Parametric Stats Tests

• To construct an equation for predicting the DV from values of the IV
• To describe the strength of the relationship between variables
• The general term for this is regression analysis

• Correlational analyses 
• These often make no distinction between predictor and response variables

• Conditional distributions
• Conditional distributions are distributions of the response variable conditional on a specified value of the predictor variable

The marginal distribution

The straight line upon which the population means lie

The mean population of Y values for a given value of X = (the intercept of the population regression line + the slope of the population regression line) x the value of X

• Population values that are needed to uniquely specify a straight line function 
• The intercept value specifies how far up the Y axis the line is 
• The slope value (regression coefficient) specifies how steep the line is

The rate at which values on the Y axis are changing from one unit change in the X axis

The slope and intercept values that provide the 'best fit' for data must be found

Score = model fit + residual

One which minimises the sum of squares of residuals

• The vertical line between data and regression line
• Observation-predicted value

Sum of products over sum of squares

The sum of products divided by the degrees of freedom

The similarity of the structure of the expressions of variance and covariance

The means of the X and Y values