chap 5 stats
Help!
|
|
||||
|---|---|---|---|---|---|
| what are the two purposes of the pearson correlation and regression descriptive aspects: to study the relationship between __ and to predict scores on one variable from scores of __- how are scores __ | interval level variables, another variable, how are scores related to one another
🗑
|
||||
| use X for the __ (__) variable and Y for the __ (__) variable | predictor (independent), criterion (dependent)
🗑
|
||||
| the pearson correlation is used to determine the extent to which .. | two variables approximate a linear relationship
🗑
|
||||
| regression is used to identify the line that best describes this relationship as determined by a statistical criterion known as the | least squares
🗑
|
||||
| the relationship between certain variables X and Y, represents the values of X on the abscissa and the values of Y on the ordinate, and the scors for each individual on the body of the graph | scatterplot
🗑
|
||||
| indicates the number of units that variable Y changes as variable X changes by one unit | slope
🗑
|
||||
| the slope (B)= (state equation) | y1-y2/ x1-x2
🗑
|
||||
| a relas X increases so does Y and as X decreases so does Y- this shows what relationship | positive
🗑
|
||||
| inverse relationship between x and y varialbes is what relationship | negative
🗑
|
||||
| point at which a line intersects the y axis when x=0 | intercept
🗑
|
||||
| intercept denoted by letter | a
🗑
|
||||
| a linear model is a __ that states the __ relationships and how they can differ in the values of their __ and values of their __ | equation, linear, intercepts, slopes
🗑
|
||||
| linear model equation | y=a+bX
🗑
|
||||
| b= the __ and a= the __ | slope, intercept
🗑
|
||||
| in the intercept the line intersects the __ axis and _=0 | y, x
🗑
|
||||
| using the linear equation we can substitute the scores on __ and get the scored predicted on __ | x, y
🗑
|
||||
| the pearson correlation coefficient can range from __ to __ and is represented as _ | -1.00 to +1.00, r
🗑
|
||||
| the sign of the correlation coefficient indicates the __ of linear approximation (+) means __ and (-) means __ | direction, direct relationship, inverse relationship
🗑
|
||||
| correlation coefficient of 0 means that there is | no linear relationship between the 2 relationships
🗑
|
||||
| the magnitude of the correlation coefficient is indexed by its | absolute value
🗑
|
||||
| the magnitude indicates the __ to which a __ is approximated | degree , linear relationship
🗑
|
||||
| the further r is in either a positive or negative direction from 0, the ... | higher the magnitude
🗑
|
||||
| we can use the __ of the __ of z scores as an index of the relationship between two variables | sum, products
🗑
|
||||
| the sum of the products of z scores can be influences by __ and __ | size of correlation and sample size
🗑
|
||||
| how do you fix this problem? | divide by N
🗑
|
||||
| we divide by N in order to have an index of correlation independent of __ and because dividing by N will always make .. | N, scores fall between -1 and +1
🗑
|
||||
| what is SSx and SSy | sum of squares for variable X and variable Y
🗑
|
||||
| sum of crossproducts is more precise and efficient because it requires __ and presents __ | fewer steps, presents fewer opportunities for rounding error
🗑
|
||||
| the fact that two variables are correlated does not necessarily imply that one variable __ the other to vary as it does, it is possible for two varailbes to be related to one another but have no __ | causes, causal relationship
🗑
|
||||
| the meaning of a certain correlation coefficient __ within the study | varies
🗑
|
||||
| in behavioral science research, correlations of __, __ and __ __ are considered significant | +/- .20 +/- .30
🗑
|
||||
| the proportion of variability in the dependent variable that can be explained by or that is associated with the independent variable | coefficient of determination (r2)
🗑
|
||||
| the proportion of variability in the dependent variable and cant be explained by and is not associated with the dependent variable | 1-r2
🗑
|
||||
| when two variables are not perfectly correlated, the statistical technique of __ can be used to identify a line that fits the data points better than any other line | regression
🗑
|
||||
| regression is determined by the | least squares criterion
🗑
|
||||
| the regression line describes the nature of the __ between the two variables | linear relationship
🗑
|
||||
| the linear relationship beetween 2 varialbes can be represented by a regression line that takes the general form of | y=a +bx
🗑
|
||||
| the slope of the regression line equation: | b=SCP/SSx
🗑
|
||||
| intercept regression line equation | a=y-bX
🗑
|
||||
| the regression line is the line that __ the __ at the value of the | intersects the yaxis at the value of the intercept
🗑
|
||||
| and the slope of this line is scuh that when x increases by one unit, | y increases by the b
🗑
|
||||
| the slope and the intercept are defined so as to minimize the ___ that the data points are from the regression line | squared vertical distances
🗑
|
||||
| the criterion for deriving the values of the slope and intercept | least squares criterion
🗑
|
||||
| the least squares criterion concerns itself with the squares of the __ and formally defines the values of the slope and intercept so as to __ the __ | discrepancy scores, minimize the sum of these squared errors
🗑
|
||||
| the amount of error for a given individual can be represented by the discrepancy between that persons __ and __ _ scores | actual and predicted y scores
🗑
|
||||
