Discrete vs Continuous data

Hi Sayed!

(1) and (2) capture the idea perfectly!

Your example, however, depends on the context.

When talking about 'weight' conceptually, the variable is continuous by nature.

With the limitations of our technology, many of the continuous variables cannot be measured in a continuous format, so they become discrete.

The most prominent example is 'Age'.
Everybody measures 'Age' in integers. It is so common and easy to do that, that age is virtually always discrete.

Most interestingly, when a baby is born, we measure its age in days.
Then we start measuring it in weeks (for a bit), then in months.
Then we do so in years, but 1.5 years, or 2.5 years (parents are still very specific about the age of their child).
At some point people start measuring them in years: 10,11,20,25, etc.
As people reach older age, it's still okay to say: in his fifties, or in her sixties. It doesn't matter if the person is 56 or 57 - for all data science purposes these two are likely to yield the same results. Due to this reason, we create age intervals: 25-35, 35-45, because we don't need that much specificity.

So you can see how 'age' is continuous, becomes discrete, and in the end becomes categorical (categories here are 25-35, 35-45).

***

What matters is:

1) can you measure a certain variable
2) does it really makes sense to measure it

In the weight example, a person doesn't care if he/she is 92.3kg or 92.4kg. In fact, the difference between these two is drinking 1 glass of water, so measuring it continuously doesn't make sense, although conceptually, the variable is continuous, thus we don't have a technological solution to the problem.

To answer your question - for all practical purposes, you are correct. For all theoretical purposes, weight is a continuous variable.

Hope this helps!
Best,
The 365 Team