Stats Flashcards
1
Q
See slope, think
A
linear regression
2
Q
sample proportion =
A
successes/trials
3
Q
odds
A
= focal and alternative, treatment and control
4
Q
expected frequency =
A
row total x column total / grand total
5
Q
poisson distribution
A
mean = variance (not SD)
include 0
6
Q
correlations
A
sanity check em
7
Q
variance
A
check its not sd
8
Q
MS
A
check its not SS
9
Q
squares
A
watch out for em!
10
Q
binomial assumptions
A
- number of trials fixed
- each trial is independent
- probability of success is the same in each trial
11
Q
poisson assumptions
A
- successes are independent
- occur with equal probability at every point in space
12
Q
poisson output
A
- if variance > mean, successes are clumped
- if variance < mean successes are more evenly distributed than expected
13
Q
Chi-squared goodness of fit assumptions
A
- population is normally distributed
- random sample
- at least 80% of expected frequencies must be >5
- none <1
14
Q
Agresti-Coull assumptions
A
- sample collected randomly
- data values independent
- normal distribution
15
Q
t-test assumptions
A
- random samples from the population
- numerical variable is normally distributed
- standard deviation (and variance) is the same in both populations