Comparing Means Flashcards
We look for a probability value (or pvalue) of?
5% or less (i.e. p<.05)
If results are not significant, results are?
NON-significant
If significant, results were
unlikely to occur by?
chance
Degrees of freedom =
n-1
Independent samples t-test
- Used in between group designs
-Use when P have only taken part
in 1 of our 2 conditions
eg. Unpaired means t-test
Unrelated samples t-test
Between groups t-test
Independent T test assumptions
Data are on a ratio or interval scale.
Data is normally distributed.
Data in one group must be independent of the other.
Variances between the two groups must be similar to each other.
Why do we conduct tests of significance?
To determine if the means between two groups are significantly different, assuming the null hyp about the population is true
What do Paired samples t-tests assume that Independent t-tests don’t?
Assumes our data is from a within-subjects study,
whereas with a one-sample t-test were are comparing one mean to a chosen value.
All test require knowledge of what 3 things?
The test statistic
The degrees of freedom
The significance value
Name versions of the t-test for within-subjects designs:
Related-samples t-test
Paired means t-test
Within subjects t-test
Repeated measures t-test
Paired Samples t-test asumptions
Data must be on an interval or ratio scale
The sample of pairs is a random sample from the population.
The difference between scores is normally distributed (not the raw score)
One Samples T test:
Comparing a single set of scores with a test value rather than another set of scores
Pearsons R (Correlations)
Parametric (i.e., for normally distributed data).
Tells us how much two variables are related to each other.
Standardises covariances to create a correlation coefficient
Correlation Coefficient is between?
Always returns a value between -1 (negative cor) and +1(positive cor)
What is A non-parametric version of the standard correlation coefficient test?
Spearmans Rho (ρ)
Ordinal data, and data that do not meet parametric assumptions.
Data is ranked and then correlations are calculated