Stats 7 - t tests

What is a t-test?

Inferential statistical test that compares one dependent (outcome) variable between two groups

How do z tests differ from t tests? How are they similar?

Z and t test both compare a dependent variable, but a t-test can do this with a smaller group and when the population standard deviation is unknown

What are degrees of freedom?

Refers to how much freedom statisticians have to vary the numbers in their data and still get the same parameters

What are the characteristics of a t curve?

A t curve is like a z curve but its cutoff is adjusted using degrees of freedom to account for sample size; like normal distribution but with more variance; lower peak, heavier tails with a smaller df

Population parameter

A numerical value that describes a characteristic of an entire population; we don’t have this for a t-test

When is a one-sample t-test used?

Used when we want to compare a sample to population estimate; testing a sample value against a claim value

What is the one-sample t-test formula?

Sample mean minus test value divided by sample error (SD divided by the square root of sample size)

What is the null hypothesis for a one-sample t-test formula?

Data point is the same as the sample mean

What are the assumptions of a t-test?

• Comparing two things

• Use NHST and p-value logic

• Random sampling

• Normally distributed

• Uniformity of variance

• Interval, ratio DV

What is the df for a one-sample t-test

n-1