Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
13 Cards in this Set
- Front
- Back
|
When to use mean |
-ratio/interval data -spread is similar on either side of midpoint
|
|
|
When to use Standard Deviation |
-ratio/interval data
how much a set of values is spread around the median |
|
|
Median |
-important to report when data are skewed
|
|
|
p-value |
the probability of any observed difference having happened by chance |
|
|
t-test |
used with ratio/interval data with normal distribution
output of this is a p-value
Only used when comparing 2 groups and the data have a normal distribution |
|
|
ANOVA |
used with ratio/interval data with normal distribution
output is a p-value
Used instead of t-test when comparing more than 2 groups |
|
|
Chi-square test |
used with nominal/ordinal data
measure of the differences between actual and expected frequencies
output is p-value and X squared
Loses accuracy when smaller than 5 cells |
|
|
Fisher's Test |
nominal/ordinal data
"correction" to chi-square test
corrects for the small cell number |
|
|
Relative Risk Reduction |
-when you want to know how often the tx worked
RRR=1-RR
|
|
|
Absolute Risk Reduction |
ARR = 100/NNT |
|
|
Number Needed to Treat |
NNT = 100/ARR |
|
|
Pearson Correlation Test |
Used for normally distributed populations and parametric data (interval/ratio) |
|
|
Spearman Rank Correlation |
-non-normally distributed populations
-one variable is ordinal |
