1.3.1 PT

measure that can have more than one value

it is plotting one variable against another variable

is a mathematical index that describes the direction and magnitude of a relationship

the value of one variable goes up while the other rises also

the value of one variable rises while the value of the other falls

the pair of variables show no sign of correlation

predicts the scores on one variable based on what we know about the score on another variable obatined from the regression line.

it is the best straight line through a set of points in a scatter diagram

homoscedastic

the best linear relation between the X and Y scores.

is a term used to describe the extend to which knowing the value of one variable predicts the value of the other

it is the regression coefficient

-a

• -is the value of Y when X=0
• *this is where the regression line crosses the Y axis

the difference between an actual score and the predicted score (predicted by the regression line)

a best fitting line keeps residuals to a minimum

0

bc the scores in correlation of both variables are in standarized form (z units)

+1- -1

-converted into Z-scores

-mean is 0

it will always be zero

-r

-is the regression coefficient when both x and y have been converted to z scores

-r

• -is a ratio used to determine the degree of variation in one variable that can be estimated from knowledge about variation in the other variable
• *this is the one we use most commonly

that there is no real relationship between teh variables in question

we want to reject it

if there is a calculated probaility that the association between the 2 variables is significantly differnt from 0 using a t-distribution

the relationship between a test score and some well defined criterion (job aptitude test and actual job performance)

correlation between 2 continouos variables

variables that can take on any values over a range of values

ex) height, weight, intelligence

asscoation between 2 sets of ranks

variables that can have only 2 values

ex) yes-no and correct-incorrect

they naturally form 2 categories

ex) gender: male or female

they reflect an underlying continuous scale forced into dichotomy

ex) passing or failing a bar exam bc although there is a range of scores collected, there can only be pass and fail

expresses the relationship between a continuous variable and an artificial dichotomous variable

ex) a biserial might try to asses relationship between pasing or failing a bar exam (artificial dichotmous) and GPA in law school (continuous)

is when one variable is continuous and the other is true dichotomous

ex) this correlation would try to find relationship between gender( true dichot) and GPA (continuous)

phi symbol

if both variables are artificially dichotomous variables

the SD of the residuals

it is the correlation coefficient sqaured

how much variation in the scores of Y we know as a fucntion what we know about X

is a measure of the nonassociation between 2 variables

is the amount of decrease observed when a regression equation is craeted for one population then applied to anohter

ex) using an regression eqution for incoming freshmans predicted GPA from their SATs. then using the same equation for next years

• using the regression equation derived using one group of subjects to predict performance in a different group of subjects
• *the ddifference between results is standard error of estimate

then significant correlation may be difficult to find even if they may actually be there


ex)trying to find relationship between scores on GRE and performance during 1st year of grad school in math departmanet. no students have been admitted with GRE verbal scores of less than 700. Further most grades in grad school are As. there isnt much variability between students there

considers the relationship among combinations of 3 or more variables

the linear combination of variables that provides the maximum discrimination between categories

ex) attempts to determine whether a set of measures predicts success or failure on a particular performance evaluation

correlation between every variable and every other variable

which items clustered with one another

*meaning if we knew how someone asnwered something we can predict with some accuracy how they would asnwer the others in that cluster

mmultidimensional

they exsit bc everyone rejects the items