| Three different forms of Quantitative Data | |||
|
Quantitative data with qualitative characteristics |
Quantitative data with comparative characteristics |
Quantitative data with quantitative characteristics |
|
| Nominal data | Ordinal data | Interval data | Ratio data |
| The data that you collect can be separated according to discrete categories or labels. Because the categories are counted, the qualitative characteristic becomes quantitative. However this data is not handled the same as other measured data
E.g. Male or Female Yes or No Blue, Red, Yellow or Green
|
Ordinal data has the same properties as nominal data with an added feature, namely we rank the data.
E.g. A hotel rating, where we say 1, 2,3,4 or 5 stars A 4 star hotel is not 4 times as good as a 1 star hotel, but it is better for sure.
We cannot perform mathematical operations using the rankings, but it does give us more information than just categorising the data |
Interval data is strictly quantitative data but interval data does not have a “true zero”. Using this data we however can measure the difference between categories with actual numbers using addition and subtraction. We cannot use multiplication or division though because double the number is not necessarily twice the amount.
E.g. Temperature, 180oC will always be 10oC more that 170oC, but we cannot bake a cake at 360oC for half the time that you would have baked it at 180oC |
Ratio data is the king of data types for statistical purposes. It has a “true zero” which means that we can perform all mathematical operations.
The distinction between ratio data and interval data is a fine line, however if the phrase “twice as much” can be used to accurately describe the relationship then the data is ratio.
E.g. 2.2meters is exactly twice as much as 1.1meter; R10 is twice as much as R5; 20yrs is half of 40yrs.
This data has true zero. In the absence of any there is nothing |
Bronwyn Swartz 