T-tests L5 Flashcards
What does a one sample T test aim to do?
Compare the mean of one sample group against a norm/population norm
What are the assumptions of a one sample T test?
The dependent variable must be measured on an interval or ratio scale
The sample is randomly selected from the underlying population
The scores form a normal distribution
What does an independent samples T test aim to do?
Compare the observed differences between the means of two independent samples (e.g. Experimental vs control, group A vs group B)
What are the assumptions of an independent samples T test?
The independent variable must be a dichotomous variable
The dependent variable must be measured on an interval or ratio scale
The scores should be normally distributed
The scores should have approximately equal variances
The sample sizes of each group should not be massively different (ideally not by more than 20 people)
How do we determine the statistical significance of the T value obtained?
Tobt is compared with the minimum t value of the region of rejection (T-crit)
If Tobt > Tcrit then we reject the H0 and T is significant
If Tobt < Tcrit then we accept our H0 and T is not significant
How is homogeneity of variance assessed?
By Levene’s test, which tests if the variances in different groups are the same
Significant = variances are not equal (<0.05)
Non significant = variances are equal (>0.05)
What does a dependent samples T test aim to do?
Compare the observed differences between the means of the same people measured at different times (e.g. pre to post season)
How is the Tcrit value determined in a dependent samples T test?
Depends upon:
- the degrees of freedom (n-1)
- whether the test is one tailed or two tailed
- your alpha level
What does the value of Tcrit depend upon?
- degrees of freedom (n-1) + (n-1)
- whether the test is one tailed or two tailed
- the alpha level chosen
