It struck me today in answering a question that statisticians have not been very helpful to those trying to learn statistics in the way they name statistical terms.

I can think of other examples (how many totally different concepts does alpha refer to in statistics?), but the term I was using today was **levels**.

Specifically, there are Multilevel models with two or more sources of random variation. A **two level model** has two sources of random variation, and can have predictors at each level. A common example is where students are sampled within schools. Predictors can be measured at the student level (eg. gender, SES, age) or the school level (enrollment, % who go on to college). The dependent variable has variation from student to student (level 1) and from school to school (level 2).

If a predictor is a fixed factor (meaning it is a categorical predictor), it can have two or more **levels**, meaning categories. In ANOVA, factors (categorical independent variables) have 2 or more levels (2 or more categories).

Then we get to **levels of measurement**: nominal, ordinal, interval, ratio. These **levels** refer to how much information a variable contains. Does it indicate a category, indicate a quantity, etc?

So, a factor with 3 levels that is measured at level 2 of a model has a nominal level of measurement.

**Analyzing Repeated Measures Data: GLM and Mixed Models Approaches**.

{ 2 comments… read them below or add one }

i am beginning to appreciate the note

i don’t understand

{ 1 trackback }