1.5 - Statistical Tests 1 Flashcards
(16 cards)
normal distribution
mean and standard deviation independent of one another
standard normal distribution
mean of 0and standard deviation of 1
compare normal distributions to standard normal distribution calculation
Z = (Y - Y)
/S
Z = probability with which value lies in a range
Y = Y value
Y = mean
S = standard deviation
null hypothesis
statement about world that can be falsified; no signal
alternative hypothesis
signal (often research hypothesis)
P-valueVariables that determine what statistical test to do? (5)
- what type of response variable
- what type of explanatory variable
- interested in differences or trends/ relationships
- paired or independent sample
- distribution
different types of response variable (4)
- continuous
- discrete/count
- proportion
- binary
different types of explanatory variables (5)
- continuous
- discrete/count
- proportion
- binary
- categorical
different types of distribution (2)
- normal
- non-normal
paired sample
first and second measurement of the same individual are not independent of each other
normal distribution (6)
- parametric tests
- powerful, easy to interpret
- use means
- require data (or residuals) to be normally distributed
- often require similar variance in groups
- can be used to answer complicated questions
non-normal distribution (6)
- non-parametric tests
- less powerful, more conservative
- use medians (data usually ranked before test)
- usually no assumptions about distribution of data
- robust
- often restrictive, cannot answer more complex Qs
how to check for normal distribution (2)
- graphically (does it look normal)
- formal tests (shapiro-wilk test)
shapiro-wilk test in R
shapiro.test(variable name)
H0 = no deviation from normal distribution
P = < 0.05 = reject H0, >0.05 = dont reject (normal)
chi-squared response variable/explanatory variable (2)
- response variable = count (observations)
- explanatory variable = categorical
chi-squared
look at one note