Lecture 12 Flashcards
Lecture 12:
What is a T-Test?
Compares the means of 2 different groups of samples (2 levels of independent variables)
Lecture 12:
If a t-test is properly designed, what can it determine?
Can determine causation
Lecture 12:
What are 3 examples of Independent Variables in t-tests?
1.) Groups of people
2.) task conditions
3.) time points
Lecture 12:
What 5 things need to be valid for an independent t-test?
1.) Data is interval or ratio (check measurement)
2.) Random selection (check research design)
3.) Normal distribution (skew & kurtosis values)
4.) Homogeneity of Variance (variance for 2 groups = similar)
5.) done prior to conduction test & may need to use a non-parametric test if assumptions are not met
Lecture 12:
What are the 6 steps to conducting an independent t-test?
1.) states the null & research hypothesis
2.) set the level of risk associated with the null hypothesis (alpha 0.05 if 2-tailed or 0.10 if 1-tailed)
3.) select appropriate test statistic ( & assumptions)
4.) compute the t-obtained test statistic value
5.) calculate the degrees of freedom
6.) determine the t-critical value
Lecture 12:
What is a Two-Tailed hypothesis?
When a difference exists between the groups so there is no prediction of whether the difference will be positive or negative
Lecture 12:
What is a One-Tailed hypothesis?
Predicts the direction of the difference between the groups
Lecture 12:
What is the T-obtained Test Statistic?
Measures the difference between the means
- similar to the Z value but corrected for small group sizes
Lecture 12:
Define “Degrees of Freedom”
The number of scores that are free to vary
—> df = n-1
Lecture 12:
How do you determine the critical value?
Critical value is determined using a table to find the specific df & alpha levels based on your hypothesis
Lecture 12:
What are 5 ways that you can improve your chances of rejecting the null hypothesis?
1.) increase degrees of freedom (larger sample, reduces t-critical)
2.) one-tailed a value (reduces t-critical)
3.) reduce the standard deviation (increase obtained t value)
4.) increase effect size (reduces t-critical)
5.) select a larger alpha value (reduces t-critical)
Lecture 12:
What is a repeated measures t-test?
Tests the same subjects twice which reduces standard error & increases degrees of freedom