Measurement scale attaches meaning to a number. The post is aligned with the Black Box Software Testing Foundations course (BBST) designed by Rebecca Fiedler, Cem Kaner, and James Bach.
You measured how wide is your mobile phone. Let’s say that your reading is 5 and four lines, you write down 5.4. But what is the scale of that reading? From the picture above, we could not conclude that. It could be inches, knots, or centimeters.
If we decide on centimeters, then your reading has meaning, your mobile phone is 5.4 cm wide. The centimeter is a standardized scale. If you texted your reading to your friend, he could “replicate” your phone wide dimension only using a meter with scale in cm. If his phone is 6 cm wide, you both could agree that his phone is wider.
We can not count things if we do not know the reading scale.
Your shirt is brighter than mine. But how much brighter?
We have different scale types.
The scale is the ratio scale when multiplication by the same number has the same meaning. You can find out if the scale is a ratio scale by comparing scale reading with another scale. Inches and cm-scale are both ratio scales because:
a statement that 20 inches (50.8 cm) is two times bigger than 10 inches (25.4 cm) is also true in cm scale.
We can compare the ratio of values for those two scales.
In interval scales, ratio comparison does not hold. Fahrenheit and Celsius’s scales are interval scales because we have the following formula:
C = (F — 32) x 5/9
You could recognize the interval scale when there is adding/subtraction in the formula.
So 100 F/50 F is 2, but 37 C/10 C is 3.7!
But difference holds because of 100 F — 75 F = 25 F and 37.8 C — 23.8 C = 13.9 C
is almost (because of rounding errors) same as 75 F — 50 F = 25 F and 23.8 C — 10 C = 13.8 C
In the ordinal scale, we only know the order of reading values. For example, the first place is better than second place, second place is better than third place. We could be sure that the first place is better than third place. But how much better, we do not know that.
In this scale, value meaning is only to create an object tag. For example, when you buy a conference ticket, that ticket has a number that only has a label purpose. It does not compare you with other participants in any way.
Originally published at https://blog.tentamen.eu on April 27, 2020.