U2 Two-Variable Statistics Flashcards
Two-Variable Analysis
demonstrates if one variable affects the other
Scatter Plot
graph that shows the relationship between 2 variables
Correlation Coefficient
( r ) a measure of how well a linear model fits a 2-variable data set
Correlation
describes how 2 variables are related
Correlation Classifications
TYPE
- Linear
- Non Linear
STRENGTH
- none
- weak
- moderate
- strong
- perfect
DIRECTION
- positive
-negative
regression
analytic technique to determine a model that fits a LOBF/COBF
interpolation
estimating a value between data points
extrapolation
estimating a value beyond data points
linear regression
determine LOBF equation
residual value
vertical deviation of a data points from the LOBF/COBF
outlier
- points that do not follow the trend
- affects LOBF drastically
external factors
influence trends and misleads predictions made from extrapolation
Coefficient of determination
( r^2) a measure of how well a COBF fits a 2-variable set of data
cause & effect
change in x makes a change in y
common cause
variable causes 2 variables to change
reverse cause & effect
dependent and independent variables are reversed in process
dependent is the…
y variable
independent is the…
x variable
accidental relationship
coincidental correlation between 2 varibales
presumed relationship
no cause & effect relationship or common cause but it does not seem to be accidental
extraneous variable
variables not being investigated that can affect the outcome
control group
may help remove/reduce the effect if an extraneous variable
bias factos
- poor wording
- bad data collection
- misinterpretation
- poor analysis
correlation coefficient strong range
(negative or positive) 1 to 0.67
correlation coefficient moderate range
(negative or positive) 0.67 to 0.33
correlation coefficient weak range
(negative or positive) 0.33 to 0
what is the correlation coefficient for a perfect correlation
(negative or positive) 1
what is the correlation coefficient for no correlation
0
hidden variable
variable that affects or obscures the relationship between two other variables, but is difficult to detect
