Repeated Measures t-test EdPuzzle Flashcards
Characteristic: Single Sample t-Test
Purpose: Compares one sample mean to a known population value
Research Question: “Does this sample differ from a known value?”
Design Type: One sample design
Number of Groups: 1 group
Time Points: 1 time point
Null Hypothesis: H0: μ = μ0
What’s Being Tested: Difference between sample mean and population value
Formula: t = (M - μ0)/(s/√n)
Degrees of Freedom: df = n - 1
Effect Size (Cohen’s d): d = (M - μ0)/s
Characteristic: Repeated Measures t- Test
Purpose: Compares means from the same group under different conditions or time points
Research Question: “Is there change across time/conditions within the same individuals?”
Design Type: Within-subjects design
Number of Groups: 1 group measured twice ( or more)
Time Points: 2+ time points (or conditions)
Null Hypothesis: H0 = μD = 0 ( or H0: μ1 = μ2)
What’s Being Tested: Mean of difference scores
Formula: t = MD/(sD/√n)
Degrees of Freedom: df = n - 1
Effect Size (Cohen’s d): d = MD/sD
In repeated measures t-tests, what is being measured according to the slide?
Mean of difference scores
What is the formula for degrees of freedom for repeated measures t-tests?
n - 1
Repeated Measures (paired, matched)
A repeated measures design involves measuring the same participants at different times or under different conditions. Think of it as a “before and after” comparison:
- A psychologist measures anxiety levels before and after therapy
- A teacher assesses student performance levels before and after a new teaching method
- A researcher tracks attitudes before and after viewing a public service announcement
The power of this approach is that each person serves as their own control, allowing us to focus specifically on the changes that occur.
Steps to conducts a Repeated Measures Design
- State the hypothesis
- Determine the critical value
- Calculate the test statistic (in this case Repeated Measures t - test)
- Make a decision
- Calculate the effect size
What is an example of a repeated measures design?
Depression scores before and after treatment
The null hypothesis states that the MEAN DIFFERENCE between the before and after scores will be _____?
= 0