# Discussion: Levels Of Measurement

THIS IS THE CLASS DISCUSSION TO BE ANSWERED WITH REFERENCES THAT CAN BE FROM THE OUTSIDE I HAVE ATTACHED A REWRITE JUST TO FOLLOW BYBECAUSE REMEMBER EVERYTHING IN WEEK 3 LESSON WILL REVERT BACK TO WEEK 1 OR 2 SO YOU WILL HAVE TO REMEMBER WHAT YOU WROTE FROM WEEK 1 AND 2…. THANKS

Discussion: Levels of Measurement

Four levels of measurement—nominal, ordinal, interval, and ratio—can be used with variables. The nominal level of measurement is used with variables that can be classified arbitrarily by numbers and words. For example, color is a nominal variable because numbers can be assigned to represent it (e.g., green = 1; red = 2). The ordinal level of measurement is used with variables that can be rank-ordered. An example of an ordinal variable is attitudes about the effectiveness of specific police practices in reducing crime. This particular variable requires the use of a Likert scale, in which the participants are asked whether they strongly agree, agree, are neutral, or disagree with a statement about the effectiveness of specific police practices. Based on the results of the Likert scale, the police practices could be rank-ordered according to their effectiveness.

The interval level of measurement is used with variables that have meaningful values attached to them with an equal distance between the values. Interval variables, however, do not have a true zero point. The most widely used example of an interval variable is Fahrenheit temperature, where the value of zero has no intrinsic meaning.

The last level of measurement is ratio—it has the same properties as the interval level of measurement but has a true zero point. Ratio variables are those that can be counted, such as the number of homicides in a year or the number of felony convictions.

One of the most important steps in data analysis and research is applying the correct level of measurement to each variable. Levels of measurement determine the types of statistics that should be used for analysis. If the level of measurement for a variable is incorrectly identified, the statistics will be skewed. In this Discussion, you explore levels of measurement.