Why dummy variables rather than one "factor" variable in modelling? I was speaking to a statistician recently who suggested that using dummy variables rather than one variable with several levels reduced the constraints on models, particular reducing the assumption of linearity. I didn't understand the explanation and was wondering if someone could make it clear?
 A: Perhaps he is saying that treating an ordinal variable as continuous (which is reasonably common) means more assumptions in the relationship to the response variable than if you treat it properly as a categorical factor (nominal or ordinal).
If you treat an ordinal variable as though it is continuous you are assuming that the differences between different adjacent levels of the scale are in some sense constant, as well as that this variable is linearly related to the response (assuming you have a linear model).
A: It is based on your research question. For the example of elementary school to high school. Let's say that you're trying to measure the Impact of educational level(niveau d'éducation) on the overall social behavior of a person. You will keep the ordinal/factor variable and eventually you will infer the overall relationship without knowing which level of education can predict social behavior better than others (for Instance, people with high school education level might exhibit better behavior than those who stopped at the elementary level).in this case,you need to use dummies.
