Stats Chap 4

• What is the independent variable known as 


  explanatory variable
• What is the dependant variable known as


  Response variable
• Which axis does the explanatory variable go on


  X axis
• Which axis does the response variable go on


  Y axis
• causal relationship


  When a change in the explanatory variable explains the change in the response variable
• When there is no relationship on a scatter diagram we say there is


  No correlation
• What is bivariate data


  Data which has pairs of values for two variables
• what does correlation describe


  the nature of the linear relationship between two variables
• for negatively correlated variables 


  one variable increases and the other decreases
• for positively correlated variables


  where one increases and the other also increases
• Warning for causal relationships


  You must consider the context of the question and use your common sense to determine if it really is a causal relationship
• The regression line of y on x is written in the form :

  y= a + bx
• What does the coefficient b tell you in the regression line 


  The change in y for each unit of x
• if the data is positive, what will the b co-efficient in the regression line be


  b will be positive
• if the data is negative, what will the b co-efficient in the regression line be


  negative
• What do we call the perfect line with the lowest sum of squared deviations


  Least squares regression line
• What might the regression line be referred to as

  Regression equation, linear regression model
• What may the y-intercept tell you in the regression Line


  The y value when x is 0
• When a regression line is a line of y on x

  We can only use it to find the value of the response variable (y-axis), given a value for the explanatory variable (x-axis)
• Interpolation


  Predicting a y-value from an x-value that is within the range of our explanatory variable
• extrapolation


  Predicting a y-value from an x-value that is beyond the range of our explanatory variable
• Why should you not extrapolate


  You get results that are unreliable as you don’t know for sure if the regression line applies to those values