| least squares criterion is limited by the fact that the sum of the discrepancies between the actual y scores and y scores predicted from the regression equation will.. | always equal zerio
🗑
|
||||
| index of predictive error | standard error of estimate
🗑
|
||||
| typical error made when predicting y from x | standard error of estimate
🗑
|
||||
| conditions of standard error of estimate (4) | x and y are interval level measurements, same individuals are measured on both x and y, there is a linear trend between x and y predictions shouldnt be made with scores on x beyond the range measured
🗑
|
||||
| x and y in Syx are __ measurements | interval
🗑
|
||||
| same individuals are measured on | both x and y
🗑
|
||||
| there is a __ trend between x and y | linear
🗑
|
||||
| predictions shouldnt be made withs cores on __ beyond the range measured | x
🗑
|
||||
| absolute magnitude of the standard error of estimate is | meaningful
🗑
|
||||
| the standard error of estimate can be compared with the __ of Y which indicated what the average error in prediction would be if one were to predict a Y score equal to the mean of Y for each individual | standard deviation
🗑
|
||||
| if two variables are related in a nonlinear way the pearson correlation will | not be sensitive to this
🗑
|
||||
| if they are related nonlinearly what are effective models | curvilinear or polynomial regression
🗑
|
||||
| the regression equation for predicting variable x from varialbe y is not the same as the regression equation for predicting variable _ from variable x | y, x
🗑
|
||||
| example of this | conversion rates
🗑
|
||||
| from a statistical perspective, the designation of one variable as X and one varialbe as Y is | arbitrary
🗑
|
||||
| the use of __ presupposes an underylying rationale for making predictions about variable y from variable x, | regression
🗑
|
||||
| if interest is merely in whether a given variable is linearly related to another __ can be applied | pearson correlation
🗑
|
||||
| from a conceptual perspective the decision of which variable to designate as X and which to designate as Y has | important implications
🗑
|
||||
| depending on the particular circumstances, the __ of the correlation when a limited portion of this range is considered might be either less than or greater than if the range had not been so restricted | magnitude
🗑
|
||||
| if two variables are linearly related, then restricting the range of one variable will __ the __ of the __ | reduce the magnitude of the correlation coefficient
🗑
|
||||
| prediction of y from x is only meaningul for the _ of __ values that formed at the basis for the calcuation of the __ equation | range of x, regression
🗑
|
||||
| we must not extend our interpretation of correlational results outside the range of the original data set- the conclusions drawn from a correlational analysis apply only to the | range of variables on which the correlation was based
🗑
|
||||
| the pearson correlation coefficient reps the extent to which two variables approximate a linear relationship for the __ of __ included in its __ | range of variables, calculation
🗑
|
||||
| when using the regression equation for standard scores.. first __ the x and y scores and then apply the __ to calculate the __ for the regression line based on standard scores | standardize x and y scores, apply formula to calculate slope
🗑
|
||||
| in regression line eq. you dont have to calculate the intercept of the regression line when standard scores are analyzed in this manner because it will always | equal zero
🗑
|
||||
| the slope of the regression line in this instance will always equal the | correlation ccoefficient
🗑
|
||||
| a correlation coefficient conveys the number of __ that one variable is predicted to change given a change of one standard score in the other variable, other things being __ | standard scores, equal
🗑
|
||||
| the magnitude and sign of a correlation coefficient can be influenced by __ | outliers
🗑
|
||||
| outliers can do what? | turn weak correlation into strong or strong correlation into weak
🗑
|
||||
| how strong a relationship is determined through | r^2=SSexplained/SStotal
🗑
|
||||
| correlation coeff R (2) | nature of relationship, strength of relationship
🗑
|
||||
| why would we standardize the slope | slope of regression line should be 0 when no linear relationship
🗑
|
||||
| nature of linear relationship is determined by the sign of | b or r
🗑
|
||||
| coef of determination indicates proportion of __ in _ explained or predicted by _ | variability, y, x,
🗑
|
Review the information in the table. When you are ready to quiz yourself you can hide individual columns or the entire table. Then you can click on the empty cells to reveal the answer. Try to recall what will be displayed before clicking the empty cell.
To hide a column, click on the column name.
To hide the entire table, click on the "Hide All" button.
You may also shuffle the rows of the table by clicking on the "Shuffle" button.
Or sort by any of the columns using the down arrow next to any column heading.
If you know all the data on any row, you can temporarily remove it by tapping the trash can to the right of the row.
To hide a column, click on the column name.
To hide the entire table, click on the "Hide All" button.
You may also shuffle the rows of the table by clicking on the "Shuffle" button.
Or sort by any of the columns using the down arrow next to any column heading.
If you know all the data on any row, you can temporarily remove it by tapping the trash can to the right of the row.
Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.
Normal Size Small Size show me how
Normal Size Small Size show me how
Created by:
lilcollins92