yr1 chap 4 correlation

• the regression line of y on x means a line in the form y = ax + b
• the regression line of x on y means a line in the form x = ay + b
• the coefficient of a regression line is the gradient, representing the change in y for each unit change in x
• data is positively correlated if the gradient of a regression line is positive
• data is negatively correlated if the gradient of a regression line is negative
• the regression line can only be used for to predict values for data which is within the range of it
• using a regression line to predict data inside the range is interpolation and is more accurate
• using a regression line to predict data outside the range is extrapolation and is less accurate
• a regression line of y on x can only be used to predict values of the dependent variable which is y, not values of the independent variable which is x
• a regression line of x on y can only be used to predict values of the dependent variable which is x, not values of the independent variable which is y
• positive correlation means when one variable increases, the other increases
• negative correlation means when one variable increases, the other decreases
• strong correlation is shown on a scatter diagram by a straight line
• weak correlation is shown on a scatter diagram by a less straight line
• if a question refers to the explanatory variable it means the independent variable