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;
32 Cards in this Set
- Front
- Back
Inferential statistics
|
ability we have to take sample data and infer to population
|
|
Census is
|
entire population
|
|
p value
|
probability that 50% shows a true population parameter
|
|
variables
|
stuff that we quantify
|
|
2 discrete variables
|
nominal/categorical and ordinal
|
|
2 continuous variables
|
interval and ratio
|
|
degrees (temp) is a(n)
|
interval variable
|
|
kelvin (temp) is a(n)
|
ratio variable because it has an absolute 0
|
|
kurtosis
|
too fat/too thin
|
|
p (rho) is a
|
parameter
|
|
hypotheses tests allow us....
|
they allow us to talk about EVIDENCE OF CAUSALITY, NOT PROOF
|
|
Don't use t test or ANOVA with
|
continuous; may lose data or something like that?
|
|
t and f tests aren't
|
aren't necessarily for normal distributions
|
|
_______ is the only central tendency measure that makes sense for categorical variables
|
mode
|
|
formula for standard deviation
|
s = square root of E(x-xbar)^2/n
|
|
z score formula for a sample
|
z = (x-xbar)/Sx
|
|
Raw score formula of SSx
|
EX^2 - ((EX)^2)/n))
Divide this all by n and take the square root to get the standard deviation (no square root for variance) |
|
t test is Pearson's r with
|
experiment
|
|
restriction of range is ______ the relationship
|
underestimating
its looking for elongation |
|
If cut off middle of scatter plot, you are _____ the relationship
|
overestimating
|
|
r^2 is
|
the coefficient of determination
|
|
If r=.42..and r^2=.16, it means...
|
we can explain 16% of the differences in x ...i think
R2 is a statistic that will give some information about the goodness of fit of a model. In regression, the R2 coefficient of determination is a statistical measure of how well the regression line approximates the real data points. An R2 of 1.0 indicates that the regression line perfectly fits the data. ask about this |
|
1 - r^2
|
coefficient of nondetermination (we cannot explain 84%?)
|
|
xbar is ________ than population mean
|
xbar is usually smaller than population mean so sometimes its n-1 in the denominator for mew
|
|
Regression of y on x...standard score formula
|
z1y = rzofx ..z1y is z score of y (prediction) and z of x is z score of x
|
|
When r = 1
|
mean of x is predicting mean of y
|
|
If r = 1 and zx=1, then
|
z1y=1
|
|
Syx will tell us _______
|
68% within 1 SE
it is the standard error of estimate |
|
If Syx is small, it is _____
|
hugging line pretty closely
|
|
Formula of Syx
|
Syz = square root of E(Y-Y1)^2/n
and Sy times the square root of 1-r^2 |
|
3 assumptions of Syx
|
1. Linear association (obvious but maybe its parabolic)
2. Homosledasticity (spread of points on regression line are same no matter on x axis) 3. Normal distribution on Y around Y1 |
|
Regression to the mean could be a _____
|
confounding variable
|