Linear Correlation and Regression Flashcards
_______________________ (PCC)
Pearson Correlation Coefficient
Correlation
•shows the __________ between two variables
•the _______ of ________ between the two variables
• The two variables should be _______
Co-relationship
Strength of relationship
numeric
Linear Correlation and Regression
• Linear mean __________
• For ___________ it does not really matter which of the variable is independent or dependent
• For ___________ it matters, there will be an independent variable and a dependent variable
straight line
correlation
Regression
On a straight line, the perfect linear relationship, has the highest value of correlation coefficient which is equal to ________
1
Linear Correlation and Regression
• In real life, we have the perfect linear relationship
T/F
F
• In real life, we don’t have perfect linear relationship
Linear Correlation and Regression
• Lowest value of correlation coefficient is ________
• The value -0.7 and 0.7 are of same strength but __________
zero
in different direction
In positive Correlation, The variables ________________ (as one variable is ________ ,the other variable is _________
move in the same direction
increasing
also increasing.
In Negative Correlation , The variables are in _________ direction
opposite
Small letter r= _____________ coefficient
• rho = ____________ coefficient
sample correlation
population correlation
Regression Coefficient
• Predict one variable from ___________________________
the knowledge of another biological measurement
Regression Coefficient
• Dependent variable: The variable you __________ (The _________ variable)
• Independent e.g. _____
want to predict; Outcome
Age
Regression Coefficient
• Regression coefficient = _____
• It is telling us the amount of ___________ per unit _______
b
change in y ; change in x
Linear equation of a straight line
y =_________
a+bx
Y= a + bx
• a= _______ ( Value of ___ when ________ )
• b= _____ (___________ )
intercept; y; x =0
slope
regression coefficient
Columns needed in calculating linear correlation and regression?
x- x
(x- x)^2
(y-y)
(y-y)^2
(x- x)(y-y)
Both regression and correlation coefficient have the same __________
numerator
Numerator of both correlation and regression is ???
Σ (x- x) (y-y)
Denominator for r (__________ coefficient ) is ??
correlation
= √( Σ(x- x)^2 (Σ (Y-Y)^2)
Denominator for b (__________ coefficient ) is ?
Regression
Σ(x- x)^2
r is generally considered strong when r value is > _____
0.7
A simple guide to interpreting r
• ____-_____ =Poor
• ____-_____ =weak
• ____-_____ = Moderate
• ____-_____ = strong.
NB: r larger than ______ is generally considered strong.
• 0- 0.25 =Poor
• 0.26-0.50 =weak
• 0.51 - 0.75 = Moderate
• 0.76 – 1.0 = strong.
strong.