# difference between interval scale and ratio scale in classifying Data

I was reading about 'different types of data'. I understood the concept of nominal and order based data. But I did not have a clear picture about the difference between interval and ratio scaled data. I know the concept of true zero, absent for interval scale while present for ratio scale, but still I was not satisfied with the concept/examples of kelvin and Celsius temperature.

Please give 2 or 3 examples with the differences. It would be helpful.

• Celsius temperature has an arbitrary zero, so ratios make no sense, e.g. $20^\circ$ C is not meaningfully twice $10^\circ$ C, which is in my view easier to understand by comparing Fahrenheit equivalents $68^\circ$ F with $50^\circ$ F. The zero separating BC (BCE) from AD (CE) is another example. – Nick Cox Apr 13 '20 at 9:51
• @NickCox,Thanks for answering. Could you please elaborate little more? – tedd Apr 13 '20 at 10:42
• This is standard stuff. Good accounts are easy to find e.g. graphpad.com/support/faq/… I don't see a clear question in your post: you just keep asking for more examples. – Nick Cox Apr 13 '20 at 10:45

Height, weight and many physical variables are ratio: If Joe weighs 75 kg and Jane weighs 50 kg. then Joe weighs 25 kg more, but also 1.5 times as much. If you change to pounds, the ratio is the same.

Temperature (in the commonly used scales) is not ratio. 100 C is not twice as hot as 50 C, and if you translate to F then it's a different ratio (212 to 122). Temp. in degrees absolute or Kelvin is ratio.