Correlation And Regression Flashcards
(29 cards)
Analysis of variance
Evaluates mean differences between two or more treatments
Why is analysis of variance important
Protects against error of inflation (type I error) by first asking, if any difference exist at all among means of groups
What advantage does analysis of variance have over t-test
Compares more than three treatment at the same time
What is the simplest way of describing the relationship between x and y
Scatter plot or diagram
What is a bivariate distribution using scatter diagram
Two scores for each individual, where an individual scores on both x and y axis
What two statistical technique are used to define the relationship between x and y
Correlation and Regression
Looking at how two variable covary or how one change correspond to the other is known as?
Correlation
In correlation study what tells us the direction and magnitude of the relationship or how variables are related and the strength of the relationship
Correlation coefficient
What is a correlation with an upward line
Positive
What is correlation with downward line
Negative
What is a correlation with constant line
No correlation
How does coefficient study vary numerically
-1.0 to +1.0
The correlation coefficient (r) closer to -1 or +1 indicates what?
Stronger relationship and the more nearly it approximate a straight line
What does a correlation of zero mean
No relationship
What does variable that tend to increase together say about r
r is positive
When too variable are indirectly proportional what does that indicate about r
R is negative
What type of variable does correlation and regression describe its relationship?
quantitative variable
What is the primary purpose of linea regression
Predicting one variable from the knowledge of the other
What does correlation truly define
Discovering whether or not a relationship exist in the first place and then specify the strength and direction of the relationship
What is the equation of a linear regression
Y=mx+b
The linear regression equation is also known as
Last squares method
What is coefficient of determination (Rsquared)
Describe how well the sample regression fits the observed data
What is the range of coefficient of determination and what does it mean
0 to 1
0 = poorest
1 = best fit
When is logistic regression used?
Used to predict the relationship when the dependent variable is a discrete nominal variable (I.e represent the presence or absence of some characteristics)
