So many regression procedures. Confused?
Regression is the work-horse of research analytics. It has been around for a long time and it probably will be around for a long time to come. Whether we always realise it or not, most of our analytical tools are in some way or another based on the concept of correlation and regression.
Let us look at a few regression-based procedures in the researchers’ toolbox:
1. Simple and multiple linear regression: Applicable if both the single dependent variable (outcome or response variable) and one or many independent variables (predictors) are measured on an interval scale. If we have more than one interval scaled outcome variable, we need to conduct multivariate regression (a multivariate General Linear Model).
2. GLM univariate analysis: If the single outcome variable is interval and some or all predictors are categorical (Regression [with a dummy], ANOVA or ANCOVA).
3. GLM multivariate analysis: If more than one outcome variables are interval and some or all predictors are categorical (Multivariate Regression, Canonical Regression, MANOVA or MANCOVA).
4. Ordinal regression analysis: If there are more than two possible outcomes (but one variable) and they are ordered.
5. Categorical regression (Optical Scaling – CATREG): If all variables are categorical. Extensions of chi-square analysis include log-linear, multi-way frequency analysis and logit/probit.
6. Discriminant function analysis: If there are two (or more than two) possible outcomes and your predictors can be considered interval. Discriminant analysis is generally preferred for polytomous outcomes rather than for a dichotomous outcome when logistic regression is more appropriate.
7. Binary logistic regression: For predicting the value of a categorical response variable with two possible outcomes (dichotomous). This procedure is generally preferred over discriminant analysis for a dichotomous outcome.
8. Multinomial logistic regression procedure: If there are more than two possible outcomes and they do not have an inherent ordering. Discriminant Analysis is often preferred for a polytomous outcome.
Its not really confusing. See elsewhere for more discussions on the above.
_________________________________________
Related Posts:
Building statistical models: Linear (OLS) regression
Which Test: Chi-Square, Logistic Regression, or Log-linear analysis
Which Test: Logistic Regression or Discriminant Function Analysis
_________________________________________
/zza82