Z Scores, t-tests, 1-Way-ANOVA Flashcards
Briefly, what is the z-score?
Where all normal distributions are the same.
How is the z-score calculated,
The mean is subtracted (so it equals 0) and that is divided by variance (so SD is 1).
What is the main problem with z-scores?
They require us them know the sample parameters mean and SD.
The t-distributions is (more/less) spread out then the z-distribution.
More.
What does the spread of the t-distribution depend on?
The uncertainty of the estimates of the mean and SD.
How is SEM directly calculated?
Use lots of sample estimates of the mean and calculate the variance.
How is SEM indirectly calculated?
Divide the standard deviation of the sample by the square root of n.
What is the main question in t test 1?
Is the sample mean the same as the hypothesised population mean?
What is t in t test 1?
Distance between sample and hypothesised men’s divided by SEM.
Why is a degree of freedom lost in t test 1?
Because an estimate of the sample mean is being used.
What is the main question in t test 2?
Are the means of two samples the same?
How is whether u1 = u2 assessed in t test 2?
Look at how many SDs data points are apart.
How is t calculated in t test 2?
Using the standard deviations of the differences between the two means.
What is generally used to assess and deal with homoscedasticity?
Levene’s test, then if p<0.05, use Satterwaithe’s correction.
What is the main question in t test 3?
Is the mean difference of scores 0?
How do all t tests express a difference between means?
As SEM: the number of SDs from the mean.
Give 2 limits of t tests.
They assume normal distributions and they can only compare 2 means.
In 1-way ANOVA, what is a direct estimate of the variance between means?
Calculate variance from the observed means.
In 1-way ANOVA, what is an indirect estimate of the variance between means?
Calculate the SEM*2, which is the expected variance based on the spread of the individual data points about their mean.
If an IV does not influence a DV, the direct and indirect estimate should be (the same/different).
The same.
If an IV does influence a DV, the direct and indirect estimate should be (the same/different).
Different.
In 1-way ANOVA, what makes up the observed value?
Grand mean + level effect + error.
What is the sum of squares between?
The variability between level means (direct method).
What is the error sum of squares?
The variability due to noice (indirect method).
What ratio is ANOVA looking at?
The observed level effects to that expected from the error.
What is effect size?
The proportion of total variability explainer by knowing which level data belongs to.
What is R squared?
The proportion of variance explained in the experiment.
What is adjusted R squared?
The estimated proportion of variance in the population that level is likely to explain.
In 1-way ANOVA, how is partial Eta-squared used?
To refer to the proportion of variance in the data explained by a factor of an interaction.
In 1-way ANOVA, how is R squared used?
To refer to the promotion of variance explained by all factors and interactions.