The statement you provided seems to touch on the levels of measurement in statistics, specifically the difference between nominal, ordinal, interval, and ratio levels.
-
Nominal Level: This is the simplest level of measurement, where data can be categorized but not ranked or ordered. Examples include gender, race, or yes/no questions.
-
Ordinal Level: This level involves ordered categories, where the order matters but the differences between the ranks are not uniform. For example, ratings (like star ratings) can indicate a rank order, such as 1 star, 2 stars, 3 stars, etc., where more stars indicate a better rating.
-
Interval Level: In this level, differences between measurements are meaningful, but there is no true zero point. An example is temperature in Celsius or Fahrenheit.
-
Ratio Level: This level includes all the properties of interval measurement, but also has a true zero point, allowing for the comparison of absolute magnitudes. Examples include height, weight, and age.
In your context, you are discussing ratings, such as a star rating system. Ratings can be seen as ordinal because they indicate rank (e.g., 4 stars is better than 3 stars). However, application to a ratio level can be a topic of debate; some might argue that if a rating system is designed so that each star represents a consistent and equal interval of quality or value (e.g., a 2-star rating is precisely half as "good" as a 4-star rating in a specific way), it might be characterized as having ratio characteristics.
In summary, if the scale is used for qualitative assessments (like satisfaction), it's best considered at least ordinal. If you believe that the differences between ratings represent meaningful, quantifiable differences with a true zero, you might argue for a ratio interpretation, but this often depends on specific context and the design of the rating system.