Linear Regression

• Cronbach's alpha to assess internal reliability/consistency
• Cronbach's alpha

  • Determines the internal consistency or average correlation of items in a questionnaire to assess its internal reliability
  • Should range between 0 and 1 (if negative, check you have  ‘reversed scored the correct items’)
• reliability/internal consistency

  Cronbach’s alpha (α) larger than .9 is deemed excellent, between .8 and .9 is considered Good, .7-.8 is Acceptable, .6-.7 is questionable, .5-.6 is poor and below .5 is considered unacceptable
• What is the difference between correlation and regression?

  ·       Correlation looks for the (symmetric) relation between variables

  ·       Correlation assumes bivariate normality

  ·       Regression examines how one or more variables (Xs) can predict another variable (Y)

  ·       Regression assumes residuals* are normally distributed and that the predicting variables are measured without error
• What is regression

  -       A hypothetical model of the relationship between two variables

  -       The model used is a linear one

  -       Therefore, we describe the relationship using the equation of a straight line

  Outcome = model + error
• yi = b0 +bix

  Regression line

  Bi – regression coefficient for the predictor ‘I’

  -       Gradient of the regression line

  -       Direction/strength of relationship

   

  B0 – intercept (value of y when x = 0

  Point at which regression line crosses the y-axis (ordinate)

   
• •       Cronbach’s alpha to assess reliability of a scale

  •       Linear regression with one predictor

  •       We assess the fit of a regression model

  •       ANOVA – F-test (is the model better than chance)

  •       R2 – what proportion of the variance in outcome score is accounted for by the model

  •       Quick overview of Regression in JASP (more in practical)

  •       Interpretation of a regression model