Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
11 Cards in this Set
- Front
- Back
What is a repeated measures design? |
one in which the IV has 2 or more conditions, and each participant takes part in all conditions. |
|
What is a paired samples t-test? |
a parametrical statistical test used to compare 2 means from the same sample, or from paired or matched pairs used for parametric data |
|
What is the t-test doing? |
checking whether the overall difference in participants' scores (between 2 conditions) is large (significant) or small enough to assume it just reflects random error (i.e. not significant) |
|
When does t-value increase/ become larger? |
when there are large differences in scores, in a consistent direction. Larger t-values are more likely to be significant |
|
What does the t-test tell us? |
How likely it is that we'll observe a certain difference (or more) given the distribution of differences expected by H0 |
|
Why do we not really need to look at the t-test value? and why do we look at the p instead? |
can't tell from the value itself whether it's significant enough to reject the H0. |
|
express the following as fractions: p < .05, p< .005, p < .0005, p < .00005 |
1/20, 1/100, 1/1000, 1/2000 |
|
What is the criteria for using a paired-samples t-test? |
must have repeated measures or matched pairs design, interval or ratio data (with difference scores that aren't skewed), >=10 participants, and only two conditions |
|
For a t-test, how do we calculate the effect size? |
Use Cohen's d, which expresses effect size in terms of standard deviation units. |
|
What are the guidelines for describing effect size in a t-test? |
small effect size - d = 0.2 to 0.5 medium " " - d = 0.5 to 0.8 large effect size - d = more than 0.8 |
|
How do we report a t-test? |
t (df) = t (to 2 d.p.), p = .XXX (one/two-tailed), Cohen's d = (to 2 d.p.) |