[Tuvshee]

When I am doing multiple calculations one after the other, do I round after every operation, of do I take the exact value and round the final value in the end?

And also, I have 0.216-0.192 = 0.024. The data I started with each have 3sf but my answer now only has two. If I want to continue calculations with my result, do I work in 2sf from then on?

[Arkcon]

When I am doing multiple calculations one after the other, do I round after every operation, of do I take the exact value and round the final value in the end?

When you're actually doing it, that can get unwieldy if there are many steps.  You should keep just 1, or maybe two more sig. figs. as you go through.  Then, at the last step, round to the correct place.  But you don't round at each step, that adds errors at each step.

And also, I have 0.216-0.192 = 0.024. The data I started with each have 3sf but my answer now only has two. If I want to continue calculations with my result, do I work in 2sf from then on?

No.  The correct answer still has 3 s.f..  You will need a placeholder.

[Tuvshee]

What do you mean my a placeholder? Is the answer then 0.0240? But can the final value have more decimal places than the starting values?

[Arkcon]

Convert each number into scientific notation, you will have different exponents, but decimal places is not how we determine significant figures.

[curiouscat]

When I am doing multiple calculations one after the other, do I round after every operation, of do I take the exact value and round the final value in the end?

It is good practice to retain more significant figures in intermediate stages of a calculation, to avoid accumulated rounding errors.

[Chartreuse]

It is practically more ideal to retain more significant figures after every calculation since you get less accurate when rounding off to lesser significant figures. Personally, I was trained in school to round of to 3 decimal places regardless of the number of whole numbers. And also, in rounding off, you follow the number of s.f.'s on the givens with the lowest number of s.f.'s

[Arkcon]

Rounding to 3 decimal places is an old way of doing things.  The convention was, you rounded to 3 decimal places when you had a digit to the left of the decimal, or 4 decimal places if you didn't, because you were using log tables to calculate, which were always printed to 4 decimal places.  My teacher in college used to lament -- the answer to significant figures was much easier, in their day.  Modern calculators have removed the hard and fast 3 or 4 decimal place rule.