search

numbas / Numbas

A completely browser-based e-assessment/e-learning system, with an emphasis on mathematics
http://www.numbas.org.uk
Apache License 2.0
199 stars 118 forks source link

Change the way number entry parts with precision restrictions work #296

Open christianp opened 10 years ago

christianp commented 10 years ago

Suppose answer is a=1.2345, and you want it entered to 2 dp. Naively setting min=max=a doesn't work, because 1.23 (what the student enters) isn't 1.2345.

This is really complicated, so we decided min and max should be rounded to the desired precision before marking.

In the future, the whole interface should be different, so you set:

And it works as above.

christianp commented 10 years ago

I'm now adding the non-strict option, so answers without trailing zeroes can be marked as correct. Should the wrong number of trailing zeroes be marked incorrect?

For example: "Write 1.4, correct to 4 d.p.". The strict correct answer is 1.4000. If we're not being strict, we also want to mark 1.4 as correct. But what about 1.40 and 1.400? My instinct is that they should be marked incorrect even when we're not being strict, because they mislead you about the degrees of precision used. Is that right?

BillFoster commented 10 years ago

A moot point ! My inclination is to mark as correct as 1.4 is just as non-strict as the others. Perhaps a topic on the user group to ask for comments?

my iPhone

On 12 Nov 2013, at 10:37, "Christian Perfect" notifications@github.com<mailto:notifications@github.com> wrote:

I'm now adding the non-strict option, so answers without trailing zeroes can be marked as correct. Should the wrong number of trailing zeroes be marked incorrect?

For example: "Write 1.4, correct to 4 d.p.". The strict correct answer is 1.4000. If we're not being strict, we also want to mark 1.4 as correct. But what about 1.40 and 1.400? My instinct is that they should be marked incorrect even when we're not being strict, because they mislead you about the degrees of precision used. Is that right?

— Reply to this email directly or view it on GitHubhttps://github.com/numbas/Numbas/issues/296#issuecomment-28283303.

christianp commented 3 years ago

I have a question where you can get two different answers depending on if you use the rounded value of one of the input variables, or the precise value. The displayed correct answer should be the one obtained using the precise value, not the midpoint of the two calculated values.