

(r-1) * (c-1)

Question 14

What are the assumptions for a chi - square test?

Answer 14

Data is frequencies // counts
Study groups are independent and each participant contributes data to only one cell.
Value of cell EXPECTED should be >5 in >80% of cells and NOT zero.

Question 15

In chi square, what does the null hypothesis (H0) show?

Answer 15

Categories are INDEPENDENT // NOT RELATED

Question 16

In chi square, what does the alternate hypothesis (H1) show?

Answer 16

Categories are NOT INDEPENDENT // ARE RELATED

Question 17

What is the correlation coefficient?

Answer 17

Describes the strength and direction of the LINEAR relationship between 2 CONTINUOUS variables.

Question 18

What does p and r stand for in correlation coefficients?

Answer 18

P = population parameter
r = sample statistic

Question 19

What are the assumptions for Pearson’s correlation?

Answer 19

2 continuous variables which are paired.
Normal distribution = NO OUTLIERS.
Relationship is LINEAR.

Question 20

What does homoscedasticity mean?

Answer 20

Shape of scatter plot is tube - like NOT cone shape.