BI250 Final Exam

Biostatistics Fundamentals | Dr. Salsbury | SP26

Frequentist Probability

The proportion of times an event will occur if we repeat the same process over and over again

The Multiplication Rule

Gives the probability of both event A and event B occuring | Pr(A and B) = Pr(A)*Pr(B)

The General Addition Rule

Gives the probability that event A or event B will occur, but not both | Pr(A or B) = Pr(A) + Pr(B) - Pr(A and B)

Conditional Probability

The probability of an event occuring given that a certain condition is met

The Fallacy of Transposed Conditional Probabilities

The incorrect assumption that P(A|B) = P(B|A)

P-value

Represents the proportion of times you would expect to get your results given that the null hypothesis is true

Type 1 Error

The false positive | Falsely rejecting the null hypothesis

Type 2 Error

The false negative | Falsely failing to reject the null hypothesis

Statistical Power

The probability that a test will correctly identify an effect when one actually exists

Binomial Test: Type of data, type of variable(s), question being tested

Counts or proportions of one variable that has two categories (eg. coin toss outcome is a variable, heads and tails are the categories), evaluates the actual number of successes against the hypothesised probability of success

Chi-Squared GOF Test: Type of data, type of variable(s), question being tested

Frequency/count data, 2 counts of 1 categorical variable, compares observed outcome to expected

Chi-Squared TOA: Type of data, type of variable(s), question being tested

Frequency or count data of 2 distinct categorical variable, determines if they are related

1 Sample T-Test: Type of data, type of variable(s), question being tested

1 continuous numerical variable, comparing sample mean to null hypothesis

2 Sample T-Test: Type of data, type of variable(s), question being tested

2 distinct continuous numerical variables, comparing the means of each sample to each other (significantly different)

Paired T-Test: Type of data, type of variable(s), question being tested

Compares the differences between two samples (eg. a before and after measurement) to the null hypothesis (which is that the difference is zero) for a continuous numerical variable

ANOVA: Type of data, type of variable(s), question being tested

Compares the means of more than two experimental groups

What are the assumptions of the chi-squared tests?

1. Data are frequencies/counts that have been placed in mutually exclusive categories- in other words you can’t be “both”

2. Observations are independent

3. Expected frequencies are not too small! No EFs should be less than one, and no more than 20% of categories should have EF less than 5

What are the assumptions of the t-tests?

1. Sampling is random

2. Data is normally distributed

3. Variances are not significantly different between two groups of a two sample test