Multiple linear regression

what "holding the others constant" means and why it is the whole point; choosing predictors from theory, not from p-values; dummy coding categorical predictors and the reference category; entry methods - Enter vs Stepwise, and why stepwise is hard to defend; the full assumption set (linearity, independence of residuals, homoscedasticity, normality of residuals, multicollinearity, influential cases) with the SPSS check for each; VIF and tolerance thresholds; reading the coefficients table; standardised vs unstandardised beta and when each belongs in a paper; hierarchical regression and R-squared change; APA reporting and the regression table.