complexity Flashcards
(20 cards)
What add’s complexity to an experimental design?
adding levels to your IV, thus increasing the number of conditions in an experiment.
t-tests compare what number of conditions?
2
when would you use a one way anova
when you have more then 2 levels to your IV
what is the F statistic
F= between groups variance/ within group variance
what is the F statistic
F= between groups variance/ within group variance
ratio between how much my groups differ from each other and how much variance is in my sample
One Way Anova’s uses a between or within subjects design?
between subjects design. dependent variable is continuous
What test do we use when we have more than 2 levels f our IV and we are using a within subjects design?
Repeated measures anova test
F Distribution
how possible it is to get a F statistics like this again if the null hypothesis is true and there is no variance.
F-distribution+or-
Always + because we squared both the numerator and denominator of the F statistic formula so it will always be +
If the null hypothesis is true what is the most likely F statistic you will get and why?
will be 1, because the ratio for f= between/within =1/1
one-tailed and two-tailed tests refer to…
t-tests
why does F (2, 33) = 11.64, p
F statistic’s have 2 degrees of freedom because the F statistic is a ratio, Thus it’s giving the DF for between and DF for within.
Thus DF 1 is the #of IV levels-1
& DF2 is sample size minus -2
Why don’t we use t-tests for more than two groups?
Because each t-test has a 5% probability of being a type 1 error (false positive) This is an issue because once we did 3 t-tests to identify which group varies from the others we now have 15% probability that our results are a false positive, these are not good odds so we use a different method.
What do we use instead t-test to measure relationship between 3 IV levels?
Post Hoc Test
What is a post hoc test?
Post Hoc tests makes a comparison between group means if a significant F value is found and we have no indication of the direction of the relationship between categorise. Whilst adjusting for possible type 1 errors.
What inflates an alpha false positive rate?
Number of groups for comparison.
what do oneway anova’s NOT tell us?
the direction in which groups differ.
whats a Bonferroni test
it divides the significance level .05 by the number of groups you have to correct inflation in the probability of a type 1 error (false positive)
Whats the Tukey correction?
It’s less conservative than the Bonferroni corrective method.
Tukey compares each level of the IV with each of your conditions. Then shows you how much these two means differ.
what is a cohen’s d?
the size of the effect
