What is the function of a t-test?
To compare group means to determine if the difference in means is likely to be caused by manipulation or random variation
How do you determine which kind of t-test to use in independent measures?
Levene’s Test
* Checks the assumption of homogeneity of variance
* Insignificant → Standard t-test
* Significant w/ normal distribution → Standard t-test
* Significant w/o normal distribution or ordinal data → Mann-Whitney U Test
What is homogeneity of variance?
The parametric assumption that each group has similar amounts of variance, meaning the analysis is insignificant (p > .050)
How do you run an independent measures t-test in SPSS?
- Analyze → Compare Means → Indep. Samples T Test
- DV → Test Variable
- IV → “Define Groups”
How do you write up the results of a Levene test statistic?
F = X.XX, p = X.XXX
What happens to df in SPSS if the assumption of homogeneity of variance is violated?
df are adjusted and made smaller to make test more accurate and conservative/less significant
What does a positive/negative t-value reveal?
Which condition has higher or lower scores, depending on the order each group was coded
How do you write up Levene test results?
t (df) = t-stat, p = .XXX
* Report M and SD
How do you write up Whitney test results?
U = XX.XX, z = XX.XX, p = .XXX
* Report mean ranks
What is sphericity?
The repeated measures equivalent to homogeneity of variance
* requires 3+ conditions
What is the Wilcoxon test?
The nonparametric equivalent to a repeated t-test for ordinal data or data that is not normally distributed
How do you write up Wilcoxon results?
There is a (significant/insignificant) difference in (variable observed) depending on (condition) (z = X.XX, p = .XXX), (report which variables had which effect) (M = XX.XX, SD = XX.XX)
What is the function of a repeated measures t-test (AKA paired samples t-test)?
To determine if there is a significant difference in the means of 2 measurements taken from the same group of participants
What is the function of a one sample t-test?
To compare the mean in a single, normally distributed sample is significantly different compared to a reference value
How do you report a one sample t-test result in APA?
t (df) = test stat, p = .XXX; M = XX.XX, SD= XX.XX)
How is a one sample t-test different from an independent measures t-test and repeated measures t-test?
A one sample t-test compares one group’s mean to a reference variable (Test value), while independent measures t-test compares 2+ group means to each other, and repeated measures t-test compares 2+ measured condition results to a single group
When is a p value considered significant vs. insignificant?
p < .050 = significant in difference
p > .050 = insignificant
