Solved – Nominal data multiple regression analysis

categorical datamultiple regression

I have been given the following data and am supposed to build a multiple regression model to be able to predict the price of a car:

  1. Price of Car – SPSS Measurement = Scale

  2. Mileage of Car – SPSS Measurement = Scale

  3. Age of Car – SPSS Measurement = Scale

  4. Number of Previous Owners – SPSS Measurement = Nominal

  5. Brand ID – SPSS Measurement = Nominal

I am not sure if the Number of Previous Owners has been categorised properly, and if I should treat it as a dummy variable. Please advice!

Thanks

Best Answer

"A variable can be treated as nominal when its values represent categories with no intrinsic ranking; for example, the department of the company in which an employee works. Examples of nominal variables include region, zip code, or religious affiliation."

"A variable can be treated as scale when its values represent ordered categories with a meaningful metric, so that distance comparisons between values are appropriate. Examples of scale variables include age in years and income in thousands of dollars."

(Source: SPSS User Guide)

Nominal deals with words such as blue eyes or brown eyes to categorize. ordinal means that there is a rank such as first second or third place, interval means that there is not a true zero, such as zero degrees does not mean no temperature, or 80 degrees is not twice as hot as 40 degrees. Ratio, is a measure where zero has meaning. Such as no cars really mean no cars and 4 cars are twice as much as 2 cars. Or such as "zero number of previous owners", meaning that the car is new and thus had no previous owner. So I think unless otherwise tabulated, your variable of interest "Number of Previous Owners" should be a scale variable.

Related Question